Oscar VenterpersName, Eric W. SandersonpersName, Ainhoa MagrachpersName, James R. AllanpersName, Jutta BeherpersName, Kendall R. JonespersName, Hugh P. PossinghampersName, William F. LaurancepersName, Peter WoodpersName, Balázs M. FeketepersName, Marc A. LevypersName, James E.M. WatsonpersName (2016date). “Global terrestrial Human Footprint maps for 1993 and 2009.” Scientific Datatitle·j.
Object of study — the spatial distribution and temporal change of cumulative human pressuresterm across non-Antarctic terrestrial environments
Research question — how eight direct and indirect pressure layers can be standardized, combined, and validated as temporally comparable global Human Footprint mapsterm
Contribution — the first temporally consistent 1993date and 2009date Human Footprint mapterm pair and the first visual validation of a cumulative pressure mapterm
Cross-disciplinary analysis
content-tags & facets:
phenomenon · claim-type · cross-domain-transfer · mechanism · claim-relation · replication-status · limitation — grounded in the full text| phenomenon | cumulative human pressure, disturbance, wilderness, and their macro-ecological and conservation consequences — “The Human footprint maps provide information on where humans are exerting pressure on natural systems, altering them from their natural states. They also provide information on where these pressures are absent, and ecosystems are likely to be operating in a more natural state. These pressure-free lands represent candidate sites for consideration as ‘Wilderness’ 16 , 17 . The new Human Footprint maps have already been used to show that recent economic and population growth has far outstripped increases in the Human Footprint, yet the most biologically diverse regions of Earth have been disproportionately impacted 18 . We anticipate that the 1993 and 2009 Human Footprint maps will find a range of additional uses, such as serving as proxies for human disturbance and wilderness, including understanding the role of human pressures in driving macro-ecological patterns 19 , 20 , species extinction risk and distribution analyses 21 , dispersal ecology 22 , conservation science and decision making 23 , and tracking progress toward policy commitments to conservation 23 , among others.” | |
| claim-type | method — “Here we use the Human Footprint methodology 13 to compile remotely-sensed and bottom-up survey information on eight variables measuring the direct and indirect human pressures on the environment in 1993 and 2009. This synthesis represents not only the most current information of its type, but also the first temporally-consistent set of Human Footprint maps, allowing for analyses of change over time. We also provide the first validation of a cumulative pressure map by adopting methods from remote sensing 15 to visually interpret human pressures in high resolution (median=0.5 m) imagery from 3114 1 km 2 random sample plots globally ( Supplementary Appendix 1 ). We then determine the level of agreement between these visually interpreted pressures and those mapped by the Human Footprint.” | |
| method — evidence | — “To create the Human Footprint maps we adopted the methods developed by Sanderson and colleagues 13 . Data on human pressures in 1993 and 2009 were collected or developed for: 1) the extent of built environments, 2) population density, 3) electric infrastructure, 4) crop lands, 5) pasture lands, 6) roads, 7) railways, and 8) navigable waterways, which are described in detail below ( Fig. 1 , step 1). To facilitate comparison across pressures we placed each human pressure within a 0–10 scale ( Fig. 1 , step 2), weighted within that range according to estimates of their relative levels of human pressure following Sanderson et al. 13 . The resulting standardized pressures were then summed together to create the standardized Human Footprint maps for all non-Antarctic land areas ( Fig. 1 , step 3). Pressures are not intended to be mutually exclusive, and many will co-occur in the same location. Three pressures only had data from a single time period, and these are treated as static in the Human Footprint maps.” | |
| object of study — evidence | — “Remotely-sensed and bottom-up survey information were compiled on eight variables measuring the direct and indirect human pressures on the environment globally in 1993 and 2009. This represents not only the most current information of its type, but also the first temporally-consistent set of Human Footprint maps. Data on human pressures were acquired or developed for: 1) built environments, 2) population density, 3) electric infrastructure, 4) crop lands, 5) pasture lands, 6) roads, 7) railways, and 8) navigable waterways. Pressures were then overlaid to create the standardized Human Footprint maps for all non-Antarctic land areas. A validation analysis using scored pressures from 3114×1 km 2 random sample plots revealed strong agreement with the Human Footprint maps. We anticipate that the Human Footprint maps will find a range of uses as proxies for human disturbance of natural systems. The updated maps should provide an increased understanding of the human pressures that drive macro-ecological patterns, as well as for tracking environmental change and informing conservation science and application.” | |
| research question — evidence | — “Here we use the Human Footprint methodology 13 to compile remotely-sensed and bottom-up survey information on eight variables measuring the direct and indirect human pressures on the environment in 1993 and 2009. This synthesis represents not only the most current information of its type, but also the first temporally-consistent set of Human Footprint maps, allowing for analyses of change over time. We also provide the first validation of a cumulative pressure map by adopting methods from remote sensing 15 to visually interpret human pressures in high resolution (median=0.5 m) imagery from 3114 1 km 2 random sample plots globally ( Supplementary Appendix 1 ). We then determine the level of agreement between these visually interpreted pressures and those mapped by the Human Footprint.” | |
| contribution — evidence | — “Here we use the Human Footprint methodology 13 to compile remotely-sensed and bottom-up survey information on eight variables measuring the direct and indirect human pressures on the environment in 1993 and 2009. This synthesis represents not only the most current information of its type, but also the first temporally-consistent set of Human Footprint maps, allowing for analyses of change over time. We also provide the first validation of a cumulative pressure map by adopting methods from remote sensing 15 to visually interpret human pressures in high resolution (median=0.5 m) imagery from 3114 1 km 2 random sample plots globally ( Supplementary Appendix 1 ). We then determine the level of agreement between these visually interpreted pressures and those mapped by the Human Footprint.” | |
| mechanism | top-down and bottom-up coupling Remote land-cover sensing is combined with systematically surveyed and modelled pressures to detect features either source alone misses. — “Human pressures on the environment are the actions taken by humans with the potential to harm nature 1 , 2 . Cumulative pressure mapping measures the breadth of these pressures by coupling top-down remote sensing of land cover change with data on additional human pressures collected ‘bottom-up’ through systematic surveys and modelling 3 , 4 . The method circumvents the limitations of using remote sensing alone, which has difficulty in detecting low intensity pressures 5 , such as linear infrastructures 6 and pasture lands 7 , and often confounds natural and anthropogenic land covers in arid and mosaic environments 8 .” | mechanisms[0] |
| mechanism | pressure standardization Each input is placed on a common zero-to-ten scale before weighted summation. — “To create the Human Footprint maps we adopted the methods developed by Sanderson and colleagues 13 . Data on human pressures in 1993 and 2009 were collected or developed for: 1) the extent of built environments, 2) population density, 3) electric infrastructure, 4) crop lands, 5) pasture lands, 6) roads, 7) railways, and 8) navigable waterways, which are described in detail below ( Fig. 1 , step 1). To facilitate comparison across pressures we placed each human pressure within a 0–10 scale ( Fig. 1 , step 2), weighted within that range according to estimates of their relative levels of human pressure following Sanderson et al. 13 . The resulting standardized pressures were then summed together to create the standardized Human Footprint maps for all non-Antarctic land areas ( Fig. 1 , step 3). Pressures are not intended to be mutually exclusive, and many will co-occur in the same location. Three pressures only had data from a single time period, and these are treated as static in the Human Footprint maps.” | mechanisms[1] |
| mechanism | weighted summation Standardized co-occurring pressure layers are summed into a zero-to-fifty Human Footprint score. — “To create the Human Footprint maps we adopted the methods developed by Sanderson and colleagues 13 . Data on human pressures in 1993 and 2009 were collected or developed for: 1) the extent of built environments, 2) population density, 3) electric infrastructure, 4) crop lands, 5) pasture lands, 6) roads, 7) railways, and 8) navigable waterways, which are described in detail below ( Fig. 1 , step 1). To facilitate comparison across pressures we placed each human pressure within a 0–10 scale ( Fig. 1 , step 2), weighted within that range according to estimates of their relative levels of human pressure following Sanderson et al. 13 . The resulting standardized pressures were then summed together to create the standardized Human Footprint maps for all non-Antarctic land areas ( Fig. 1 , step 3). Pressures are not intended to be mutually exclusive, and many will co-occur in the same location. Three pressures only had data from a single time period, and these are treated as static in the Human Footprint maps.” | mechanisms[2] |
| mechanism | DMSP-OLS threshold Pixels with calibrated digital number above twenty are treated as built environments. — “To map built environments, we used the Defence Meteorological Satellite Program Operational Line Scanner (DMSP-OLS) composite images which gives the annual average brightness of 30 arc second (~1 km at the equator) pixels in units of digital numbers (DN) 28 . These data are provided for each year from 1992 to 2012. We extracted data for the years 1994 (1993 was excluded due to anomalies in the data), and 2009, and both datasets were then inter-calibrated to facilitate comparison 29 . Using the DMSP-OLS datasets, we considered pixels to be ‘built’ if they exhibited a calibrated DN greater than 20. We selected this threshold based on a global analyses of the implications of a range of thresholds for mapped extent of cities 30 , and visual validation against Landsat imagery for 10 cities spread globally.” | mechanisms[3] |
| mechanism | logarithmic scaling Population pressure rises logarithmically with density and saturates at one thousand people per square kilometre. — “Human population density is scored in this way under the assumption that the pressures people induce on their local natural systems increase logarithmically with increasing population density, and saturate at a level of 1000 people per km 2 .” | mechanisms[4] |
| mechanism | equal quintile approach Night-light digital numbers are binned and rescaled into comparable zero-to-ten scores. — “To include these pressures, we used the inter-calibrated DMSP-OLS layers 28 used for the built environments mapping. The equations for intercalibrating across years are second order quadratics trained using data from Sicily, which was chosen as it had negligible infrastructure change over this period and where DN average roughly 14 (ref. 28 ). For our purposes, DN values of six or less where excluded from consideration prior to calibration of data, as the shape of the quadratic function leads to severe distortion of very low DN values. The inter-calibrated DN data from 1994 were then rescaled using an equal quintile approach into a 0–10 scale ( Table 2 ). To scale the data, we divided the calibrated night light data into 10 equal sample bins (each bin with a DN greater than 1 contains the same number of pixels) based on the DN values and then assigned them scores of 1 through 10, starting with the lowest DN bin. DN values of 0 were assigned a score of 0. The thresholds used to bin the 1994 data where then used to convert the 2009 data into a comparable 0–10 scale.” | mechanisms[5] |
| mechanism | exponential access decay Road and water access pressures decline exponentially to fifteen kilometres from the feature. — “We acquired data on the distribution of roads from gROADS 48 , and excluded all trails and private roads, which were inconsistently mapped, with only a subset of countries mapping their linear infrastructure to this resolution. The dataset is the most comprehensive publicly available database on roads, which compiles nationally mapped road data spanning the period 1980–2000 and has a spatial accuracy of around 500 m. The gROADS data do not include all minor roads, and therefore should be viewed as a map of the major roadways. We mapped the direct and indirect influence of roads by assigning an pressure score of 8 for 0.5 km out for either side of roads, and access pressures were awarded a score of 4 at 0.5 km and decaying exponentially out to 15 km either side of the road ( Table 2 ).” | mechanisms[6] |
| mechanism | high-resolution visual interpretation Independent image interpretation supplies reference pressure scores for random validation plots. — “High resolution images were used to visually interpret human pressures in 3460×1 km 2 sample plots randomly located across the Earth’s non-Antarctic land areas ( Fig. 3a ). Images for these plots were obtained from World Imagery 52 , which provides one meter or better satellite and aerial imagery in many parts of the world and lower resolution satellite imagery worldwide. The map features 0.3 m resolution imagery across the continental United States and parts of Western Europe, as well as many parts of the world, with concentrations in South America, Eastern Europe, India, Japan, the Middle East and Northern Africa, Southern Africa, Australia, and New Zealand. The imagery used for the validation plots had a median resolution of 0.5 meters and a median acquisition year of 2010. Comparable imagery was not available for the 1993 time period, and therefore only the 2009 map underwent validation.” | mechanisms[7] |
| mechanism | certainty filter Cloud-obscured and insufficient-resolution plots are discarded before agreement analysis. — “For the visual interpretation, the extent of built environments, crop lands, pasture lands, roads, human settlements, infrastructures and navigable waterways, were recorded using a standard key for identifying these features ( Supplementary Appendix 1 ). Shape, size, texture and colour of features in the imagery were important characteristics for identifying human pressures on the environment. Interpretations were also marked as ‘certain’ or ‘uncertain’, and the year and resolution of the interpreted image was recorded. The 346 ‘uncertain’ plots were discarded, leaving 3114 validation plots ( Fig. 3a ). In general, plots were classified as ‘uncertain’ for two reasons; either because cloud cover obscured the image, or because only medium resolution (15 m) imagery was available for the plot, preventing accurate interpretation of the image. The human footprint score for each plot was determined in ArcGIS, and the visual and Human Footprint scores were then normalized to a 0–1 scale. As we only retained plots for which visual interpretations of the images were determined to be ‘certain’, we consider the visual score to be the true state of in-situ pressures for the plots.” | mechanisms[8] |
| mechanism | 20% match threshold Mapped and visual scores are categorical matches when they differ by no more than twenty percent. — “Two statistics were used to determine Human Footprint performance, root mean squared error (RMSE) 53 and the Cohen kappa statistic of agreement 54 . The RMSE is a dimensioned (expresses average error in the units of variable of interest) error metric for numerical predictions, and tends to heavily punish large errors. The Kappa statistic expresses the agreement between two categorical datasets corrected for the expected agreement, which is based on a random allocation given the relative class sizes. When calculating the kappa statistic, the Human Footprint score was considered as a match to the visual score if they were within 20% of one another on the 0–1 scale.” | mechanisms[9] |
| mechanism | individual pressure layers Publishing components separately lets users construct region- or species-specific alternative scoring schemes. — “The individual pressure maps were developed to be globally consistent, using a scoring approach originally developed by Sanderson and colleagues 13 . However, in some regions and for some species groups, alternate scores may be better suited for reflecting the pressures exerted by humans on nature. We therefore provide the individual pressure layers that compose the Human Footprint maps, thereby allowing data developers to create alternate scoring schemes that better suit their purposes, as well facilitating the addition of new or alternate data sources.” | mechanisms[10] |
| mechanism | temporal comparison Consistent 1993 and 2009 layers make changes in cumulative pressure measurable through time. — “Mapping human pressures to the environment is an essential first step to identifying priority areas for protection or restoration of natural systems. Understanding the spatial distribution of pressures, as well as their change through time, also provides insights for studies on macro-ecological patterns. The Human Footprint maps for 1993 and 2009 represent the first temporally-consistent maps of the human footprint, as well as much more up-to-date information on cumulative pressures than is currently available. Moreover, the 2009 Human Footprint map is the first cumulative pressure map to have undergone an accuracy assessment.” | mechanisms[11] |
| builds-on | Sanderson and colleagues — “To create the Human Footprint maps we adopted the methods developed by Sanderson and colleagues 13 . Data on human pressures in 1993 and 2009 were collected or developed for: 1) the extent of built environments, 2) population density, 3) electric infrastructure, 4) crop lands, 5) pasture lands, 6) roads, 7) railways, and 8) navigable waterways, which are described in detail below ( Fig. 1 , step 1). To facilitate comparison across pressures we placed each human pressure within a 0–10 scale ( Fig. 1 , step 2), weighted within that range according to estimates of their relative levels of human pressure following Sanderson et al. 13 . The resulting standardized pressures were then summed together to create the standardized Human Footprint maps for all non-Antarctic land areas ( Fig. 1 , step 3). Pressures are not intended to be mutually exclusive, and many will co-occur in the same location. Three pressures only had data from a single time period, and these are treated as static in the Human Footprint maps.” | claim_relations[0] |
| extends | Human Footprint — “Cumulative pressure maps have been developed at regional 9 , 10 and global scales 11 , 12 . The ‘Human Footprint’ was first released in 2002 using data primarily from the early 1990s (approximately 1993) on eight human pressures globally, making it the most complete, highest resolution and globally-consistent terrestrial dataset on cumulative human pressures on the environment 13 . It has been used in a large number of ecological and conservation analyses, and still receives around 100 citations each year, particularly from its data users. However, the Human Footprint is a static and dated view of human pressures on the environment. With many of Earth’s systems experiencing pressures close to or beyond safe levels 14 , there is a strong need for an up-to-date understanding of the spatial and temporal trends in human pressures.” | claim_relations[1] |
| validates | cumulative pressure map — “Here we use the Human Footprint methodology 13 to compile remotely-sensed and bottom-up survey information on eight variables measuring the direct and indirect human pressures on the environment in 1993 and 2009. This synthesis represents not only the most current information of its type, but also the first temporally-consistent set of Human Footprint maps, allowing for analyses of change over time. We also provide the first validation of a cumulative pressure map by adopting methods from remote sensing 15 to visually interpret human pressures in high resolution (median=0.5 m) imagery from 3114 1 km 2 random sample plots globally ( Supplementary Appendix 1 ). We then determine the level of agreement between these visually interpreted pressures and those mapped by the Human Footprint.” | claim_relations[2] |
| sense | Human Footprint as cumulative pressure rather than ecological state or impact — “Moreover, our work is subject to three primary limitations. First, like all attempts to map cumulative pressures we did not fully account for all human pressures. Some of the missing and static pressures, such as invasive species and pollution, may be closely associated with pressures we did consider, and therefore their inclusion may not affect our overall results. Second, a lack of available data resulted in three of our pressures being static through time, which would cause an underestimation of Human Footprint expansion if these pressures expanded at a higher than average rate. Third, the Human Footprint measures the pressure humans place on nature, not the realized ‘state’ or ‘impacts’ on natural systems or their biodiversity. Significant scope exists to determine how natural systems respond to cumulating human pressures, and if non-linearity or thresholds exist where pressures lead to accelerated impacts.” | senses[0] |
| sense | temporally consistent as comparable methodology across 1993 and 2009 — “Here we use the Human Footprint methodology 13 to compile remotely-sensed and bottom-up survey information on eight variables measuring the direct and indirect human pressures on the environment in 1993 and 2009. This synthesis represents not only the most current information of its type, but also the first temporally-consistent set of Human Footprint maps, allowing for analyses of change over time. We also provide the first validation of a cumulative pressure map by adopting methods from remote sensing 15 to visually interpret human pressures in high resolution (median=0.5 m) imagery from 3114 1 km 2 random sample plots globally ( Supplementary Appendix 1 ). We then determine the level of agreement between these visually interpreted pressures and those mapped by the Human Footprint.” | senses[1] |
| sense | built environment as urban buildings, paving, and parks — “Built environments are human produced areas that provide the setting for human activity. In the context of the human footprint, we take these areas to be primarily urban settings, including buildings, paved land and urban parks. Built environments do not provide viable habitats for many species of conservation concern, nor do they provide high levels of ecosystem services 24–27 . As such, built environments were assigned a pressure score of 10.” | senses[2] |
| sense | night lights as sparse electric infrastructure outside the built threshold — “The high sensitivity of the DMSP-OLS 28 dataset provides a means for mapping the sparser electric infrastructure typical of more rural and suburban areas. In 2009, 79% of the lights registered in the DMSP-OLS dataset had a Digital Number less than 20, and are therefore not included in our ‘built environments’ layers. However, these lower DN values are often important human infrastructures, such as rural housing or working landscapes, with associated pressures on natural environments.” | senses[3] |
| sense | roads as direct, edge-effect, and access pressures — “As one of humanity’s most prolific linear infrastructures, roads are an important direct driver of habitat conversion 44 . Beyond simply reducing the extent of suitable habitat, roads can act as population sinks for many species through traffic induced mortality 45 . Roads also fragment otherwise contiguous blocks of habitat, and create edge effects such as reduced humidity 6 and increased fire frequency that reach well beyond the roads immediate footprint 46 . Finally, roads provide conduits for humans to access nature, bringing hunters and nature users into otherwise wilderness locations 47 .” | senses[4] |
| sense | railways as direct pressure without broad access pressure — “While railways are an important component of our global transport system, their pressure on the environment differs in nature from that of our road networks. By modifying a linear swath of habitat, railways exert direct pressure where they are constructed, similar to roads. However, as passengers seldom disembark from trains in places other than rail stations, railways do not provide a means of accessing the natural environments along their borders. To map railways we used the same dataset as was used in the original footprint 31 , as no update of this dataset or alternate source has been developed. The direct pressure of railways where assigned a pressure score of 8 for a distance of 0.5 km on either side of the railway.” | senses[5] |
| sense | visual validation as high-resolution independent plot interpretation — “For the visual interpretation, the extent of built environments, crop lands, pasture lands, roads, human settlements, infrastructures and navigable waterways, were recorded using a standard key for identifying these features ( Supplementary Appendix 1 ). Shape, size, texture and colour of features in the imagery were important characteristics for identifying human pressures on the environment. Interpretations were also marked as ‘certain’ or ‘uncertain’, and the year and resolution of the interpreted image was recorded. The 346 ‘uncertain’ plots were discarded, leaving 3114 validation plots ( Fig. 3a ). In general, plots were classified as ‘uncertain’ for two reasons; either because cloud cover obscured the image, or because only medium resolution (15 m) imagery was available for the plot, preventing accurate interpretation of the image. The human footprint score for each plot was determined in ArcGIS, and the visual and Human Footprint scores were then normalized to a 0–1 scale. As we only retained plots for which visual interpretations of the images were determined to be ‘certain’, we consider the visual score to be the true state of in-situ pressures for the plots.” | senses[6] |
| sense | conservative map as more likely to omit a present pressure than invent one — “There is strong agreement between the Human Footprint measure of pressure and pressures scored by visual interpretation of high resolution imagery. The RMSE for the 3114 validation plots was 0.125 on the normalized 0–1 scale, indicating an average error of approximately 13%. The Kappa statistic was 0.737 ( P <0.01), also indicating good agreement between the Human Footprint and the validation dataset. Of the 3114×1 km 2 validation plots, the Human Footprint scored 94 of them 20% higher than the visual score and 263 of them 20% lower. The remaining 2757 plots (88.5%) were within 20% agreement. While this represents good agreement, it appears that the Human Footprint is to some extent susceptible to mapping pressures as absent in locations where they are actually present. The maps should therefore be considered as conservative estimates of human pressures on the environment. The Kappa statistic measure of agreement is sensitive to the threshold used to consider plots as a ‘match’. If we apply a more stringent threshold for agreement of within 15% of one another, the Kappa statistic falls to 0.565 (moderate agreement), and if we apply a less stringent threshold of within 25%, the Kappa statistic increases to 0.856 (very high agreement).” | senses[7] |
| sense | pressure layers as reusable ingredients for alternate local scores — “The individual pressure maps were developed to be globally consistent, using a scoring approach originally developed by Sanderson and colleagues 13 . However, in some regions and for some species groups, alternate scores may be better suited for reflecting the pressures exerted by humans on nature. We therefore provide the individual pressure layers that compose the Human Footprint maps, thereby allowing data developers to create alternate scoring schemes that better suit their purposes, as well facilitating the addition of new or alternate data sources.” | senses[8] |
| sense | Human Footprint map name reserved for the unaltered distributed data — “While we welcome the opportunity to contribute intellectually and as co-authors to research projects that incorporate our datasets into their work, we make the data freely available without restriction for non-commercial use and redistribution. The data may be altered from their original form, and redistributed if done so free of charge and with a full description of any alterations to the original data. We do however ask that term ‘Human Footprint map’ be used only when referring to the unaltered data in the Human Footprint 7-zip file, and not to alternative versions of the data created by data users, and that the data be cited following the template at the end of this manuscript.” | senses[9] |
| control | uncertain validation plots — held — “For the visual interpretation, the extent of built environments, crop lands, pasture lands, roads, human settlements, infrastructures and navigable waterways, were recorded using a standard key for identifying these features ( Supplementary Appendix 1 ). Shape, size, texture and colour of features in the imagery were important characteristics for identifying human pressures on the environment. Interpretations were also marked as ‘certain’ or ‘uncertain’, and the year and resolution of the interpreted image was recorded. The 346 ‘uncertain’ plots were discarded, leaving 3114 validation plots ( Fig. 3a ). In general, plots were classified as ‘uncertain’ for two reasons; either because cloud cover obscured the image, or because only medium resolution (15 m) imagery was available for the plot, preventing accurate interpretation of the image. The human footprint score for each plot was determined in ArcGIS, and the visual and Human Footprint scores were then normalized to a 0–1 scale. As we only retained plots for which visual interpretations of the images were determined to be ‘certain’, we consider the visual score to be the true state of in-situ pressures for the plots.” — rules out cloud-obscured or insufficient-resolution visual interpretations — “For the visual interpretation, the extent of built environments, crop lands, pasture lands, roads, human settlements, infrastructures and navigable waterways, were recorded using a standard key for identifying these features ( Supplementary Appendix 1 ). Shape, size, texture and colour of features in the imagery were important characteristics for identifying human pressures on the environment. Interpretations were also marked as ‘certain’ or ‘uncertain’, and the year and resolution of the interpreted image was recorded. The 346 ‘uncertain’ plots were discarded, leaving 3114 validation plots ( Fig. 3a ). In general, plots were classified as ‘uncertain’ for two reasons; either because cloud cover obscured the image, or because only medium resolution (15 m) imagery was available for the plot, preventing accurate interpretation of the image. The human footprint score for each plot was determined in ArcGIS, and the visual and Human Footprint scores were then normalized to a 0–1 scale. As we only retained plots for which visual interpretations of the images were determined to be ‘certain’, we consider the visual score to be the true state of in-situ pressures for the plots.” | controls[0] |
| control | 15% and 25% match thresholds — partial — “There is strong agreement between the Human Footprint measure of pressure and pressures scored by visual interpretation of high resolution imagery. The RMSE for the 3114 validation plots was 0.125 on the normalized 0–1 scale, indicating an average error of approximately 13%. The Kappa statistic was 0.737 ( P <0.01), also indicating good agreement between the Human Footprint and the validation dataset. Of the 3114×1 km 2 validation plots, the Human Footprint scored 94 of them 20% higher than the visual score and 263 of them 20% lower. The remaining 2757 plots (88.5%) were within 20% agreement. While this represents good agreement, it appears that the Human Footprint is to some extent susceptible to mapping pressures as absent in locations where they are actually present. The maps should therefore be considered as conservative estimates of human pressures on the environment. The Kappa statistic measure of agreement is sensitive to the threshold used to consider plots as a ‘match’. If we apply a more stringent threshold for agreement of within 15% of one another, the Kappa statistic falls to 0.565 (moderate agreement), and if we apply a less stringent threshold of within 25%, the Kappa statistic increases to 0.856 (very high agreement).” — rules out dependence of the agreement verdict on the default 20% threshold — “There is strong agreement between the Human Footprint measure of pressure and pressures scored by visual interpretation of high resolution imagery. The RMSE for the 3114 validation plots was 0.125 on the normalized 0–1 scale, indicating an average error of approximately 13%. The Kappa statistic was 0.737 ( P <0.01), also indicating good agreement between the Human Footprint and the validation dataset. Of the 3114×1 km 2 validation plots, the Human Footprint scored 94 of them 20% higher than the visual score and 263 of them 20% lower. The remaining 2757 plots (88.5%) were within 20% agreement. While this represents good agreement, it appears that the Human Footprint is to some extent susceptible to mapping pressures as absent in locations where they are actually present. The maps should therefore be considered as conservative estimates of human pressures on the environment. The Kappa statistic measure of agreement is sensitive to the threshold used to consider plots as a ‘match’. If we apply a more stringent threshold for agreement of within 15% of one another, the Kappa statistic falls to 0.565 (moderate agreement), and if we apply a less stringent threshold of within 25%, the Kappa statistic increases to 0.856 (very high agreement).” | controls[1] |
| control | Landsat imagery for 10 cities — rules out unsupported selection of the built-environment brightness threshold — “To map built environments, we used the Defence Meteorological Satellite Program Operational Line Scanner (DMSP-OLS) composite images which gives the annual average brightness of 30 arc second (~1 km at the equator) pixels in units of digital numbers (DN) 28 . These data are provided for each year from 1992 to 2012. We extracted data for the years 1994 (1993 was excluded due to anomalies in the data), and 2009, and both datasets were then inter-calibrated to facilitate comparison 29 . Using the DMSP-OLS datasets, we considered pixels to be ‘built’ if they exhibited a calibrated DN greater than 20. We selected this threshold based on a global analyses of the implications of a range of thresholds for mapped extent of cities 30 , and visual validation against Landsat imagery for 10 cities spread globally.” | controls[2] |
| note | Sanderson and colleagues — The map explicitly adopts the earlier Human Footprint method. — “To create the Human Footprint maps we adopted the methods developed by Sanderson and colleagues 13 . Data on human pressures in 1993 and 2009 were collected or developed for: 1) the extent of built environments, 2) population density, 3) electric infrastructure, 4) crop lands, 5) pasture lands, 6) roads, 7) railways, and 8) navigable waterways, which are described in detail below ( Fig. 1 , step 1). To facilitate comparison across pressures we placed each human pressure within a 0–10 scale ( Fig. 1 , step 2), weighted within that range according to estimates of their relative levels of human pressure following Sanderson et al. 13 . The resulting standardized pressures were then summed together to create the standardized Human Footprint maps for all non-Antarctic land areas ( Fig. 1 , step 3). Pressures are not intended to be mutually exclusive, and many will co-occur in the same location. Three pressures only had data from a single time period, and these are treated as static in the Human Footprint maps.” | claim_relations[0] |
| note | Human Footprint — The paper updates the static early-1990s release into a comparable time series. — “Cumulative pressure maps have been developed at regional 9 , 10 and global scales 11 , 12 . The ‘Human Footprint’ was first released in 2002 using data primarily from the early 1990s (approximately 1993) on eight human pressures globally, making it the most complete, highest resolution and globally-consistent terrestrial dataset on cumulative human pressures on the environment 13 . It has been used in a large number of ecological and conservation analyses, and still receives around 100 citations each year, particularly from its data users. However, the Human Footprint is a static and dated view of human pressures on the environment. With many of Earth’s systems experiencing pressures close to or beyond safe levels 14 , there is a strong need for an up-to-date understanding of the spatial and temporal trends in human pressures.” | claim_relations[1] |
| weight | Human Footprint — 0.95 — “Cumulative pressure maps have been developed at regional 9 , 10 and global scales 11 , 12 . The ‘Human Footprint’ was first released in 2002 using data primarily from the early 1990s (approximately 1993) on eight human pressures globally, making it the most complete, highest resolution and globally-consistent terrestrial dataset on cumulative human pressures on the environment 13 . It has been used in a large number of ecological and conservation analyses, and still receives around 100 citations each year, particularly from its data users. However, the Human Footprint is a static and dated view of human pressures on the environment. With many of Earth’s systems experiencing pressures close to or beyond safe levels 14 , there is a strong need for an up-to-date understanding of the spatial and temporal trends in human pressures.” | claim_relations[1] |
| note | cumulative pressure map — The article reports the first validation of this map class. — “Here we use the Human Footprint methodology 13 to compile remotely-sensed and bottom-up survey information on eight variables measuring the direct and indirect human pressures on the environment in 1993 and 2009. This synthesis represents not only the most current information of its type, but also the first temporally-consistent set of Human Footprint maps, allowing for analyses of change over time. We also provide the first validation of a cumulative pressure map by adopting methods from remote sensing 15 to visually interpret human pressures in high resolution (median=0.5 m) imagery from 3114 1 km 2 random sample plots globally ( Supplementary Appendix 1 ). We then determine the level of agreement between these visually interpreted pressures and those mapped by the Human Footprint.” | claim_relations[2] |
| note | built environments —operationalised-as→ pressure score — Exact assigned score. — “Built environments are human produced areas that provide the setting for human activity. In the context of the human footprint, we take these areas to be primarily urban settings, including buildings, paved land and urban parks. Built environments do not provide viable habitats for many species of conservation concern, nor do they provide high levels of ecosystem services 24–27 . As such, built environments were assigned a pressure score of 10.” | concept_edges[0] |
| relation | built environments —operationalised-as→ pressure score — Built environments are assigned a pressure score of 10. — “Built environments are human produced areas that provide the setting for human activity. In the context of the human footprint, we take these areas to be primarily urban settings, including buildings, paved land and urban parks. Built environments do not provide viable habitats for many species of conservation concern, nor do they provide high levels of ecosystem services 24–27 . As such, built environments were assigned a pressure score of 10.” | concept_edges[0] |
| note | human population density —operationalised-as→ pressure score — Exact thresholded scoring rule. — “Human population density was mapped using the Gridded Population of the World dataset developed by the Centre for International Earth Science Information Network (CIESEN) 37 . The dataset provides a ~4 km×~4 km gridded summary of population census data for the years 1990 and 2010, which we downscaled using bilinear sampling in ArcGIS 10.1 to match the 1 km 2 resolution of the other datasets. For all locations with more than 1000 people·km −2 , we assigned a pressure score of 10 ( Table 2 ). For more sparsely populated areas with densities lower than 1000 people·km −2 , we logarithmically scaled the pressure score using,” | concept_edges[1] |
| relation | human population density —operationalised-as→ pressure score — Population density is mapped into a pressure score that reaches 10 above 1000 people per square kilometre. — “Human population density was mapped using the Gridded Population of the World dataset developed by the Centre for International Earth Science Information Network (CIESEN) 37 . The dataset provides a ~4 km×~4 km gridded summary of population census data for the years 1990 and 2010, which we downscaled using bilinear sampling in ArcGIS 10.1 to match the 1 km 2 resolution of the other datasets. For all locations with more than 1000 people·km −2 , we assigned a pressure score of 10 ( Table 2 ). For more sparsely populated areas with densities lower than 1000 people·km −2 , we logarithmically scaled the pressure score using,” | concept_edges[1] |
| audit | crop lands —operationalised-as→ pressure score — Audit 2026-09-28: DOUBTFUL - read against its full span and kept, but the span writes 'intensive agriculture', not 'crop lands' (named earlier in the section). — “Spatial data on remotely sensed agriculture extent in 1992 were extracted from the UMD Land Cover Classification 40 , and for 2009 from GlobCover 41 . Although intensive agriculture often results in whole-scale ecosystem conversion, we gave it a pressure score of 7 ( Table 2 ), which is lower than built environments because of their less impervious cover.” | concept_edges[2] |
| note | crop lands —operationalised-as→ pressure score — Exact assigned score. — “Spatial data on remotely sensed agriculture extent in 1992 were extracted from the UMD Land Cover Classification 40 , and for 2009 from GlobCover 41 . Although intensive agriculture often results in whole-scale ecosystem conversion, we gave it a pressure score of 7 ( Table 2 ), which is lower than built environments because of their less impervious cover.” | concept_edges[2] |
| relation | crop lands —operationalised-as→ pressure score — Intensive crop land is assigned a pressure score of 7. — “Spatial data on remotely sensed agriculture extent in 1992 were extracted from the UMD Land Cover Classification 40 , and for 2009 from GlobCover 41 . Although intensive agriculture often results in whole-scale ecosystem conversion, we gave it a pressure score of 7 ( Table 2 ), which is lower than built environments because of their less impervious cover.” | concept_edges[2] |
| note | pasture lands —operationalised-as→ pressure score — Exact scaled score. — “Pasture lands cover 22% of the Earth’s land base or almost twice that of agricultural crops 42 , making them the most extensive direct human pressure on the environment. Land grazed by domesticated herbivores is often degraded through a combination of fencing, intensive browsing, soil compaction, invasive grasses and other species, and altered fire regimes 43 . We mapped grazing lands for the year 2000 using a spatial dataset that combines agricultural census data with satellite derived land cover to map pasture extent 42 . We assigned pasture a pressure score of 4, which was then scaled from 0–4 using the percent pasture for each 1 km 2 pixel.” | concept_edges[3] |
| relation | pasture lands —operationalised-as→ pressure score — Pasture is assigned a pressure score up to 4 according to percent pasture. — “Pasture lands cover 22% of the Earth’s land base or almost twice that of agricultural crops 42 , making them the most extensive direct human pressure on the environment. Land grazed by domesticated herbivores is often degraded through a combination of fencing, intensive browsing, soil compaction, invasive grasses and other species, and altered fire regimes 43 . We mapped grazing lands for the year 2000 using a spatial dataset that combines agricultural census data with satellite derived land cover to map pasture extent 42 . We assigned pasture a pressure score of 4, which was then scaled from 0–4 using the percent pasture for each 1 km 2 pixel.” | concept_edges[3] |
| note | roads —operationalised-as→ pressure score — Exact distance-dependent score. — “We acquired data on the distribution of roads from gROADS 48 , and excluded all trails and private roads, which were inconsistently mapped, with only a subset of countries mapping their linear infrastructure to this resolution. The dataset is the most comprehensive publicly available database on roads, which compiles nationally mapped road data spanning the period 1980–2000 and has a spatial accuracy of around 500 m. The gROADS data do not include all minor roads, and therefore should be viewed as a map of the major roadways. We mapped the direct and indirect influence of roads by assigning an pressure score of 8 for 0.5 km out for either side of roads, and access pressures were awarded a score of 4 at 0.5 km and decaying exponentially out to 15 km either side of the road ( Table 2 ).” | concept_edges[4] |
| relation | roads —operationalised-as→ pressure score — Road pressure is scored 8 directly and 4 for access before exponential decay. — “We acquired data on the distribution of roads from gROADS 48 , and excluded all trails and private roads, which were inconsistently mapped, with only a subset of countries mapping their linear infrastructure to this resolution. The dataset is the most comprehensive publicly available database on roads, which compiles nationally mapped road data spanning the period 1980–2000 and has a spatial accuracy of around 500 m. The gROADS data do not include all minor roads, and therefore should be viewed as a map of the major roadways. We mapped the direct and indirect influence of roads by assigning an pressure score of 8 for 0.5 km out for either side of roads, and access pressures were awarded a score of 4 at 0.5 km and decaying exponentially out to 15 km either side of the road ( Table 2 ).” | concept_edges[4] |
| note | railways —operationalised-as→ pressure score — Exact distance-dependent score. — “While railways are an important component of our global transport system, their pressure on the environment differs in nature from that of our road networks. By modifying a linear swath of habitat, railways exert direct pressure where they are constructed, similar to roads. However, as passengers seldom disembark from trains in places other than rail stations, railways do not provide a means of accessing the natural environments along their borders. To map railways we used the same dataset as was used in the original footprint 31 , as no update of this dataset or alternate source has been developed. The direct pressure of railways where assigned a pressure score of 8 for a distance of 0.5 km on either side of the railway.” | concept_edges[5] |
| relation | railways —operationalised-as→ pressure score — Railway direct pressure is assigned a score of 8 within half a kilometre. — “While railways are an important component of our global transport system, their pressure on the environment differs in nature from that of our road networks. By modifying a linear swath of habitat, railways exert direct pressure where they are constructed, similar to roads. However, as passengers seldom disembark from trains in places other than rail stations, railways do not provide a means of accessing the natural environments along their borders. To map railways we used the same dataset as was used in the original footprint 31 , as no update of this dataset or alternate source has been developed. The direct pressure of railways where assigned a pressure score of 8 for a distance of 0.5 km on either side of the railway.” | concept_edges[5] |
| audit | navigable waterways —operationalised-as→ pressure score — Audit 2026-09-28: DOUBTFUL - read against its full span and kept, but the span writes 'awarded a score of 4', not 'pressure score'. — “Navigable rivers layers were created for the years 1994 and 2009, and combined with the navigable coasts and inland seas layers to create the final navigable waterways layers. The access pressure from navigable water bodies were awarded a score of 4 adjacent to the water body, decaying exponentially out to 15 km.” | concept_edges[6] |
| note | navigable waterways —operationalised-as→ pressure score — Exact distance-dependent score. — “Navigable rivers layers were created for the years 1994 and 2009, and combined with the navigable coasts and inland seas layers to create the final navigable waterways layers. The access pressure from navigable water bodies were awarded a score of 4 adjacent to the water body, decaying exponentially out to 15 km.” | concept_edges[6] |
| relation | navigable waterways —operationalised-as→ pressure score — Navigable-water access pressure is assigned a score of 4 and decays to 15 km. — “Navigable rivers layers were created for the years 1994 and 2009, and combined with the navigable coasts and inland seas layers to create the final navigable waterways layers. The access pressure from navigable water bodies were awarded a score of 4 adjacent to the water body, decaying exponentially out to 15 km.” | concept_edges[6] |
| audit | human population density —causes→ pressures people induce on their local natural systems — Audit 2026-09-28: DOUBTFUL - read against its full span and kept, but the span states this as the modelling ASSUMPTION behind the score. — “Human population density is scored in this way under the assumption that the pressures people induce on their local natural systems increase logarithmically with increasing population density, and saturate at a level of 1000 people per km 2 .” | concept_edges[7] |
| effect direction | human population density —causes→ pressures people induce on their local natural systems — increases — “Human population density is scored in this way under the assumption that the pressures people induce on their local natural systems increase logarithmically with increasing population density, and saturate at a level of 1000 people per km 2 .” | concept_edges[7] |
| note | human population density —causes→ pressures people induce on their local natural systems — Explicitly stated as an assumption rather than a measured causal result. — “Human population density is scored in this way under the assumption that the pressures people induce on their local natural systems increase logarithmically with increasing population density, and saturate at a level of 1000 people per km 2 .” | concept_edges[7] |
| relation | human population density —causes→ pressures people induce on their local natural systems — The scoring assumes pressures increase logarithmically with increasing population density. — “Human population density is scored in this way under the assumption that the pressures people induce on their local natural systems increase logarithmically with increasing population density, and saturate at a level of 1000 people per km 2 .” | concept_edges[7] |
| weight | human population density —causes→ pressures people induce on their local natural systems — 0.7 — “Human population density is scored in this way under the assumption that the pressures people induce on their local natural systems increase logarithmically with increasing population density, and saturate at a level of 1000 people per km 2 .” | concept_edges[7] |
| effect direction | roads —causes→ habitat conversion — increases — “As one of humanity’s most prolific linear infrastructures, roads are an important direct driver of habitat conversion 44 . Beyond simply reducing the extent of suitable habitat, roads can act as population sinks for many species through traffic induced mortality 45 . Roads also fragment otherwise contiguous blocks of habitat, and create edge effects such as reduced humidity 6 and increased fire frequency that reach well beyond the roads immediate footprint 46 . Finally, roads provide conduits for humans to access nature, bringing hunters and nature users into otherwise wilderness locations 47 .” | concept_edges[8] |
| note | roads —causes→ habitat conversion — Flat positive causal claim. — “As one of humanity’s most prolific linear infrastructures, roads are an important direct driver of habitat conversion 44 . Beyond simply reducing the extent of suitable habitat, roads can act as population sinks for many species through traffic induced mortality 45 . Roads also fragment otherwise contiguous blocks of habitat, and create edge effects such as reduced humidity 6 and increased fire frequency that reach well beyond the roads immediate footprint 46 . Finally, roads provide conduits for humans to access nature, bringing hunters and nature users into otherwise wilderness locations 47 .” | concept_edges[8] |
| relation | roads —causes→ habitat conversion — Roads are an important direct driver of habitat conversion. — “As one of humanity’s most prolific linear infrastructures, roads are an important direct driver of habitat conversion 44 . Beyond simply reducing the extent of suitable habitat, roads can act as population sinks for many species through traffic induced mortality 45 . Roads also fragment otherwise contiguous blocks of habitat, and create edge effects such as reduced humidity 6 and increased fire frequency that reach well beyond the roads immediate footprint 46 . Finally, roads provide conduits for humans to access nature, bringing hunters and nature users into otherwise wilderness locations 47 .” | concept_edges[8] |
| weight | roads —causes→ habitat conversion — 0.95 — “As one of humanity’s most prolific linear infrastructures, roads are an important direct driver of habitat conversion 44 . Beyond simply reducing the extent of suitable habitat, roads can act as population sinks for many species through traffic induced mortality 45 . Roads also fragment otherwise contiguous blocks of habitat, and create edge effects such as reduced humidity 6 and increased fire frequency that reach well beyond the roads immediate footprint 46 . Finally, roads provide conduits for humans to access nature, bringing hunters and nature users into otherwise wilderness locations 47 .” | concept_edges[8] |
| effect direction | roads —causes→ humidity — reduces — “As one of humanity’s most prolific linear infrastructures, roads are an important direct driver of habitat conversion 44 . Beyond simply reducing the extent of suitable habitat, roads can act as population sinks for many species through traffic induced mortality 45 . Roads also fragment otherwise contiguous blocks of habitat, and create edge effects such as reduced humidity 6 and increased fire frequency that reach well beyond the roads immediate footprint 46 . Finally, roads provide conduits for humans to access nature, bringing hunters and nature users into otherwise wilderness locations 47 .” | concept_edges[9] |
| note | roads —causes→ humidity — Flat negative effect with the sign preserved. — “As one of humanity’s most prolific linear infrastructures, roads are an important direct driver of habitat conversion 44 . Beyond simply reducing the extent of suitable habitat, roads can act as population sinks for many species through traffic induced mortality 45 . Roads also fragment otherwise contiguous blocks of habitat, and create edge effects such as reduced humidity 6 and increased fire frequency that reach well beyond the roads immediate footprint 46 . Finally, roads provide conduits for humans to access nature, bringing hunters and nature users into otherwise wilderness locations 47 .” | concept_edges[9] |
| relation | roads —causes→ humidity — Road edge effects reduce humidity beyond the immediate footprint. — “As one of humanity’s most prolific linear infrastructures, roads are an important direct driver of habitat conversion 44 . Beyond simply reducing the extent of suitable habitat, roads can act as population sinks for many species through traffic induced mortality 45 . Roads also fragment otherwise contiguous blocks of habitat, and create edge effects such as reduced humidity 6 and increased fire frequency that reach well beyond the roads immediate footprint 46 . Finally, roads provide conduits for humans to access nature, bringing hunters and nature users into otherwise wilderness locations 47 .” | concept_edges[9] |
| weight | roads —causes→ humidity — 0.95 — “As one of humanity’s most prolific linear infrastructures, roads are an important direct driver of habitat conversion 44 . Beyond simply reducing the extent of suitable habitat, roads can act as population sinks for many species through traffic induced mortality 45 . Roads also fragment otherwise contiguous blocks of habitat, and create edge effects such as reduced humidity 6 and increased fire frequency that reach well beyond the roads immediate footprint 46 . Finally, roads provide conduits for humans to access nature, bringing hunters and nature users into otherwise wilderness locations 47 .” | concept_edges[9] |
| effect direction | roads —causes→ fire frequency — increases — “As one of humanity’s most prolific linear infrastructures, roads are an important direct driver of habitat conversion 44 . Beyond simply reducing the extent of suitable habitat, roads can act as population sinks for many species through traffic induced mortality 45 . Roads also fragment otherwise contiguous blocks of habitat, and create edge effects such as reduced humidity 6 and increased fire frequency that reach well beyond the roads immediate footprint 46 . Finally, roads provide conduits for humans to access nature, bringing hunters and nature users into otherwise wilderness locations 47 .” | concept_edges[10] |
| note | roads —causes→ fire frequency — Flat positive effect with the sign preserved. — “As one of humanity’s most prolific linear infrastructures, roads are an important direct driver of habitat conversion 44 . Beyond simply reducing the extent of suitable habitat, roads can act as population sinks for many species through traffic induced mortality 45 . Roads also fragment otherwise contiguous blocks of habitat, and create edge effects such as reduced humidity 6 and increased fire frequency that reach well beyond the roads immediate footprint 46 . Finally, roads provide conduits for humans to access nature, bringing hunters and nature users into otherwise wilderness locations 47 .” | concept_edges[10] |
| relation | roads —causes→ fire frequency — Road edge effects increase fire frequency. — “As one of humanity’s most prolific linear infrastructures, roads are an important direct driver of habitat conversion 44 . Beyond simply reducing the extent of suitable habitat, roads can act as population sinks for many species through traffic induced mortality 45 . Roads also fragment otherwise contiguous blocks of habitat, and create edge effects such as reduced humidity 6 and increased fire frequency that reach well beyond the roads immediate footprint 46 . Finally, roads provide conduits for humans to access nature, bringing hunters and nature users into otherwise wilderness locations 47 .” | concept_edges[10] |
| weight | roads —causes→ fire frequency — 0.95 — “As one of humanity’s most prolific linear infrastructures, roads are an important direct driver of habitat conversion 44 . Beyond simply reducing the extent of suitable habitat, roads can act as population sinks for many species through traffic induced mortality 45 . Roads also fragment otherwise contiguous blocks of habitat, and create edge effects such as reduced humidity 6 and increased fire frequency that reach well beyond the roads immediate footprint 46 . Finally, roads provide conduits for humans to access nature, bringing hunters and nature users into otherwise wilderness locations 47 .” | concept_edges[10] |
| note | RMSE —is-a→ error metric — Direct categorical definition. — “Two statistics were used to determine Human Footprint performance, root mean squared error (RMSE) 53 and the Cohen kappa statistic of agreement 54 . The RMSE is a dimensioned (expresses average error in the units of variable of interest) error metric for numerical predictions, and tends to heavily punish large errors. The Kappa statistic expresses the agreement between two categorical datasets corrected for the expected agreement, which is based on a random allocation given the relative class sizes. When calculating the kappa statistic, the Human Footprint score was considered as a match to the visual score if they were within 20% of one another on the 0–1 scale.” | concept_edges[12] |
| relation | RMSE —is-a→ error metric — RMSE is a dimensioned error metric for numerical predictions. — “Two statistics were used to determine Human Footprint performance, root mean squared error (RMSE) 53 and the Cohen kappa statistic of agreement 54 . The RMSE is a dimensioned (expresses average error in the units of variable of interest) error metric for numerical predictions, and tends to heavily punish large errors. The Kappa statistic expresses the agreement between two categorical datasets corrected for the expected agreement, which is based on a random allocation given the relative class sizes. When calculating the kappa statistic, the Human Footprint score was considered as a match to the visual score if they were within 20% of one another on the 0–1 scale.” | concept_edges[12] |
| audit | Human Footprint maps —distinct-from→ realized ‘state’ — Audit 2026-09-28: DOUBTFUL - read against its full span and kept, but the span writes 'the Human Footprint', not 'Human Footprint maps'. — “Moreover, our work is subject to three primary limitations. First, like all attempts to map cumulative pressures we did not fully account for all human pressures. Some of the missing and static pressures, such as invasive species and pollution, may be closely associated with pressures we did consider, and therefore their inclusion may not affect our overall results. Second, a lack of available data resulted in three of our pressures being static through time, which would cause an underestimation of Human Footprint expansion if these pressures expanded at a higher than average rate. Third, the Human Footprint measures the pressure humans place on nature, not the realized ‘state’ or ‘impacts’ on natural systems or their biodiversity. Significant scope exists to determine how natural systems respond to cumulating human pressures, and if non-linearity or thresholds exist where pressures lead to accelerated impacts.” | concept_edges[13] |
| note | Human Footprint maps —distinct-from→ realized ‘state’ — Flat explicit distinction. — “Moreover, our work is subject to three primary limitations. First, like all attempts to map cumulative pressures we did not fully account for all human pressures. Some of the missing and static pressures, such as invasive species and pollution, may be closely associated with pressures we did consider, and therefore their inclusion may not affect our overall results. Second, a lack of available data resulted in three of our pressures being static through time, which would cause an underestimation of Human Footprint expansion if these pressures expanded at a higher than average rate. Third, the Human Footprint measures the pressure humans place on nature, not the realized ‘state’ or ‘impacts’ on natural systems or their biodiversity. Significant scope exists to determine how natural systems respond to cumulating human pressures, and if non-linearity or thresholds exist where pressures lead to accelerated impacts.” | concept_edges[13] |
| relation | Human Footprint maps —distinct-from→ realized ‘state’ — The Human Footprint measures human pressure, not the realized state of natural systems. — “Moreover, our work is subject to three primary limitations. First, like all attempts to map cumulative pressures we did not fully account for all human pressures. Some of the missing and static pressures, such as invasive species and pollution, may be closely associated with pressures we did consider, and therefore their inclusion may not affect our overall results. Second, a lack of available data resulted in three of our pressures being static through time, which would cause an underestimation of Human Footprint expansion if these pressures expanded at a higher than average rate. Third, the Human Footprint measures the pressure humans place on nature, not the realized ‘state’ or ‘impacts’ on natural systems or their biodiversity. Significant scope exists to determine how natural systems respond to cumulating human pressures, and if non-linearity or thresholds exist where pressures lead to accelerated impacts.” | concept_edges[13] |
| weight | Human Footprint maps —distinct-from→ realized ‘state’ — 0.95 — “Moreover, our work is subject to three primary limitations. First, like all attempts to map cumulative pressures we did not fully account for all human pressures. Some of the missing and static pressures, such as invasive species and pollution, may be closely associated with pressures we did consider, and therefore their inclusion may not affect our overall results. Second, a lack of available data resulted in three of our pressures being static through time, which would cause an underestimation of Human Footprint expansion if these pressures expanded at a higher than average rate. Third, the Human Footprint measures the pressure humans place on nature, not the realized ‘state’ or ‘impacts’ on natural systems or their biodiversity. Significant scope exists to determine how natural systems respond to cumulating human pressures, and if non-linearity or thresholds exist where pressures lead to accelerated impacts.” | concept_edges[13] |
| audit | Human Footprint maps —distinct-from→ ‘impacts’ on natural systems — Audit 2026-09-28: DOUBTFUL - read against its full span and kept, but the span writes 'the Human Footprint', not 'Human Footprint maps'. — “Moreover, our work is subject to three primary limitations. First, like all attempts to map cumulative pressures we did not fully account for all human pressures. Some of the missing and static pressures, such as invasive species and pollution, may be closely associated with pressures we did consider, and therefore their inclusion may not affect our overall results. Second, a lack of available data resulted in three of our pressures being static through time, which would cause an underestimation of Human Footprint expansion if these pressures expanded at a higher than average rate. Third, the Human Footprint measures the pressure humans place on nature, not the realized ‘state’ or ‘impacts’ on natural systems or their biodiversity. Significant scope exists to determine how natural systems respond to cumulating human pressures, and if non-linearity or thresholds exist where pressures lead to accelerated impacts.” | concept_edges[14] |
| note | Human Footprint maps —distinct-from→ ‘impacts’ on natural systems — Flat explicit distinction. — “Moreover, our work is subject to three primary limitations. First, like all attempts to map cumulative pressures we did not fully account for all human pressures. Some of the missing and static pressures, such as invasive species and pollution, may be closely associated with pressures we did consider, and therefore their inclusion may not affect our overall results. Second, a lack of available data resulted in three of our pressures being static through time, which would cause an underestimation of Human Footprint expansion if these pressures expanded at a higher than average rate. Third, the Human Footprint measures the pressure humans place on nature, not the realized ‘state’ or ‘impacts’ on natural systems or their biodiversity. Significant scope exists to determine how natural systems respond to cumulating human pressures, and if non-linearity or thresholds exist where pressures lead to accelerated impacts.” | concept_edges[14] |
| relation | Human Footprint maps —distinct-from→ ‘impacts’ on natural systems — The Human Footprint measures pressure, not impacts on natural systems. — “Moreover, our work is subject to three primary limitations. First, like all attempts to map cumulative pressures we did not fully account for all human pressures. Some of the missing and static pressures, such as invasive species and pollution, may be closely associated with pressures we did consider, and therefore their inclusion may not affect our overall results. Second, a lack of available data resulted in three of our pressures being static through time, which would cause an underestimation of Human Footprint expansion if these pressures expanded at a higher than average rate. Third, the Human Footprint measures the pressure humans place on nature, not the realized ‘state’ or ‘impacts’ on natural systems or their biodiversity. Significant scope exists to determine how natural systems respond to cumulating human pressures, and if non-linearity or thresholds exist where pressures lead to accelerated impacts.” | concept_edges[14] |
| weight | Human Footprint maps —distinct-from→ ‘impacts’ on natural systems — 0.95 — “Moreover, our work is subject to three primary limitations. First, like all attempts to map cumulative pressures we did not fully account for all human pressures. Some of the missing and static pressures, such as invasive species and pollution, may be closely associated with pressures we did consider, and therefore their inclusion may not affect our overall results. Second, a lack of available data resulted in three of our pressures being static through time, which would cause an underestimation of Human Footprint expansion if these pressures expanded at a higher than average rate. Third, the Human Footprint measures the pressure humans place on nature, not the realized ‘state’ or ‘impacts’ on natural systems or their biodiversity. Significant scope exists to determine how natural systems respond to cumulating human pressures, and if non-linearity or thresholds exist where pressures lead to accelerated impacts.” | concept_edges[14] |
| outcome | control — held — “For the visual interpretation, the extent of built environments, crop lands, pasture lands, roads, human settlements, infrastructures and navigable waterways, were recorded using a standard key for identifying these features ( Supplementary Appendix 1 ). Shape, size, texture and colour of features in the imagery were important characteristics for identifying human pressures on the environment. Interpretations were also marked as ‘certain’ or ‘uncertain’, and the year and resolution of the interpreted image was recorded. The 346 ‘uncertain’ plots were discarded, leaving 3114 validation plots ( Fig. 3a ). In general, plots were classified as ‘uncertain’ for two reasons; either because cloud cover obscured the image, or because only medium resolution (15 m) imagery was available for the plot, preventing accurate interpretation of the image. The human footprint score for each plot was determined in ArcGIS, and the visual and Human Footprint scores were then normalized to a 0–1 scale. As we only retained plots for which visual interpretations of the images were determined to be ‘certain’, we consider the visual score to be the true state of in-situ pressures for the plots.” | controls[0] |
| outcome | control — partial — “There is strong agreement between the Human Footprint measure of pressure and pressures scored by visual interpretation of high resolution imagery. The RMSE for the 3114 validation plots was 0.125 on the normalized 0–1 scale, indicating an average error of approximately 13%. The Kappa statistic was 0.737 ( P <0.01), also indicating good agreement between the Human Footprint and the validation dataset. Of the 3114×1 km 2 validation plots, the Human Footprint scored 94 of them 20% higher than the visual score and 263 of them 20% lower. The remaining 2757 plots (88.5%) were within 20% agreement. While this represents good agreement, it appears that the Human Footprint is to some extent susceptible to mapping pressures as absent in locations where they are actually present. The maps should therefore be considered as conservative estimates of human pressures on the environment. The Kappa statistic measure of agreement is sensitive to the threshold used to consider plots as a ‘match’. If we apply a more stringent threshold for agreement of within 15% of one another, the Kappa statistic falls to 0.565 (moderate agreement), and if we apply a less stringent threshold of within 25%, the Kappa statistic increases to 0.856 (very high agreement).” | controls[1] |
| relation | cumulative pressure mapping → remote sensing — Coupling remote sensing with surveyed and modelled data circumvents the limits of remote sensing alone. — “Human pressures on the environment are the actions taken by humans with the potential to harm nature 1 , 2 . Cumulative pressure mapping measures the breadth of these pressures by coupling top-down remote sensing of land cover change with data on additional human pressures collected ‘bottom-up’ through systematic surveys and modelling 3 , 4 . The method circumvents the limitations of using remote sensing alone, which has difficulty in detecting low intensity pressures 5 , such as linear infrastructures 6 and pasture lands 7 , and often confounds natural and anthropogenic land covers in arid and mosaic environments 8 .” | unhoused_relations[0] |
| weight | cumulative pressure mapping → remote sensing — 0.95 — “Human pressures on the environment are the actions taken by humans with the potential to harm nature 1 , 2 . Cumulative pressure mapping measures the breadth of these pressures by coupling top-down remote sensing of land cover change with data on additional human pressures collected ‘bottom-up’ through systematic surveys and modelling 3 , 4 . The method circumvents the limitations of using remote sensing alone, which has difficulty in detecting low intensity pressures 5 , such as linear infrastructures 6 and pasture lands 7 , and often confounds natural and anthropogenic land covers in arid and mosaic environments 8 .” | unhoused_relations[0] |
| relation | Human Footprint maps → human disturbance — The maps are intended as proxies for human disturbance. — “Remotely-sensed and bottom-up survey information were compiled on eight variables measuring the direct and indirect human pressures on the environment globally in 1993 and 2009. This represents not only the most current information of its type, but also the first temporally-consistent set of Human Footprint maps. Data on human pressures were acquired or developed for: 1) built environments, 2) population density, 3) electric infrastructure, 4) crop lands, 5) pasture lands, 6) roads, 7) railways, and 8) navigable waterways. Pressures were then overlaid to create the standardized Human Footprint maps for all non-Antarctic land areas. A validation analysis using scored pressures from 3114×1 km 2 random sample plots revealed strong agreement with the Human Footprint maps. We anticipate that the Human Footprint maps will find a range of uses as proxies for human disturbance of natural systems. The updated maps should provide an increased understanding of the human pressures that drive macro-ecological patterns, as well as for tracking environmental change and informing conservation science and application.” | unhoused_relations[1] |
| weight | Human Footprint maps → human disturbance — 0.7 — “Remotely-sensed and bottom-up survey information were compiled on eight variables measuring the direct and indirect human pressures on the environment globally in 1993 and 2009. This represents not only the most current information of its type, but also the first temporally-consistent set of Human Footprint maps. Data on human pressures were acquired or developed for: 1) built environments, 2) population density, 3) electric infrastructure, 4) crop lands, 5) pasture lands, 6) roads, 7) railways, and 8) navigable waterways. Pressures were then overlaid to create the standardized Human Footprint maps for all non-Antarctic land areas. A validation analysis using scored pressures from 3114×1 km 2 random sample plots revealed strong agreement with the Human Footprint maps. We anticipate that the Human Footprint maps will find a range of uses as proxies for human disturbance of natural systems. The updated maps should provide an increased understanding of the human pressures that drive macro-ecological patterns, as well as for tracking environmental change and informing conservation science and application.” | unhoused_relations[1] |
| relation | Human Footprint maps → Wilderness — Pressure-free map cells identify candidate wilderness sites. — “The Human footprint maps provide information on where humans are exerting pressure on natural systems, altering them from their natural states. They also provide information on where these pressures are absent, and ecosystems are likely to be operating in a more natural state. These pressure-free lands represent candidate sites for consideration as ‘Wilderness’ 16 , 17 . The new Human Footprint maps have already been used to show that recent economic and population growth has far outstripped increases in the Human Footprint, yet the most biologically diverse regions of Earth have been disproportionately impacted 18 . We anticipate that the 1993 and 2009 Human Footprint maps will find a range of additional uses, such as serving as proxies for human disturbance and wilderness, including understanding the role of human pressures in driving macro-ecological patterns 19 , 20 , species extinction risk and distribution analyses 21 , dispersal ecology 22 , conservation science and decision making 23 , and tracking progress toward policy commitments to conservation 23 , among others.” | unhoused_relations[2] |
| weight | Human Footprint maps → Wilderness — 0.7 — “The Human footprint maps provide information on where humans are exerting pressure on natural systems, altering them from their natural states. They also provide information on where these pressures are absent, and ecosystems are likely to be operating in a more natural state. These pressure-free lands represent candidate sites for consideration as ‘Wilderness’ 16 , 17 . The new Human Footprint maps have already been used to show that recent economic and population growth has far outstripped increases in the Human Footprint, yet the most biologically diverse regions of Earth have been disproportionately impacted 18 . We anticipate that the 1993 and 2009 Human Footprint maps will find a range of additional uses, such as serving as proxies for human disturbance and wilderness, including understanding the role of human pressures in driving macro-ecological patterns 19 , 20 , species extinction risk and distribution analyses 21 , dispersal ecology 22 , conservation science and decision making 23 , and tracking progress toward policy commitments to conservation 23 , among others.” | unhoused_relations[2] |
| relation | DMSP-OLS → over-glow — Sensor hypersensitivity detects over-glow adjacent to built environments. — “The DMSP-OLS has limitations for the purpose of mapping human settlements, including hyper sensitivity of the sensors causing detection of over-glow adjacent to built environments 30 and bright lights associated with gas flaring from oil production facilities 29 . However, no other data exist to map built environments in a consistent way globally over our time horizon. While other datasets provide a one year snap shot of urban extent, they cannot be compared across time due to large differences in the methodologies used 31–33 , and the wildly contrasting extents in mapped built environments.” | unhoused_relations[3] |
| weight | DMSP-OLS → over-glow — 0.95 — “The DMSP-OLS has limitations for the purpose of mapping human settlements, including hyper sensitivity of the sensors causing detection of over-glow adjacent to built environments 30 and bright lights associated with gas flaring from oil production facilities 29 . However, no other data exist to map built environments in a consistent way globally over our time horizon. While other datasets provide a one year snap shot of urban extent, they cannot be compared across time due to large differences in the methodologies used 31–33 , and the wildly contrasting extents in mapped built environments.” | unhoused_relations[3] |
| relation | DMSP-OLS → gas flaring — Bright gas-flaring signals can be confused with built environments. — “The DMSP-OLS has limitations for the purpose of mapping human settlements, including hyper sensitivity of the sensors causing detection of over-glow adjacent to built environments 30 and bright lights associated with gas flaring from oil production facilities 29 . However, no other data exist to map built environments in a consistent way globally over our time horizon. While other datasets provide a one year snap shot of urban extent, they cannot be compared across time due to large differences in the methodologies used 31–33 , and the wildly contrasting extents in mapped built environments.” | unhoused_relations[4] |
| weight | DMSP-OLS → gas flaring — 0.95 — “The DMSP-OLS has limitations for the purpose of mapping human settlements, including hyper sensitivity of the sensors causing detection of over-glow adjacent to built environments 30 and bright lights associated with gas flaring from oil production facilities 29 . However, no other data exist to map built environments in a consistent way globally over our time horizon. While other datasets provide a one year snap shot of urban extent, they cannot be compared across time due to large differences in the methodologies used 31–33 , and the wildly contrasting extents in mapped built environments.” | unhoused_relations[4] |
| relation | pasture lands → direct human pressure — Pasture is the most extensive direct pressure and degradation combines fencing, browsing, compaction, invasive species, and altered fire. — “Pasture lands cover 22% of the Earth’s land base or almost twice that of agricultural crops 42 , making them the most extensive direct human pressure on the environment. Land grazed by domesticated herbivores is often degraded through a combination of fencing, intensive browsing, soil compaction, invasive grasses and other species, and altered fire regimes 43 . We mapped grazing lands for the year 2000 using a spatial dataset that combines agricultural census data with satellite derived land cover to map pasture extent 42 . We assigned pasture a pressure score of 4, which was then scaled from 0–4 using the percent pasture for each 1 km 2 pixel.” | unhoused_relations[5] |
| weight | pasture lands → direct human pressure — 0.95 — “Pasture lands cover 22% of the Earth’s land base or almost twice that of agricultural crops 42 , making them the most extensive direct human pressure on the environment. Land grazed by domesticated herbivores is often degraded through a combination of fencing, intensive browsing, soil compaction, invasive grasses and other species, and altered fire regimes 43 . We mapped grazing lands for the year 2000 using a spatial dataset that combines agricultural census data with satellite derived land cover to map pasture extent 42 . We assigned pasture a pressure score of 4, which was then scaled from 0–4 using the percent pasture for each 1 km 2 pixel.” | unhoused_relations[5] |
| relation | roads → population sinks — Roads can act as population sinks through traffic-induced mortality. — “As one of humanity’s most prolific linear infrastructures, roads are an important direct driver of habitat conversion 44 . Beyond simply reducing the extent of suitable habitat, roads can act as population sinks for many species through traffic induced mortality 45 . Roads also fragment otherwise contiguous blocks of habitat, and create edge effects such as reduced humidity 6 and increased fire frequency that reach well beyond the roads immediate footprint 46 . Finally, roads provide conduits for humans to access nature, bringing hunters and nature users into otherwise wilderness locations 47 .” | unhoused_relations[6] |
| weight | roads → population sinks — 0.7 — “As one of humanity’s most prolific linear infrastructures, roads are an important direct driver of habitat conversion 44 . Beyond simply reducing the extent of suitable habitat, roads can act as population sinks for many species through traffic induced mortality 45 . Roads also fragment otherwise contiguous blocks of habitat, and create edge effects such as reduced humidity 6 and increased fire frequency that reach well beyond the roads immediate footprint 46 . Finally, roads provide conduits for humans to access nature, bringing hunters and nature users into otherwise wilderness locations 47 .” | unhoused_relations[6] |
| relation | railways → direct pressure — Railways exert direct pressure but do not enable access along most borders. — “While railways are an important component of our global transport system, their pressure on the environment differs in nature from that of our road networks. By modifying a linear swath of habitat, railways exert direct pressure where they are constructed, similar to roads. However, as passengers seldom disembark from trains in places other than rail stations, railways do not provide a means of accessing the natural environments along their borders. To map railways we used the same dataset as was used in the original footprint 31 , as no update of this dataset or alternate source has been developed. The direct pressure of railways where assigned a pressure score of 8 for a distance of 0.5 km on either side of the railway.” | unhoused_relations[7] |
| weight | railways → direct pressure — 0.95 — “While railways are an important component of our global transport system, their pressure on the environment differs in nature from that of our road networks. By modifying a linear swath of habitat, railways exert direct pressure where they are constructed, similar to roads. However, as passengers seldom disembark from trains in places other than rail stations, railways do not provide a means of accessing the natural environments along their borders. To map railways we used the same dataset as was used in the original footprint 31 , as no update of this dataset or alternate source has been developed. The direct pressure of railways where assigned a pressure score of 8 for a distance of 0.5 km on either side of the railway.” | unhoused_relations[7] |
| relation | Rivers → nature — Navigable rivers act as access conduits into nature. — “Like roads, coastlines and navigable rivers act as conduits for people to access nature. While all coastlines are theoretically navigable, for the purposes of the human footprint we only considered coasts 31 as navigable for 80 km either direction of signs of a human settlement, which were mapped as a night lights signal with a DN 28 greater than 6 within 4 km of the coast. We chose 80 km as an approximation of the distance a vessel can travel and return during daylight hours. As new settlements can arise to make new sections of coast navigable, coastal layers were generated for the years 1994 and 2009.” | unhoused_relations[8] |
| weight | Rivers → nature — 0.95 — “Like roads, coastlines and navigable rivers act as conduits for people to access nature. While all coastlines are theoretically navigable, for the purposes of the human footprint we only considered coasts 31 as navigable for 80 km either direction of signs of a human settlement, which were mapped as a night lights signal with a DN 28 greater than 6 within 4 km of the coast. We chose 80 km as an approximation of the distance a vessel can travel and return during daylight hours. As new settlements can arise to make new sections of coast navigable, coastal layers were generated for the years 1994 and 2009.” | unhoused_relations[8] |
| relation | individual pressure layers → alternate scoring schemes — Published component layers allow developers to construct alternative scores. — “The individual pressure maps were developed to be globally consistent, using a scoring approach originally developed by Sanderson and colleagues 13 . However, in some regions and for some species groups, alternate scores may be better suited for reflecting the pressures exerted by humans on nature. We therefore provide the individual pressure layers that compose the Human Footprint maps, thereby allowing data developers to create alternate scoring schemes that better suit their purposes, as well facilitating the addition of new or alternate data sources.” | unhoused_relations[9] |
| weight | individual pressure layers → alternate scoring schemes — 0.95 — “The individual pressure maps were developed to be globally consistent, using a scoring approach originally developed by Sanderson and colleagues 13 . However, in some regions and for some species groups, alternate scores may be better suited for reflecting the pressures exerted by humans on nature. We therefore provide the individual pressure layers that compose the Human Footprint maps, thereby allowing data developers to create alternate scoring schemes that better suit their purposes, as well facilitating the addition of new or alternate data sources.” | unhoused_relations[9] |
| relation | uncertain → visual score — Discarding uncertain plots protects the visual score from cloud and coarse-resolution errors. — “For the visual interpretation, the extent of built environments, crop lands, pasture lands, roads, human settlements, infrastructures and navigable waterways, were recorded using a standard key for identifying these features ( Supplementary Appendix 1 ). Shape, size, texture and colour of features in the imagery were important characteristics for identifying human pressures on the environment. Interpretations were also marked as ‘certain’ or ‘uncertain’, and the year and resolution of the interpreted image was recorded. The 346 ‘uncertain’ plots were discarded, leaving 3114 validation plots ( Fig. 3a ). In general, plots were classified as ‘uncertain’ for two reasons; either because cloud cover obscured the image, or because only medium resolution (15 m) imagery was available for the plot, preventing accurate interpretation of the image. The human footprint score for each plot was determined in ArcGIS, and the visual and Human Footprint scores were then normalized to a 0–1 scale. As we only retained plots for which visual interpretations of the images were determined to be ‘certain’, we consider the visual score to be the true state of in-situ pressures for the plots.” | unhoused_relations[10] |
| relation | Human Footprint maps → conservative estimates — Missing present pressures makes the maps conservative estimates. — “There is strong agreement between the Human Footprint measure of pressure and pressures scored by visual interpretation of high resolution imagery. The RMSE for the 3114 validation plots was 0.125 on the normalized 0–1 scale, indicating an average error of approximately 13%. The Kappa statistic was 0.737 ( P <0.01), also indicating good agreement between the Human Footprint and the validation dataset. Of the 3114×1 km 2 validation plots, the Human Footprint scored 94 of them 20% higher than the visual score and 263 of them 20% lower. The remaining 2757 plots (88.5%) were within 20% agreement. While this represents good agreement, it appears that the Human Footprint is to some extent susceptible to mapping pressures as absent in locations where they are actually present. The maps should therefore be considered as conservative estimates of human pressures on the environment. The Kappa statistic measure of agreement is sensitive to the threshold used to consider plots as a ‘match’. If we apply a more stringent threshold for agreement of within 15% of one another, the Kappa statistic falls to 0.565 (moderate agreement), and if we apply a less stringent threshold of within 25%, the Kappa statistic increases to 0.856 (very high agreement).” | unhoused_relations[11] |
| weight | Human Footprint maps → conservative estimates — 0.95 — “There is strong agreement between the Human Footprint measure of pressure and pressures scored by visual interpretation of high resolution imagery. The RMSE for the 3114 validation plots was 0.125 on the normalized 0–1 scale, indicating an average error of approximately 13%. The Kappa statistic was 0.737 ( P <0.01), also indicating good agreement between the Human Footprint and the validation dataset. Of the 3114×1 km 2 validation plots, the Human Footprint scored 94 of them 20% higher than the visual score and 263 of them 20% lower. The remaining 2757 plots (88.5%) were within 20% agreement. While this represents good agreement, it appears that the Human Footprint is to some extent susceptible to mapping pressures as absent in locations where they are actually present. The maps should therefore be considered as conservative estimates of human pressures on the environment. The Kappa statistic measure of agreement is sensitive to the threshold used to consider plots as a ‘match’. If we apply a more stringent threshold for agreement of within 15% of one another, the Kappa statistic falls to 0.565 (moderate agreement), and if we apply a less stringent threshold of within 25%, the Kappa statistic increases to 0.856 (very high agreement).” | unhoused_relations[11] |
| relation | thresholds → Kappa statistic — Changing the match threshold changes the Kappa statistic substantially. — “There is strong agreement between the Human Footprint measure of pressure and pressures scored by visual interpretation of high resolution imagery. The RMSE for the 3114 validation plots was 0.125 on the normalized 0–1 scale, indicating an average error of approximately 13%. The Kappa statistic was 0.737 ( P <0.01), also indicating good agreement between the Human Footprint and the validation dataset. Of the 3114×1 km 2 validation plots, the Human Footprint scored 94 of them 20% higher than the visual score and 263 of them 20% lower. The remaining 2757 plots (88.5%) were within 20% agreement. While this represents good agreement, it appears that the Human Footprint is to some extent susceptible to mapping pressures as absent in locations where they are actually present. The maps should therefore be considered as conservative estimates of human pressures on the environment. The Kappa statistic measure of agreement is sensitive to the threshold used to consider plots as a ‘match’. If we apply a more stringent threshold for agreement of within 15% of one another, the Kappa statistic falls to 0.565 (moderate agreement), and if we apply a less stringent threshold of within 25%, the Kappa statistic increases to 0.856 (very high agreement).” | unhoused_relations[12] |
| relation | static → Human Footprint — Static pressure layers can underestimate Human Footprint expansion. — “Moreover, our work is subject to three primary limitations. First, like all attempts to map cumulative pressures we did not fully account for all human pressures. Some of the missing and static pressures, such as invasive species and pollution, may be closely associated with pressures we did consider, and therefore their inclusion may not affect our overall results. Second, a lack of available data resulted in three of our pressures being static through time, which would cause an underestimation of Human Footprint expansion if these pressures expanded at a higher than average rate. Third, the Human Footprint measures the pressure humans place on nature, not the realized ‘state’ or ‘impacts’ on natural systems or their biodiversity. Significant scope exists to determine how natural systems respond to cumulating human pressures, and if non-linearity or thresholds exist where pressures lead to accelerated impacts.” | unhoused_relations[13] |
| weight | static → Human Footprint — 0.95 — “Moreover, our work is subject to three primary limitations. First, like all attempts to map cumulative pressures we did not fully account for all human pressures. Some of the missing and static pressures, such as invasive species and pollution, may be closely associated with pressures we did consider, and therefore their inclusion may not affect our overall results. Second, a lack of available data resulted in three of our pressures being static through time, which would cause an underestimation of Human Footprint expansion if these pressures expanded at a higher than average rate. Third, the Human Footprint measures the pressure humans place on nature, not the realized ‘state’ or ‘impacts’ on natural systems or their biodiversity. Significant scope exists to determine how natural systems respond to cumulating human pressures, and if non-linearity or thresholds exist where pressures lead to accelerated impacts.” | unhoused_relations[13] |
| relation | Human Footprint → non-linearity — The dataset leaves open whether natural-system responses are nonlinear or thresholded. — “Moreover, our work is subject to three primary limitations. First, like all attempts to map cumulative pressures we did not fully account for all human pressures. Some of the missing and static pressures, such as invasive species and pollution, may be closely associated with pressures we did consider, and therefore their inclusion may not affect our overall results. Second, a lack of available data resulted in three of our pressures being static through time, which would cause an underestimation of Human Footprint expansion if these pressures expanded at a higher than average rate. Third, the Human Footprint measures the pressure humans place on nature, not the realized ‘state’ or ‘impacts’ on natural systems or their biodiversity. Significant scope exists to determine how natural systems respond to cumulating human pressures, and if non-linearity or thresholds exist where pressures lead to accelerated impacts.” | unhoused_relations[14] |
| weight | Human Footprint → non-linearity — 0.7 — “Moreover, our work is subject to three primary limitations. First, like all attempts to map cumulative pressures we did not fully account for all human pressures. Some of the missing and static pressures, such as invasive species and pollution, may be closely associated with pressures we did consider, and therefore their inclusion may not affect our overall results. Second, a lack of available data resulted in three of our pressures being static through time, which would cause an underestimation of Human Footprint expansion if these pressures expanded at a higher than average rate. Third, the Human Footprint measures the pressure humans place on nature, not the realized ‘state’ or ‘impacts’ on natural systems or their biodiversity. Significant scope exists to determine how natural systems respond to cumulating human pressures, and if non-linearity or thresholds exist where pressures lead to accelerated impacts.” | unhoused_relations[14] |
| relation | Human Footprint → thresholds — The dataset leaves open whether thresholds produce accelerated impacts. — “Moreover, our work is subject to three primary limitations. First, like all attempts to map cumulative pressures we did not fully account for all human pressures. Some of the missing and static pressures, such as invasive species and pollution, may be closely associated with pressures we did consider, and therefore their inclusion may not affect our overall results. Second, a lack of available data resulted in three of our pressures being static through time, which would cause an underestimation of Human Footprint expansion if these pressures expanded at a higher than average rate. Third, the Human Footprint measures the pressure humans place on nature, not the realized ‘state’ or ‘impacts’ on natural systems or their biodiversity. Significant scope exists to determine how natural systems respond to cumulating human pressures, and if non-linearity or thresholds exist where pressures lead to accelerated impacts.” | unhoused_relations[15] |
| weight | Human Footprint → thresholds — 0.7 — “Moreover, our work is subject to three primary limitations. First, like all attempts to map cumulative pressures we did not fully account for all human pressures. Some of the missing and static pressures, such as invasive species and pollution, may be closely associated with pressures we did consider, and therefore their inclusion may not affect our overall results. Second, a lack of available data resulted in three of our pressures being static through time, which would cause an underestimation of Human Footprint expansion if these pressures expanded at a higher than average rate. Third, the Human Footprint measures the pressure humans place on nature, not the realized ‘state’ or ‘impacts’ on natural systems or their biodiversity. Significant scope exists to determine how natural systems respond to cumulating human pressures, and if non-linearity or thresholds exist where pressures lead to accelerated impacts.” | unhoused_relations[15] |
| relation | Human Footprint maps → priority areas — Mapping pressures is a first step toward identifying protection or restoration priorities. — “Mapping human pressures to the environment is an essential first step to identifying priority areas for protection or restoration of natural systems. Understanding the spatial distribution of pressures, as well as their change through time, also provides insights for studies on macro-ecological patterns. The Human Footprint maps for 1993 and 2009 represent the first temporally-consistent maps of the human footprint, as well as much more up-to-date information on cumulative pressures than is currently available. Moreover, the 2009 Human Footprint map is the first cumulative pressure map to have undergone an accuracy assessment.” | unhoused_relations[16] |
| relation | Human Footprint map → Human Footprint maps — The authors reserve the singular name for unaltered distributed data, not user-created alternatives. — “While we welcome the opportunity to contribute intellectually and as co-authors to research projects that incorporate our datasets into their work, we make the data freely available without restriction for non-commercial use and redistribution. The data may be altered from their original form, and redistributed if done so free of charge and with a full description of any alterations to the original data. We do however ask that term ‘Human Footprint map’ be used only when referring to the unaltered data in the Human Footprint 7-zip file, and not to alternative versions of the data created by data users, and that the data be cited following the template at the end of this manuscript.” | unhoused_relations[17] |
| no type for this | cumulative pressure mapping → remote sensing — Coupling remote sensing with surveyed and modelled data circumvents the limits of remote sensing alone. — “Human pressures on the environment are the actions taken by humans with the potential to harm nature 1 , 2 . Cumulative pressure mapping measures the breadth of these pressures by coupling top-down remote sensing of land cover change with data on additional human pressures collected ‘bottom-up’ through systematic surveys and modelling 3 , 4 . The method circumvents the limitations of using remote sensing alone, which has difficulty in detecting low intensity pressures 5 , such as linear infrastructures 6 and pasture lands 7 , and often confounds natural and anthropogenic land covers in arid and mosaic environments 8 .” | unhoused_relations[0] |
| no type for this | Human Footprint maps → human disturbance — The maps are intended as proxies for human disturbance. — “Remotely-sensed and bottom-up survey information were compiled on eight variables measuring the direct and indirect human pressures on the environment globally in 1993 and 2009. This represents not only the most current information of its type, but also the first temporally-consistent set of Human Footprint maps. Data on human pressures were acquired or developed for: 1) built environments, 2) population density, 3) electric infrastructure, 4) crop lands, 5) pasture lands, 6) roads, 7) railways, and 8) navigable waterways. Pressures were then overlaid to create the standardized Human Footprint maps for all non-Antarctic land areas. A validation analysis using scored pressures from 3114×1 km 2 random sample plots revealed strong agreement with the Human Footprint maps. We anticipate that the Human Footprint maps will find a range of uses as proxies for human disturbance of natural systems. The updated maps should provide an increased understanding of the human pressures that drive macro-ecological patterns, as well as for tracking environmental change and informing conservation science and application.” | unhoused_relations[1] |
| no type for this | Human Footprint maps → Wilderness — Pressure-free map cells identify candidate wilderness sites. — “The Human footprint maps provide information on where humans are exerting pressure on natural systems, altering them from their natural states. They also provide information on where these pressures are absent, and ecosystems are likely to be operating in a more natural state. These pressure-free lands represent candidate sites for consideration as ‘Wilderness’ 16 , 17 . The new Human Footprint maps have already been used to show that recent economic and population growth has far outstripped increases in the Human Footprint, yet the most biologically diverse regions of Earth have been disproportionately impacted 18 . We anticipate that the 1993 and 2009 Human Footprint maps will find a range of additional uses, such as serving as proxies for human disturbance and wilderness, including understanding the role of human pressures in driving macro-ecological patterns 19 , 20 , species extinction risk and distribution analyses 21 , dispersal ecology 22 , conservation science and decision making 23 , and tracking progress toward policy commitments to conservation 23 , among others.” | unhoused_relations[2] |
| no type for this | DMSP-OLS → over-glow — Sensor hypersensitivity detects over-glow adjacent to built environments. — “The DMSP-OLS has limitations for the purpose of mapping human settlements, including hyper sensitivity of the sensors causing detection of over-glow adjacent to built environments 30 and bright lights associated with gas flaring from oil production facilities 29 . However, no other data exist to map built environments in a consistent way globally over our time horizon. While other datasets provide a one year snap shot of urban extent, they cannot be compared across time due to large differences in the methodologies used 31–33 , and the wildly contrasting extents in mapped built environments.” | unhoused_relations[3] |
| no type for this | DMSP-OLS → gas flaring — Bright gas-flaring signals can be confused with built environments. — “The DMSP-OLS has limitations for the purpose of mapping human settlements, including hyper sensitivity of the sensors causing detection of over-glow adjacent to built environments 30 and bright lights associated with gas flaring from oil production facilities 29 . However, no other data exist to map built environments in a consistent way globally over our time horizon. While other datasets provide a one year snap shot of urban extent, they cannot be compared across time due to large differences in the methodologies used 31–33 , and the wildly contrasting extents in mapped built environments.” | unhoused_relations[4] |
| no type for this | pasture lands → direct human pressure — Pasture is the most extensive direct pressure and degradation combines fencing, browsing, compaction, invasive species, and altered fire. — “Pasture lands cover 22% of the Earth’s land base or almost twice that of agricultural crops 42 , making them the most extensive direct human pressure on the environment. Land grazed by domesticated herbivores is often degraded through a combination of fencing, intensive browsing, soil compaction, invasive grasses and other species, and altered fire regimes 43 . We mapped grazing lands for the year 2000 using a spatial dataset that combines agricultural census data with satellite derived land cover to map pasture extent 42 . We assigned pasture a pressure score of 4, which was then scaled from 0–4 using the percent pasture for each 1 km 2 pixel.” | unhoused_relations[5] |
| no type for this | roads → population sinks — Roads can act as population sinks through traffic-induced mortality. — “As one of humanity’s most prolific linear infrastructures, roads are an important direct driver of habitat conversion 44 . Beyond simply reducing the extent of suitable habitat, roads can act as population sinks for many species through traffic induced mortality 45 . Roads also fragment otherwise contiguous blocks of habitat, and create edge effects such as reduced humidity 6 and increased fire frequency that reach well beyond the roads immediate footprint 46 . Finally, roads provide conduits for humans to access nature, bringing hunters and nature users into otherwise wilderness locations 47 .” | unhoused_relations[6] |
| no type for this | railways → direct pressure — Railways exert direct pressure but do not enable access along most borders. — “While railways are an important component of our global transport system, their pressure on the environment differs in nature from that of our road networks. By modifying a linear swath of habitat, railways exert direct pressure where they are constructed, similar to roads. However, as passengers seldom disembark from trains in places other than rail stations, railways do not provide a means of accessing the natural environments along their borders. To map railways we used the same dataset as was used in the original footprint 31 , as no update of this dataset or alternate source has been developed. The direct pressure of railways where assigned a pressure score of 8 for a distance of 0.5 km on either side of the railway.” | unhoused_relations[7] |
| no type for this | Rivers → nature — Navigable rivers act as access conduits into nature. — “Like roads, coastlines and navigable rivers act as conduits for people to access nature. While all coastlines are theoretically navigable, for the purposes of the human footprint we only considered coasts 31 as navigable for 80 km either direction of signs of a human settlement, which were mapped as a night lights signal with a DN 28 greater than 6 within 4 km of the coast. We chose 80 km as an approximation of the distance a vessel can travel and return during daylight hours. As new settlements can arise to make new sections of coast navigable, coastal layers were generated for the years 1994 and 2009.” | unhoused_relations[8] |
| no type for this | individual pressure layers → alternate scoring schemes — Published component layers allow developers to construct alternative scores. — “The individual pressure maps were developed to be globally consistent, using a scoring approach originally developed by Sanderson and colleagues 13 . However, in some regions and for some species groups, alternate scores may be better suited for reflecting the pressures exerted by humans on nature. We therefore provide the individual pressure layers that compose the Human Footprint maps, thereby allowing data developers to create alternate scoring schemes that better suit their purposes, as well facilitating the addition of new or alternate data sources.” | unhoused_relations[9] |
| no type for this | uncertain → visual score — Discarding uncertain plots protects the visual score from cloud and coarse-resolution errors. — “For the visual interpretation, the extent of built environments, crop lands, pasture lands, roads, human settlements, infrastructures and navigable waterways, were recorded using a standard key for identifying these features ( Supplementary Appendix 1 ). Shape, size, texture and colour of features in the imagery were important characteristics for identifying human pressures on the environment. Interpretations were also marked as ‘certain’ or ‘uncertain’, and the year and resolution of the interpreted image was recorded. The 346 ‘uncertain’ plots were discarded, leaving 3114 validation plots ( Fig. 3a ). In general, plots were classified as ‘uncertain’ for two reasons; either because cloud cover obscured the image, or because only medium resolution (15 m) imagery was available for the plot, preventing accurate interpretation of the image. The human footprint score for each plot was determined in ArcGIS, and the visual and Human Footprint scores were then normalized to a 0–1 scale. As we only retained plots for which visual interpretations of the images were determined to be ‘certain’, we consider the visual score to be the true state of in-situ pressures for the plots.” | unhoused_relations[10] |
| no type for this | Human Footprint maps → conservative estimates — Missing present pressures makes the maps conservative estimates. — “There is strong agreement between the Human Footprint measure of pressure and pressures scored by visual interpretation of high resolution imagery. The RMSE for the 3114 validation plots was 0.125 on the normalized 0–1 scale, indicating an average error of approximately 13%. The Kappa statistic was 0.737 ( P <0.01), also indicating good agreement between the Human Footprint and the validation dataset. Of the 3114×1 km 2 validation plots, the Human Footprint scored 94 of them 20% higher than the visual score and 263 of them 20% lower. The remaining 2757 plots (88.5%) were within 20% agreement. While this represents good agreement, it appears that the Human Footprint is to some extent susceptible to mapping pressures as absent in locations where they are actually present. The maps should therefore be considered as conservative estimates of human pressures on the environment. The Kappa statistic measure of agreement is sensitive to the threshold used to consider plots as a ‘match’. If we apply a more stringent threshold for agreement of within 15% of one another, the Kappa statistic falls to 0.565 (moderate agreement), and if we apply a less stringent threshold of within 25%, the Kappa statistic increases to 0.856 (very high agreement).” | unhoused_relations[11] |
| no type for this | thresholds → Kappa statistic — Changing the match threshold changes the Kappa statistic substantially. — “There is strong agreement between the Human Footprint measure of pressure and pressures scored by visual interpretation of high resolution imagery. The RMSE for the 3114 validation plots was 0.125 on the normalized 0–1 scale, indicating an average error of approximately 13%. The Kappa statistic was 0.737 ( P <0.01), also indicating good agreement between the Human Footprint and the validation dataset. Of the 3114×1 km 2 validation plots, the Human Footprint scored 94 of them 20% higher than the visual score and 263 of them 20% lower. The remaining 2757 plots (88.5%) were within 20% agreement. While this represents good agreement, it appears that the Human Footprint is to some extent susceptible to mapping pressures as absent in locations where they are actually present. The maps should therefore be considered as conservative estimates of human pressures on the environment. The Kappa statistic measure of agreement is sensitive to the threshold used to consider plots as a ‘match’. If we apply a more stringent threshold for agreement of within 15% of one another, the Kappa statistic falls to 0.565 (moderate agreement), and if we apply a less stringent threshold of within 25%, the Kappa statistic increases to 0.856 (very high agreement).” | unhoused_relations[12] |
| no type for this | static → Human Footprint — Static pressure layers can underestimate Human Footprint expansion. — “Moreover, our work is subject to three primary limitations. First, like all attempts to map cumulative pressures we did not fully account for all human pressures. Some of the missing and static pressures, such as invasive species and pollution, may be closely associated with pressures we did consider, and therefore their inclusion may not affect our overall results. Second, a lack of available data resulted in three of our pressures being static through time, which would cause an underestimation of Human Footprint expansion if these pressures expanded at a higher than average rate. Third, the Human Footprint measures the pressure humans place on nature, not the realized ‘state’ or ‘impacts’ on natural systems or their biodiversity. Significant scope exists to determine how natural systems respond to cumulating human pressures, and if non-linearity or thresholds exist where pressures lead to accelerated impacts.” | unhoused_relations[13] |
| no type for this | Human Footprint → non-linearity — The dataset leaves open whether natural-system responses are nonlinear or thresholded. — “Moreover, our work is subject to three primary limitations. First, like all attempts to map cumulative pressures we did not fully account for all human pressures. Some of the missing and static pressures, such as invasive species and pollution, may be closely associated with pressures we did consider, and therefore their inclusion may not affect our overall results. Second, a lack of available data resulted in three of our pressures being static through time, which would cause an underestimation of Human Footprint expansion if these pressures expanded at a higher than average rate. Third, the Human Footprint measures the pressure humans place on nature, not the realized ‘state’ or ‘impacts’ on natural systems or their biodiversity. Significant scope exists to determine how natural systems respond to cumulating human pressures, and if non-linearity or thresholds exist where pressures lead to accelerated impacts.” | unhoused_relations[14] |
| no type for this | Human Footprint → thresholds — The dataset leaves open whether thresholds produce accelerated impacts. — “Moreover, our work is subject to three primary limitations. First, like all attempts to map cumulative pressures we did not fully account for all human pressures. Some of the missing and static pressures, such as invasive species and pollution, may be closely associated with pressures we did consider, and therefore their inclusion may not affect our overall results. Second, a lack of available data resulted in three of our pressures being static through time, which would cause an underestimation of Human Footprint expansion if these pressures expanded at a higher than average rate. Third, the Human Footprint measures the pressure humans place on nature, not the realized ‘state’ or ‘impacts’ on natural systems or their biodiversity. Significant scope exists to determine how natural systems respond to cumulating human pressures, and if non-linearity or thresholds exist where pressures lead to accelerated impacts.” | unhoused_relations[15] |
| no type for this | Human Footprint maps → priority areas — Mapping pressures is a first step toward identifying protection or restoration priorities. — “Mapping human pressures to the environment is an essential first step to identifying priority areas for protection or restoration of natural systems. Understanding the spatial distribution of pressures, as well as their change through time, also provides insights for studies on macro-ecological patterns. The Human Footprint maps for 1993 and 2009 represent the first temporally-consistent maps of the human footprint, as well as much more up-to-date information on cumulative pressures than is currently available. Moreover, the 2009 Human Footprint map is the first cumulative pressure map to have undergone an accuracy assessment.” | unhoused_relations[16] |
| no type for this | Human Footprint map → Human Footprint maps — The authors reserve the singular name for unaltered distributed data, not user-created alternatives. — “While we welcome the opportunity to contribute intellectually and as co-authors to research projects that incorporate our datasets into their work, we make the data freely available without restriction for non-commercial use and redistribution. The data may be altered from their original form, and redistributed if done so free of charge and with a full description of any alterations to the original data. We do however ask that term ‘Human Footprint map’ be used only when referring to the unaltered data in the Human Footprint 7-zip file, and not to alternative versions of the data created by data users, and that the data be cited following the template at the end of this manuscript.” | unhoused_relations[17] |
| limitation | Remote sensing alone has difficulty detecting low-intensity linear infrastructure and pasture and can confuse natural with anthropogenic cover. — “Human pressures on the environment are the actions taken by humans with the potential to harm nature 1 , 2 . Cumulative pressure mapping measures the breadth of these pressures by coupling top-down remote sensing of land cover change with data on additional human pressures collected ‘bottom-up’ through systematic surveys and modelling 3 , 4 . The method circumvents the limitations of using remote sensing alone, which has difficulty in detecting low intensity pressures 5 , such as linear infrastructures 6 and pasture lands 7 , and often confounds natural and anthropogenic land covers in arid and mosaic environments 8 .” | limitations[0] |
| limitation | The original Human Footprint was a static and dated early-1990s view. — “Cumulative pressure maps have been developed at regional 9 , 10 and global scales 11 , 12 . The ‘Human Footprint’ was first released in 2002 using data primarily from the early 1990s (approximately 1993) on eight human pressures globally, making it the most complete, highest resolution and globally-consistent terrestrial dataset on cumulative human pressures on the environment 13 . It has been used in a large number of ecological and conservation analyses, and still receives around 100 citations each year, particularly from its data users. However, the Human Footprint is a static and dated view of human pressures on the environment. With many of Earth’s systems experiencing pressures close to or beyond safe levels 14 , there is a strong need for an up-to-date understanding of the spatial and temporal trends in human pressures.” | limitations[1] |
| limitation | Three pressure layers lacked a second time point and were treated as static. — “To create the Human Footprint maps we adopted the methods developed by Sanderson and colleagues 13 . Data on human pressures in 1993 and 2009 were collected or developed for: 1) the extent of built environments, 2) population density, 3) electric infrastructure, 4) crop lands, 5) pasture lands, 6) roads, 7) railways, and 8) navigable waterways, which are described in detail below ( Fig. 1 , step 1). To facilitate comparison across pressures we placed each human pressure within a 0–10 scale ( Fig. 1 , step 2), weighted within that range according to estimates of their relative levels of human pressure following Sanderson et al. 13 . The resulting standardized pressures were then summed together to create the standardized Human Footprint maps for all non-Antarctic land areas ( Fig. 1 , step 3). Pressures are not intended to be mutually exclusive, and many will co-occur in the same location. Three pressures only had data from a single time period, and these are treated as static in the Human Footprint maps.” | limitations[2] |
| limitation | DMSP-OLS over-glow and gas flaring can be falsely detected as built environments. — “The DMSP-OLS has limitations for the purpose of mapping human settlements, including hyper sensitivity of the sensors causing detection of over-glow adjacent to built environments 30 and bright lights associated with gas flaring from oil production facilities 29 . However, no other data exist to map built environments in a consistent way globally over our time horizon. While other datasets provide a one year snap shot of urban extent, they cannot be compared across time due to large differences in the methodologies used 31–33 , and the wildly contrasting extents in mapped built environments.” | limitations[3] |
| limitation | No alternative global settlement dataset was temporally comparable across the study horizon. — “The DMSP-OLS has limitations for the purpose of mapping human settlements, including hyper sensitivity of the sensors causing detection of over-glow adjacent to built environments 30 and bright lights associated with gas flaring from oil production facilities 29 . However, no other data exist to map built environments in a consistent way globally over our time horizon. While other datasets provide a one year snap shot of urban extent, they cannot be compared across time due to large differences in the methodologies used 31–33 , and the wildly contrasting extents in mapped built environments.” | limitations[4] |
| limitation | gROADS omits minor roads and inconsistently mapped private roads and trails. — “We acquired data on the distribution of roads from gROADS 48 , and excluded all trails and private roads, which were inconsistently mapped, with only a subset of countries mapping their linear infrastructure to this resolution. The dataset is the most comprehensive publicly available database on roads, which compiles nationally mapped road data spanning the period 1980–2000 and has a spatial accuracy of around 500 m. The gROADS data do not include all minor roads, and therefore should be viewed as a map of the major roadways. We mapped the direct and indirect influence of roads by assigning an pressure score of 8 for 0.5 km out for either side of roads, and access pressures were awarded a score of 4 at 0.5 km and decaying exponentially out to 15 km either side of the road ( Table 2 ).” | limitations[5] |
| limitation | Comparable high-resolution imagery was unavailable for 1993, so only the 2009 map was validated. — “High resolution images were used to visually interpret human pressures in 3460×1 km 2 sample plots randomly located across the Earth’s non-Antarctic land areas ( Fig. 3a ). Images for these plots were obtained from World Imagery 52 , which provides one meter or better satellite and aerial imagery in many parts of the world and lower resolution satellite imagery worldwide. The map features 0.3 m resolution imagery across the continental United States and parts of Western Europe, as well as many parts of the world, with concentrations in South America, Eastern Europe, India, Japan, the Middle East and Northern Africa, Southern Africa, Australia, and New Zealand. The imagery used for the validation plots had a median resolution of 0.5 meters and a median acquisition year of 2010. Comparable imagery was not available for the 1993 time period, and therefore only the 2009 map underwent validation.” | limitations[6] |
| limitation | The map sometimes records pressures as absent where visual validation finds them, making it conservative. — “There is strong agreement between the Human Footprint measure of pressure and pressures scored by visual interpretation of high resolution imagery. The RMSE for the 3114 validation plots was 0.125 on the normalized 0–1 scale, indicating an average error of approximately 13%. The Kappa statistic was 0.737 ( P <0.01), also indicating good agreement between the Human Footprint and the validation dataset. Of the 3114×1 km 2 validation plots, the Human Footprint scored 94 of them 20% higher than the visual score and 263 of them 20% lower. The remaining 2757 plots (88.5%) were within 20% agreement. While this represents good agreement, it appears that the Human Footprint is to some extent susceptible to mapping pressures as absent in locations where they are actually present. The maps should therefore be considered as conservative estimates of human pressures on the environment. The Kappa statistic measure of agreement is sensitive to the threshold used to consider plots as a ‘match’. If we apply a more stringent threshold for agreement of within 15% of one another, the Kappa statistic falls to 0.565 (moderate agreement), and if we apply a less stringent threshold of within 25%, the Kappa statistic increases to 0.856 (very high agreement).” | limitations[7] |
| limitation | Kappa agreement depends strongly on the chosen match threshold. — “There is strong agreement between the Human Footprint measure of pressure and pressures scored by visual interpretation of high resolution imagery. The RMSE for the 3114 validation plots was 0.125 on the normalized 0–1 scale, indicating an average error of approximately 13%. The Kappa statistic was 0.737 ( P <0.01), also indicating good agreement between the Human Footprint and the validation dataset. Of the 3114×1 km 2 validation plots, the Human Footprint scored 94 of them 20% higher than the visual score and 263 of them 20% lower. The remaining 2757 plots (88.5%) were within 20% agreement. While this represents good agreement, it appears that the Human Footprint is to some extent susceptible to mapping pressures as absent in locations where they are actually present. The maps should therefore be considered as conservative estimates of human pressures on the environment. The Kappa statistic measure of agreement is sensitive to the threshold used to consider plots as a ‘match’. If we apply a more stringent threshold for agreement of within 15% of one another, the Kappa statistic falls to 0.565 (moderate agreement), and if we apply a less stringent threshold of within 25%, the Kappa statistic increases to 0.856 (very high agreement).” | limitations[8] |
| limitation | The mapping does not include every human pressure, including some invasive species and pollution. — “Moreover, our work is subject to three primary limitations. First, like all attempts to map cumulative pressures we did not fully account for all human pressures. Some of the missing and static pressures, such as invasive species and pollution, may be closely associated with pressures we did consider, and therefore their inclusion may not affect our overall results. Second, a lack of available data resulted in three of our pressures being static through time, which would cause an underestimation of Human Footprint expansion if these pressures expanded at a higher than average rate. Third, the Human Footprint measures the pressure humans place on nature, not the realized ‘state’ or ‘impacts’ on natural systems or their biodiversity. Significant scope exists to determine how natural systems respond to cumulating human pressures, and if non-linearity or thresholds exist where pressures lead to accelerated impacts.” | limitations[9] |
| limitation | Static layers can underestimate expansion when those pressures grow faster than average. — “Moreover, our work is subject to three primary limitations. First, like all attempts to map cumulative pressures we did not fully account for all human pressures. Some of the missing and static pressures, such as invasive species and pollution, may be closely associated with pressures we did consider, and therefore their inclusion may not affect our overall results. Second, a lack of available data resulted in three of our pressures being static through time, which would cause an underestimation of Human Footprint expansion if these pressures expanded at a higher than average rate. Third, the Human Footprint measures the pressure humans place on nature, not the realized ‘state’ or ‘impacts’ on natural systems or their biodiversity. Significant scope exists to determine how natural systems respond to cumulating human pressures, and if non-linearity or thresholds exist where pressures lead to accelerated impacts.” | limitations[10] |
| limitation | Human Footprint measures pressure rather than realized ecosystem state or biodiversity impact. — “Moreover, our work is subject to three primary limitations. First, like all attempts to map cumulative pressures we did not fully account for all human pressures. Some of the missing and static pressures, such as invasive species and pollution, may be closely associated with pressures we did consider, and therefore their inclusion may not affect our overall results. Second, a lack of available data resulted in three of our pressures being static through time, which would cause an underestimation of Human Footprint expansion if these pressures expanded at a higher than average rate. Third, the Human Footprint measures the pressure humans place on nature, not the realized ‘state’ or ‘impacts’ on natural systems or their biodiversity. Significant scope exists to determine how natural systems respond to cumulating human pressures, and if non-linearity or thresholds exist where pressures lead to accelerated impacts.” | limitations[11] |
| limitation | Responses may be nonlinear or exhibit thresholds that accelerate impacts, which the dataset does not determine. — “Moreover, our work is subject to three primary limitations. First, like all attempts to map cumulative pressures we did not fully account for all human pressures. Some of the missing and static pressures, such as invasive species and pollution, may be closely associated with pressures we did consider, and therefore their inclusion may not affect our overall results. Second, a lack of available data resulted in three of our pressures being static through time, which would cause an underestimation of Human Footprint expansion if these pressures expanded at a higher than average rate. Third, the Human Footprint measures the pressure humans place on nature, not the realized ‘state’ or ‘impacts’ on natural systems or their biodiversity. Significant scope exists to determine how natural systems respond to cumulating human pressures, and if non-linearity or thresholds exist where pressures lead to accelerated impacts.” | limitations[12] |
Content tags
content-observations:
finding · quantity · variable · definition · key-term — grounded in the full text| finding | The maps are the first temporally consistent global Human Footprint series for 1993 and 2009. — “Remotely-sensed and bottom-up survey information were compiled on eight variables measuring the direct and indirect human pressures on the environment globally in 1993 and 2009. This represents not only the most current information of its type, but also the first temporally-consistent set of Human Footprint maps. Data on human pressures were acquired or developed for: 1) built environments, 2) population density, 3) electric infrastructure, 4) crop lands, 5) pasture lands, 6) roads, 7) railways, and 8) navigable waterways. Pressures were then overlaid to create the standardized Human Footprint maps for all non-Antarctic land areas. A validation analysis using scored pressures from 3114×1 km 2 random sample plots revealed strong agreement with the Human Footprint maps. We anticipate that the Human Footprint maps will find a range of uses as proxies for human disturbance of natural systems. The updated maps should provide an increased understanding of the human pressures that drive macro-ecological patterns, as well as for tracking environmental change and informing conservation science and application.” | content_tags.findings[0] |
| finding | Visual scores from 3,114 one-square-kilometre plots strongly agree with the mapped Human Footprint. — “Remotely-sensed and bottom-up survey information were compiled on eight variables measuring the direct and indirect human pressures on the environment globally in 1993 and 2009. This represents not only the most current information of its type, but also the first temporally-consistent set of Human Footprint maps. Data on human pressures were acquired or developed for: 1) built environments, 2) population density, 3) electric infrastructure, 4) crop lands, 5) pasture lands, 6) roads, 7) railways, and 8) navigable waterways. Pressures were then overlaid to create the standardized Human Footprint maps for all non-Antarctic land areas. A validation analysis using scored pressures from 3114×1 km 2 random sample plots revealed strong agreement with the Human Footprint maps. We anticipate that the Human Footprint maps will find a range of uses as proxies for human disturbance of natural systems. The updated maps should provide an increased understanding of the human pressures that drive macro-ecological patterns, as well as for tracking environmental change and informing conservation science and application.” | content_tags.findings[1] |
| finding | The original Human Footprint remained widely used but was a static and dated view of pressure. — “Cumulative pressure maps have been developed at regional 9 , 10 and global scales 11 , 12 . The ‘Human Footprint’ was first released in 2002 using data primarily from the early 1990s (approximately 1993) on eight human pressures globally, making it the most complete, highest resolution and globally-consistent terrestrial dataset on cumulative human pressures on the environment 13 . It has been used in a large number of ecological and conservation analyses, and still receives around 100 citations each year, particularly from its data users. However, the Human Footprint is a static and dated view of human pressures on the environment. With many of Earth’s systems experiencing pressures close to or beyond safe levels 14 , there is a strong need for an up-to-date understanding of the spatial and temporal trends in human pressures.” | content_tags.findings[2] |
| finding | The new maps support change-over-time analysis and supply the first validation of a cumulative pressure map. — “Here we use the Human Footprint methodology 13 to compile remotely-sensed and bottom-up survey information on eight variables measuring the direct and indirect human pressures on the environment in 1993 and 2009. This synthesis represents not only the most current information of its type, but also the first temporally-consistent set of Human Footprint maps, allowing for analyses of change over time. We also provide the first validation of a cumulative pressure map by adopting methods from remote sensing 15 to visually interpret human pressures in high resolution (median=0.5 m) imagery from 3114 1 km 2 random sample plots globally ( Supplementary Appendix 1 ). We then determine the level of agreement between these visually interpreted pressures and those mapped by the Human Footprint.” | content_tags.findings[3] |
| finding | Recent economic and population growth greatly exceeded Human Footprint growth, while biodiverse regions were disproportionately affected. — “The Human footprint maps provide information on where humans are exerting pressure on natural systems, altering them from their natural states. They also provide information on where these pressures are absent, and ecosystems are likely to be operating in a more natural state. These pressure-free lands represent candidate sites for consideration as ‘Wilderness’ 16 , 17 . The new Human Footprint maps have already been used to show that recent economic and population growth has far outstripped increases in the Human Footprint, yet the most biologically diverse regions of Earth have been disproportionately impacted 18 . We anticipate that the 1993 and 2009 Human Footprint maps will find a range of additional uses, such as serving as proxies for human disturbance and wilderness, including understanding the role of human pressures in driving macro-ecological patterns 19 , 20 , species extinction risk and distribution analyses 21 , dispersal ecology 22 , conservation science and decision making 23 , and tracking progress toward policy commitments to conservation 23 , among others.” | content_tags.findings[4] |
| finding | Eight standardized pressure layers were summed, with overlapping pressures allowed to co-occur. — “To create the Human Footprint maps we adopted the methods developed by Sanderson and colleagues 13 . Data on human pressures in 1993 and 2009 were collected or developed for: 1) the extent of built environments, 2) population density, 3) electric infrastructure, 4) crop lands, 5) pasture lands, 6) roads, 7) railways, and 8) navigable waterways, which are described in detail below ( Fig. 1 , step 1). To facilitate comparison across pressures we placed each human pressure within a 0–10 scale ( Fig. 1 , step 2), weighted within that range according to estimates of their relative levels of human pressure following Sanderson et al. 13 . The resulting standardized pressures were then summed together to create the standardized Human Footprint maps for all non-Antarctic land areas ( Fig. 1 , step 3). Pressures are not intended to be mutually exclusive, and many will co-occur in the same location. Three pressures only had data from a single time period, and these are treated as static in the Human Footprint maps.” | content_tags.findings[5] |
| finding | The non-Antarctic terrestrial analysis contains about 134.1 million one-square-kilometre pixels. — “We used ArcGIS 10.1 to integrate spatial data on human pressures. Analyses were conducted in Mollowedie equal area projection at the 1 km 2 resolution, yielding ~134.1 million pixels for Earth’s non-Antarctic terrestrial surface. For any grid cell, the Human Footprint can range between 0–50. The following sections and Table 1 (available online only) describe in detail the source data for each pressure, the processing steps applied, and the rationale behind the pressure weighting, and the output datasets created.” | content_tags.findings[6] |
| finding | DMSP-OLS is temporally consistent but over-detects glow and gas flaring near settlements. — “The DMSP-OLS has limitations for the purpose of mapping human settlements, including hyper sensitivity of the sensors causing detection of over-glow adjacent to built environments 30 and bright lights associated with gas flaring from oil production facilities 29 . However, no other data exist to map built environments in a consistent way globally over our time horizon. While other datasets provide a one year snap shot of urban extent, they cannot be compared across time due to large differences in the methodologies used 31–33 , and the wildly contrasting extents in mapped built environments.” | content_tags.findings[7] |
| finding | Pasture is the most extensive direct human pressure and covers 22 percent of Earth's land base. — “Pasture lands cover 22% of the Earth’s land base or almost twice that of agricultural crops 42 , making them the most extensive direct human pressure on the environment. Land grazed by domesticated herbivores is often degraded through a combination of fencing, intensive browsing, soil compaction, invasive grasses and other species, and altered fire regimes 43 . We mapped grazing lands for the year 2000 using a spatial dataset that combines agricultural census data with satellite derived land cover to map pasture extent 42 . We assigned pasture a pressure score of 4, which was then scaled from 0–4 using the percent pasture for each 1 km 2 pixel.” | content_tags.findings[8] |
| finding | Roads drive conversion, mortality, fragmentation, edge effects, and human access to nature. — “As one of humanity’s most prolific linear infrastructures, roads are an important direct driver of habitat conversion 44 . Beyond simply reducing the extent of suitable habitat, roads can act as population sinks for many species through traffic induced mortality 45 . Roads also fragment otherwise contiguous blocks of habitat, and create edge effects such as reduced humidity 6 and increased fire frequency that reach well beyond the roads immediate footprint 46 . Finally, roads provide conduits for humans to access nature, bringing hunters and nature users into otherwise wilderness locations 47 .” | content_tags.findings[9] |
| finding | Railways exert direct pressure but do not provide broad access along their borders. — “While railways are an important component of our global transport system, their pressure on the environment differs in nature from that of our road networks. By modifying a linear swath of habitat, railways exert direct pressure where they are constructed, similar to roads. However, as passengers seldom disembark from trains in places other than rail stations, railways do not provide a means of accessing the natural environments along their borders. To map railways we used the same dataset as was used in the original footprint 31 , as no update of this dataset or alternate source has been developed. The direct pressure of railways where assigned a pressure score of 8 for a distance of 0.5 km on either side of the railway.” | content_tags.findings[10] |
| finding | The retained visual validation plots excluded cloud-obscured and insufficient-resolution imagery. — “For the visual interpretation, the extent of built environments, crop lands, pasture lands, roads, human settlements, infrastructures and navigable waterways, were recorded using a standard key for identifying these features ( Supplementary Appendix 1 ). Shape, size, texture and colour of features in the imagery were important characteristics for identifying human pressures on the environment. Interpretations were also marked as ‘certain’ or ‘uncertain’, and the year and resolution of the interpreted image was recorded. The 346 ‘uncertain’ plots were discarded, leaving 3114 validation plots ( Fig. 3a ). In general, plots were classified as ‘uncertain’ for two reasons; either because cloud cover obscured the image, or because only medium resolution (15 m) imagery was available for the plot, preventing accurate interpretation of the image. The human footprint score for each plot was determined in ArcGIS, and the visual and Human Footprint scores were then normalized to a 0–1 scale. As we only retained plots for which visual interpretations of the images were determined to be ‘certain’, we consider the visual score to be the true state of in-situ pressures for the plots.” | content_tags.findings[11] |
| finding | Validation produced RMSE 0.125, Kappa 0.737, and 88.5 percent of plots within 20 percent agreement. — “There is strong agreement between the Human Footprint measure of pressure and pressures scored by visual interpretation of high resolution imagery. The RMSE for the 3114 validation plots was 0.125 on the normalized 0–1 scale, indicating an average error of approximately 13%. The Kappa statistic was 0.737 ( P <0.01), also indicating good agreement between the Human Footprint and the validation dataset. Of the 3114×1 km 2 validation plots, the Human Footprint scored 94 of them 20% higher than the visual score and 263 of them 20% lower. The remaining 2757 plots (88.5%) were within 20% agreement. While this represents good agreement, it appears that the Human Footprint is to some extent susceptible to mapping pressures as absent in locations where they are actually present. The maps should therefore be considered as conservative estimates of human pressures on the environment. The Kappa statistic measure of agreement is sensitive to the threshold used to consider plots as a ‘match’. If we apply a more stringent threshold for agreement of within 15% of one another, the Kappa statistic falls to 0.565 (moderate agreement), and if we apply a less stringent threshold of within 25%, the Kappa statistic increases to 0.856 (very high agreement).” | content_tags.findings[12] |
| finding | The map tends to miss pressures that are present and should be treated as a conservative estimate. — “There is strong agreement between the Human Footprint measure of pressure and pressures scored by visual interpretation of high resolution imagery. The RMSE for the 3114 validation plots was 0.125 on the normalized 0–1 scale, indicating an average error of approximately 13%. The Kappa statistic was 0.737 ( P <0.01), also indicating good agreement between the Human Footprint and the validation dataset. Of the 3114×1 km 2 validation plots, the Human Footprint scored 94 of them 20% higher than the visual score and 263 of them 20% lower. The remaining 2757 plots (88.5%) were within 20% agreement. While this represents good agreement, it appears that the Human Footprint is to some extent susceptible to mapping pressures as absent in locations where they are actually present. The maps should therefore be considered as conservative estimates of human pressures on the environment. The Kappa statistic measure of agreement is sensitive to the threshold used to consider plots as a ‘match’. If we apply a more stringent threshold for agreement of within 15% of one another, the Kappa statistic falls to 0.565 (moderate agreement), and if we apply a less stringent threshold of within 25%, the Kappa statistic increases to 0.856 (very high agreement).” | content_tags.findings[13] |
| finding | Agreement varies geographically and is weakest in temperate broadleaf/mixed forest and boreal biomes. — “While agreement is generally strong, there is some geographic variation in the RMSE results comparing the Human Footprint scores and those derived from visual interpretation ( Fig. 3b ). By calculating RMSE for all biomes that contain at least 100 of the 3114 sample plots, we found that agreement was strongest in the Tundra biome and the Temperate grasslands, savannas and shrublands biomes ( Table 4 ). Agreement was weakest in the Temperate broadleaf and mixed forest biome and the Boreal.” | content_tags.findings[14] |
| finding | The dataset measures human pressure rather than realized ecosystem state or biodiversity impacts. — “Moreover, our work is subject to three primary limitations. First, like all attempts to map cumulative pressures we did not fully account for all human pressures. Some of the missing and static pressures, such as invasive species and pollution, may be closely associated with pressures we did consider, and therefore their inclusion may not affect our overall results. Second, a lack of available data resulted in three of our pressures being static through time, which would cause an underestimation of Human Footprint expansion if these pressures expanded at a higher than average rate. Third, the Human Footprint measures the pressure humans place on nature, not the realized ‘state’ or ‘impacts’ on natural systems or their biodiversity. Significant scope exists to determine how natural systems respond to cumulating human pressures, and if non-linearity or thresholds exist where pressures lead to accelerated impacts.” | content_tags.findings[15] |
| quantity | eight — direct and indirect human-pressure variables — “Remotely-sensed and bottom-up survey information were compiled on eight variables measuring the direct and indirect human pressures on the environment globally in 1993 and 2009. This represents not only the most current information of its type, but also the first temporally-consistent set of Human Footprint maps. Data on human pressures were acquired or developed for: 1) built environments, 2) population density, 3) electric infrastructure, 4) crop lands, 5) pasture lands, 6) roads, 7) railways, and 8) navigable waterways. Pressures were then overlaid to create the standardized Human Footprint maps for all non-Antarctic land areas. A validation analysis using scored pressures from 3114×1 km 2 random sample plots revealed strong agreement with the Human Footprint maps. We anticipate that the Human Footprint maps will find a range of uses as proxies for human disturbance of natural systems. The updated maps should provide an increased understanding of the human pressures that drive macro-ecological patterns, as well as for tracking environmental change and informing conservation science and application.” | content_tags.quantities[0] |
| quantity | 1993 and 2009 — two temporally comparable map years — “Remotely-sensed and bottom-up survey information were compiled on eight variables measuring the direct and indirect human pressures on the environment globally in 1993 and 2009. This represents not only the most current information of its type, but also the first temporally-consistent set of Human Footprint maps. Data on human pressures were acquired or developed for: 1) built environments, 2) population density, 3) electric infrastructure, 4) crop lands, 5) pasture lands, 6) roads, 7) railways, and 8) navigable waterways. Pressures were then overlaid to create the standardized Human Footprint maps for all non-Antarctic land areas. A validation analysis using scored pressures from 3114×1 km 2 random sample plots revealed strong agreement with the Human Footprint maps. We anticipate that the Human Footprint maps will find a range of uses as proxies for human disturbance of natural systems. The updated maps should provide an increased understanding of the human pressures that drive macro-ecological patterns, as well as for tracking environmental change and informing conservation science and application.” | content_tags.quantities[1] |
| quantity | 0–10 — standardized scale for each individual pressure — “To create the Human Footprint maps we adopted the methods developed by Sanderson and colleagues 13 . Data on human pressures in 1993 and 2009 were collected or developed for: 1) the extent of built environments, 2) population density, 3) electric infrastructure, 4) crop lands, 5) pasture lands, 6) roads, 7) railways, and 8) navigable waterways, which are described in detail below ( Fig. 1 , step 1). To facilitate comparison across pressures we placed each human pressure within a 0–10 scale ( Fig. 1 , step 2), weighted within that range according to estimates of their relative levels of human pressure following Sanderson et al. 13 . The resulting standardized pressures were then summed together to create the standardized Human Footprint maps for all non-Antarctic land areas ( Fig. 1 , step 3). Pressures are not intended to be mutually exclusive, and many will co-occur in the same location. Three pressures only had data from a single time period, and these are treated as static in the Human Footprint maps.” | content_tags.quantities[2] |
| quantity | 0–50 — possible total Human Footprint range per grid cell — “We used ArcGIS 10.1 to integrate spatial data on human pressures. Analyses were conducted in Mollowedie equal area projection at the 1 km 2 resolution, yielding ~134.1 million pixels for Earth’s non-Antarctic terrestrial surface. For any grid cell, the Human Footprint can range between 0–50. The following sections and Table 1 (available online only) describe in detail the source data for each pressure, the processing steps applied, and the rationale behind the pressure weighting, and the output datasets created.” | content_tags.quantities[3] |
| quantity | ~134.1 million pixels — one-square-kilometre cells covering non-Antarctic land — “We used ArcGIS 10.1 to integrate spatial data on human pressures. Analyses were conducted in Mollowedie equal area projection at the 1 km 2 resolution, yielding ~134.1 million pixels for Earth’s non-Antarctic terrestrial surface. For any grid cell, the Human Footprint can range between 0–50. The following sections and Table 1 (available online only) describe in detail the source data for each pressure, the processing steps applied, and the rationale behind the pressure weighting, and the output datasets created.” | content_tags.quantities[4] |
| quantity | 22% — share of Earth's land base covered by pasture — “Pasture lands cover 22% of the Earth’s land base or almost twice that of agricultural crops 42 , making them the most extensive direct human pressure on the environment. Land grazed by domesticated herbivores is often degraded through a combination of fencing, intensive browsing, soil compaction, invasive grasses and other species, and altered fire regimes 43 . We mapped grazing lands for the year 2000 using a spatial dataset that combines agricultural census data with satellite derived land cover to map pasture extent 42 . We assigned pasture a pressure score of 4, which was then scaled from 0–4 using the percent pasture for each 1 km 2 pixel.” | content_tags.quantities[5] |
| quantity | 3114 — retained validation plots — “For the visual interpretation, the extent of built environments, crop lands, pasture lands, roads, human settlements, infrastructures and navigable waterways, were recorded using a standard key for identifying these features ( Supplementary Appendix 1 ). Shape, size, texture and colour of features in the imagery were important characteristics for identifying human pressures on the environment. Interpretations were also marked as ‘certain’ or ‘uncertain’, and the year and resolution of the interpreted image was recorded. The 346 ‘uncertain’ plots were discarded, leaving 3114 validation plots ( Fig. 3a ). In general, plots were classified as ‘uncertain’ for two reasons; either because cloud cover obscured the image, or because only medium resolution (15 m) imagery was available for the plot, preventing accurate interpretation of the image. The human footprint score for each plot was determined in ArcGIS, and the visual and Human Footprint scores were then normalized to a 0–1 scale. As we only retained plots for which visual interpretations of the images were determined to be ‘certain’, we consider the visual score to be the true state of in-situ pressures for the plots.” | content_tags.quantities[6] |
| quantity | 0.125 — global normalized RMSE — “There is strong agreement between the Human Footprint measure of pressure and pressures scored by visual interpretation of high resolution imagery. The RMSE for the 3114 validation plots was 0.125 on the normalized 0–1 scale, indicating an average error of approximately 13%. The Kappa statistic was 0.737 ( P <0.01), also indicating good agreement between the Human Footprint and the validation dataset. Of the 3114×1 km 2 validation plots, the Human Footprint scored 94 of them 20% higher than the visual score and 263 of them 20% lower. The remaining 2757 plots (88.5%) were within 20% agreement. While this represents good agreement, it appears that the Human Footprint is to some extent susceptible to mapping pressures as absent in locations where they are actually present. The maps should therefore be considered as conservative estimates of human pressures on the environment. The Kappa statistic measure of agreement is sensitive to the threshold used to consider plots as a ‘match’. If we apply a more stringent threshold for agreement of within 15% of one another, the Kappa statistic falls to 0.565 (moderate agreement), and if we apply a less stringent threshold of within 25%, the Kappa statistic increases to 0.856 (very high agreement).” | content_tags.quantities[7] |
| quantity | 0.737 — Kappa at the 20-percent match threshold — “There is strong agreement between the Human Footprint measure of pressure and pressures scored by visual interpretation of high resolution imagery. The RMSE for the 3114 validation plots was 0.125 on the normalized 0–1 scale, indicating an average error of approximately 13%. The Kappa statistic was 0.737 ( P <0.01), also indicating good agreement between the Human Footprint and the validation dataset. Of the 3114×1 km 2 validation plots, the Human Footprint scored 94 of them 20% higher than the visual score and 263 of them 20% lower. The remaining 2757 plots (88.5%) were within 20% agreement. While this represents good agreement, it appears that the Human Footprint is to some extent susceptible to mapping pressures as absent in locations where they are actually present. The maps should therefore be considered as conservative estimates of human pressures on the environment. The Kappa statistic measure of agreement is sensitive to the threshold used to consider plots as a ‘match’. If we apply a more stringent threshold for agreement of within 15% of one another, the Kappa statistic falls to 0.565 (moderate agreement), and if we apply a less stringent threshold of within 25%, the Kappa statistic increases to 0.856 (very high agreement).” | content_tags.quantities[8] |
| quantity | 88.5% — validation plots within 20-percent agreement — “There is strong agreement between the Human Footprint measure of pressure and pressures scored by visual interpretation of high resolution imagery. The RMSE for the 3114 validation plots was 0.125 on the normalized 0–1 scale, indicating an average error of approximately 13%. The Kappa statistic was 0.737 ( P <0.01), also indicating good agreement between the Human Footprint and the validation dataset. Of the 3114×1 km 2 validation plots, the Human Footprint scored 94 of them 20% higher than the visual score and 263 of them 20% lower. The remaining 2757 plots (88.5%) were within 20% agreement. While this represents good agreement, it appears that the Human Footprint is to some extent susceptible to mapping pressures as absent in locations where they are actually present. The maps should therefore be considered as conservative estimates of human pressures on the environment. The Kappa statistic measure of agreement is sensitive to the threshold used to consider plots as a ‘match’. If we apply a more stringent threshold for agreement of within 15% of one another, the Kappa statistic falls to 0.565 (moderate agreement), and if we apply a less stringent threshold of within 25%, the Kappa statistic increases to 0.856 (very high agreement).” | content_tags.quantities[9] |
| quantity | 0.565 — Kappa at the stricter 15-percent threshold — “There is strong agreement between the Human Footprint measure of pressure and pressures scored by visual interpretation of high resolution imagery. The RMSE for the 3114 validation plots was 0.125 on the normalized 0–1 scale, indicating an average error of approximately 13%. The Kappa statistic was 0.737 ( P <0.01), also indicating good agreement between the Human Footprint and the validation dataset. Of the 3114×1 km 2 validation plots, the Human Footprint scored 94 of them 20% higher than the visual score and 263 of them 20% lower. The remaining 2757 plots (88.5%) were within 20% agreement. While this represents good agreement, it appears that the Human Footprint is to some extent susceptible to mapping pressures as absent in locations where they are actually present. The maps should therefore be considered as conservative estimates of human pressures on the environment. The Kappa statistic measure of agreement is sensitive to the threshold used to consider plots as a ‘match’. If we apply a more stringent threshold for agreement of within 15% of one another, the Kappa statistic falls to 0.565 (moderate agreement), and if we apply a less stringent threshold of within 25%, the Kappa statistic increases to 0.856 (very high agreement).” | content_tags.quantities[10] |
| quantity | 0.856 — Kappa at the looser 25-percent threshold — “There is strong agreement between the Human Footprint measure of pressure and pressures scored by visual interpretation of high resolution imagery. The RMSE for the 3114 validation plots was 0.125 on the normalized 0–1 scale, indicating an average error of approximately 13%. The Kappa statistic was 0.737 ( P <0.01), also indicating good agreement between the Human Footprint and the validation dataset. Of the 3114×1 km 2 validation plots, the Human Footprint scored 94 of them 20% higher than the visual score and 263 of them 20% lower. The remaining 2757 plots (88.5%) were within 20% agreement. While this represents good agreement, it appears that the Human Footprint is to some extent susceptible to mapping pressures as absent in locations where they are actually present. The maps should therefore be considered as conservative estimates of human pressures on the environment. The Kappa statistic measure of agreement is sensitive to the threshold used to consider plots as a ‘match’. If we apply a more stringent threshold for agreement of within 15% of one another, the Kappa statistic falls to 0.565 (moderate agreement), and if we apply a less stringent threshold of within 25%, the Kappa statistic increases to 0.856 (very high agreement).” | content_tags.quantities[11] |
| variable | built environments (maximum-weight direct pressure layer) — “Built environments are human produced areas that provide the setting for human activity. In the context of the human footprint, we take these areas to be primarily urban settings, including buildings, paved land and urban parks. Built environments do not provide viable habitats for many species of conservation concern, nor do they provide high levels of ecosystem services 24–27 . As such, built environments were assigned a pressure score of 10.” | content_tags.variables[0] |
| variable | population density (logarithmically scaled local-pressure layer) — “Human population density is scored in this way under the assumption that the pressures people induce on their local natural systems increase logarithmically with increasing population density, and saturate at a level of 1000 people per km 2 .” | content_tags.variables[1] |
| variable | electric infrastructure (night-light-derived pressure layer) — “To create the Human Footprint maps we adopted the methods developed by Sanderson and colleagues 13 . Data on human pressures in 1993 and 2009 were collected or developed for: 1) the extent of built environments, 2) population density, 3) electric infrastructure, 4) crop lands, 5) pasture lands, 6) roads, 7) railways, and 8) navigable waterways, which are described in detail below ( Fig. 1 , step 1). To facilitate comparison across pressures we placed each human pressure within a 0–10 scale ( Fig. 1 , step 2), weighted within that range according to estimates of their relative levels of human pressure following Sanderson et al. 13 . The resulting standardized pressures were then summed together to create the standardized Human Footprint maps for all non-Antarctic land areas ( Fig. 1 , step 3). Pressures are not intended to be mutually exclusive, and many will co-occur in the same location. Three pressures only had data from a single time period, and these are treated as static in the Human Footprint maps.” | content_tags.variables[2] |
| variable | crop lands (intensive-agriculture pressure layer) — “To create the Human Footprint maps we adopted the methods developed by Sanderson and colleagues 13 . Data on human pressures in 1993 and 2009 were collected or developed for: 1) the extent of built environments, 2) population density, 3) electric infrastructure, 4) crop lands, 5) pasture lands, 6) roads, 7) railways, and 8) navigable waterways, which are described in detail below ( Fig. 1 , step 1). To facilitate comparison across pressures we placed each human pressure within a 0–10 scale ( Fig. 1 , step 2), weighted within that range according to estimates of their relative levels of human pressure following Sanderson et al. 13 . The resulting standardized pressures were then summed together to create the standardized Human Footprint maps for all non-Antarctic land areas ( Fig. 1 , step 3). Pressures are not intended to be mutually exclusive, and many will co-occur in the same location. Three pressures only had data from a single time period, and these are treated as static in the Human Footprint maps.” | content_tags.variables[3] |
| variable | pasture lands (grazing pressure layer) — “Pasture lands cover 22% of the Earth’s land base or almost twice that of agricultural crops 42 , making them the most extensive direct human pressure on the environment. Land grazed by domesticated herbivores is often degraded through a combination of fencing, intensive browsing, soil compaction, invasive grasses and other species, and altered fire regimes 43 . We mapped grazing lands for the year 2000 using a spatial dataset that combines agricultural census data with satellite derived land cover to map pasture extent 42 . We assigned pasture a pressure score of 4, which was then scaled from 0–4 using the percent pasture for each 1 km 2 pixel.” | content_tags.variables[4] |
| variable | roads (direct, edge-effect, and access pressure layer) — “We acquired data on the distribution of roads from gROADS 48 , and excluded all trails and private roads, which were inconsistently mapped, with only a subset of countries mapping their linear infrastructure to this resolution. The dataset is the most comprehensive publicly available database on roads, which compiles nationally mapped road data spanning the period 1980–2000 and has a spatial accuracy of around 500 m. The gROADS data do not include all minor roads, and therefore should be viewed as a map of the major roadways. We mapped the direct and indirect influence of roads by assigning an pressure score of 8 for 0.5 km out for either side of roads, and access pressures were awarded a score of 4 at 0.5 km and decaying exponentially out to 15 km either side of the road ( Table 2 ).” | content_tags.variables[5] |
| variable | railways (direct linear pressure layer without diffuse access) — “While railways are an important component of our global transport system, their pressure on the environment differs in nature from that of our road networks. By modifying a linear swath of habitat, railways exert direct pressure where they are constructed, similar to roads. However, as passengers seldom disembark from trains in places other than rail stations, railways do not provide a means of accessing the natural environments along their borders. To map railways we used the same dataset as was used in the original footprint 31 , as no update of this dataset or alternate source has been developed. The direct pressure of railways where assigned a pressure score of 8 for a distance of 0.5 km on either side of the railway.” | content_tags.variables[6] |
| variable | navigable waterways (water-based access pressure layer) — “Navigable rivers layers were created for the years 1994 and 2009, and combined with the navigable coasts and inland seas layers to create the final navigable waterways layers. The access pressure from navigable water bodies were awarded a score of 4 adjacent to the water body, decaying exponentially out to 15 km.” | content_tags.variables[7] |
| variable | pressure score (zero-to-ten relative intensity assigned to each layer) — “To create the Human Footprint maps we adopted the methods developed by Sanderson and colleagues 13 . Data on human pressures in 1993 and 2009 were collected or developed for: 1) the extent of built environments, 2) population density, 3) electric infrastructure, 4) crop lands, 5) pasture lands, 6) roads, 7) railways, and 8) navigable waterways, which are described in detail below ( Fig. 1 , step 1). To facilitate comparison across pressures we placed each human pressure within a 0–10 scale ( Fig. 1 , step 2), weighted within that range according to estimates of their relative levels of human pressure following Sanderson et al. 13 . The resulting standardized pressures were then summed together to create the standardized Human Footprint maps for all non-Antarctic land areas ( Fig. 1 , step 3). Pressures are not intended to be mutually exclusive, and many will co-occur in the same location. Three pressures only had data from a single time period, and these are treated as static in the Human Footprint maps.” | content_tags.variables[8] |
| variable | visual score (interpreted in-situ pressure used as validation reference) — “For the visual interpretation, the extent of built environments, crop lands, pasture lands, roads, human settlements, infrastructures and navigable waterways, were recorded using a standard key for identifying these features ( Supplementary Appendix 1 ). Shape, size, texture and colour of features in the imagery were important characteristics for identifying human pressures on the environment. Interpretations were also marked as ‘certain’ or ‘uncertain’, and the year and resolution of the interpreted image was recorded. The 346 ‘uncertain’ plots were discarded, leaving 3114 validation plots ( Fig. 3a ). In general, plots were classified as ‘uncertain’ for two reasons; either because cloud cover obscured the image, or because only medium resolution (15 m) imagery was available for the plot, preventing accurate interpretation of the image. The human footprint score for each plot was determined in ArcGIS, and the visual and Human Footprint scores were then normalized to a 0–1 scale. As we only retained plots for which visual interpretations of the images were determined to be ‘certain’, we consider the visual score to be the true state of in-situ pressures for the plots.” | content_tags.variables[9] |
| variable | RMSE (continuous prediction-error metric) — “Two statistics were used to determine Human Footprint performance, root mean squared error (RMSE) 53 and the Cohen kappa statistic of agreement 54 . The RMSE is a dimensioned (expresses average error in the units of variable of interest) error metric for numerical predictions, and tends to heavily punish large errors. The Kappa statistic expresses the agreement between two categorical datasets corrected for the expected agreement, which is based on a random allocation given the relative class sizes. When calculating the kappa statistic, the Human Footprint score was considered as a match to the visual score if they were within 20% of one another on the 0–1 scale.” | content_tags.variables[10] |
| variable | Kappa statistic (threshold-sensitive categorical agreement metric) — “Two statistics were used to determine Human Footprint performance, root mean squared error (RMSE) 53 and the Cohen kappa statistic of agreement 54 . The RMSE is a dimensioned (expresses average error in the units of variable of interest) error metric for numerical predictions, and tends to heavily punish large errors. The Kappa statistic expresses the agreement between two categorical datasets corrected for the expected agreement, which is based on a random allocation given the relative class sizes. When calculating the kappa statistic, the Human Footprint score was considered as a match to the visual score if they were within 20% of one another on the 0–1 scale.” | content_tags.variables[11] |
| definition | human pressures on the environment — “Human pressures on the environment are the actions taken by humans with the potential to harm nature 1 , 2 . Cumulative pressure mapping measures the breadth of these pressures by coupling top-down remote sensing of land cover change with data on additional human pressures collected ‘bottom-up’ through systematic surveys and modelling 3 , 4 . The method circumvents the limitations of using remote sensing alone, which has difficulty in detecting low intensity pressures 5 , such as linear infrastructures 6 and pasture lands 7 , and often confounds natural and anthropogenic land covers in arid and mosaic environments 8 .” | content_tags.definitions[0] |
| definition | cumulative pressure mapping — “Human pressures on the environment are the actions taken by humans with the potential to harm nature 1 , 2 . Cumulative pressure mapping measures the breadth of these pressures by coupling top-down remote sensing of land cover change with data on additional human pressures collected ‘bottom-up’ through systematic surveys and modelling 3 , 4 . The method circumvents the limitations of using remote sensing alone, which has difficulty in detecting low intensity pressures 5 , such as linear infrastructures 6 and pasture lands 7 , and often confounds natural and anthropogenic land covers in arid and mosaic environments 8 .” | content_tags.definitions[1] |
| definition | Human Footprint — “To create the Human Footprint maps we adopted the methods developed by Sanderson and colleagues 13 . Data on human pressures in 1993 and 2009 were collected or developed for: 1) the extent of built environments, 2) population density, 3) electric infrastructure, 4) crop lands, 5) pasture lands, 6) roads, 7) railways, and 8) navigable waterways, which are described in detail below ( Fig. 1 , step 1). To facilitate comparison across pressures we placed each human pressure within a 0–10 scale ( Fig. 1 , step 2), weighted within that range according to estimates of their relative levels of human pressure following Sanderson et al. 13 . The resulting standardized pressures were then summed together to create the standardized Human Footprint maps for all non-Antarctic land areas ( Fig. 1 , step 3). Pressures are not intended to be mutually exclusive, and many will co-occur in the same location. Three pressures only had data from a single time period, and these are treated as static in the Human Footprint maps.” | content_tags.definitions[2] |
| definition | built environments — “Built environments are human produced areas that provide the setting for human activity. In the context of the human footprint, we take these areas to be primarily urban settings, including buildings, paved land and urban parks. Built environments do not provide viable habitats for many species of conservation concern, nor do they provide high levels of ecosystem services 24–27 . As such, built environments were assigned a pressure score of 10.” | content_tags.definitions[3] |
| definition | population density — “Human population density was mapped using the Gridded Population of the World dataset developed by the Centre for International Earth Science Information Network (CIESEN) 37 . The dataset provides a ~4 km×~4 km gridded summary of population census data for the years 1990 and 2010, which we downscaled using bilinear sampling in ArcGIS 10.1 to match the 1 km 2 resolution of the other datasets. For all locations with more than 1000 people·km −2 , we assigned a pressure score of 10 ( Table 2 ). For more sparsely populated areas with densities lower than 1000 people·km −2 , we logarithmically scaled the pressure score using,” | content_tags.definitions[4] |
| definition | night-time lights — “The high sensitivity of the DMSP-OLS 28 dataset provides a means for mapping the sparser electric infrastructure typical of more rural and suburban areas. In 2009, 79% of the lights registered in the DMSP-OLS dataset had a Digital Number less than 20, and are therefore not included in our ‘built environments’ layers. However, these lower DN values are often important human infrastructures, such as rural housing or working landscapes, with associated pressures on natural environments.” | content_tags.definitions[5] |
| definition | crop lands — “Crop lands vary in their structure from intensely managed monocultures receiving high inputs of pesticides and fertilizers, to mosaic agricultures such as slash and burn methods that can support intermediate levels of natural values 38 , 39 . For the purposes of the human footprint, we focused only on intensive agriculture because of its greater direct pressure on the environment, as well as to circumvent the shortcomings of using remotely sensed data to map mosaic agriculture globally, namely the tendency to confound agriculture mosaics with natural woodland and savannah ecosystems 8 .” | content_tags.definitions[6] |
| definition | pasture lands — “Pasture lands cover 22% of the Earth’s land base or almost twice that of agricultural crops 42 , making them the most extensive direct human pressure on the environment. Land grazed by domesticated herbivores is often degraded through a combination of fencing, intensive browsing, soil compaction, invasive grasses and other species, and altered fire regimes 43 . We mapped grazing lands for the year 2000 using a spatial dataset that combines agricultural census data with satellite derived land cover to map pasture extent 42 . We assigned pasture a pressure score of 4, which was then scaled from 0–4 using the percent pasture for each 1 km 2 pixel.” | content_tags.definitions[7] |
| definition | navigable rivers — “Rivers were considered as navigable if their depth was greater than 2 m and there were signs of nighttime lights (DN>=6) within 4km of their banks, or if contiguous with a navigable coast or large inland lake, and then for a distance of 80 km or until stream depth is likely to prevent boat traffic ( Table 2 ). To map rivers and their depth we used the hydrosheds (hydrological data and maps based on shuttle elevation derivatives at multiple scales) 49 dataset on stream discharge, and the following formulae from 50 , 51 :” | content_tags.definitions[8] |
| definition | RMSE — “Two statistics were used to determine Human Footprint performance, root mean squared error (RMSE) 53 and the Cohen kappa statistic of agreement 54 . The RMSE is a dimensioned (expresses average error in the units of variable of interest) error metric for numerical predictions, and tends to heavily punish large errors. The Kappa statistic expresses the agreement between two categorical datasets corrected for the expected agreement, which is based on a random allocation given the relative class sizes. When calculating the kappa statistic, the Human Footprint score was considered as a match to the visual score if they were within 20% of one another on the 0–1 scale.” | content_tags.definitions[9] |
| definition | Kappa statistic — “Two statistics were used to determine Human Footprint performance, root mean squared error (RMSE) 53 and the Cohen kappa statistic of agreement 54 . The RMSE is a dimensioned (expresses average error in the units of variable of interest) error metric for numerical predictions, and tends to heavily punish large errors. The Kappa statistic expresses the agreement between two categorical datasets corrected for the expected agreement, which is based on a random allocation given the relative class sizes. When calculating the kappa statistic, the Human Footprint score was considered as a match to the visual score if they were within 20% of one another on the 0–1 scale.” | content_tags.definitions[10] |
| definition | Human Footprint map — “While we welcome the opportunity to contribute intellectually and as co-authors to research projects that incorporate our datasets into their work, we make the data freely available without restriction for non-commercial use and redistribution. The data may be altered from their original form, and redistributed if done so free of charge and with a full description of any alterations to the original data. We do however ask that term ‘Human Footprint map’ be used only when referring to the unaltered data in the Human Footprint 7-zip file, and not to alternative versions of the data created by data users, and that the data be cited following the template at the end of this manuscript.” | content_tags.definitions[11] |
Concepts
terms this paper carries, each with the sentence it was cut from — the layer that joins one paper to another. Terms with no quote yet are on CONCEPTS.html.
| concept | Human pressures on the environment — not defined in this paper's own words; human actions capable of harming nature — “Human pressures on the environment are the actions taken by humans with the potential to harm nature 1 , 2 . Cumulative pressure mapping measures the breadth of these pressures by coupling top-down remote sensing of land cover change with data on additional human pressures collected ‘bottom-up’ through systematic surveys and modelling 3 , 4 . The method circumvents the limitations of using remote sensing alone, which has difficulty in detecting low intensity pressures 5 , such as linear infrastructures 6 and pasture lands 7 , and often confounds natural and anthropogenic land covers in arid and mosaic environments 8 .” | concepts[0] |
| concept | cumulative pressure mapping — not defined in this paper's own words; combined mapping of multiple direct and indirect pressures — “Human pressures on the environment are the actions taken by humans with the potential to harm nature 1 , 2 . Cumulative pressure mapping measures the breadth of these pressures by coupling top-down remote sensing of land cover change with data on additional human pressures collected ‘bottom-up’ through systematic surveys and modelling 3 , 4 . The method circumvents the limitations of using remote sensing alone, which has difficulty in detecting low intensity pressures 5 , such as linear infrastructures 6 and pasture lands 7 , and often confounds natural and anthropogenic land covers in arid and mosaic environments 8 .” | concepts[1] |
| concept | top-down remote sensing — not defined in this paper's own words; land-cover-change observation — “Human pressures on the environment are the actions taken by humans with the potential to harm nature 1 , 2 . Cumulative pressure mapping measures the breadth of these pressures by coupling top-down remote sensing of land cover change with data on additional human pressures collected ‘bottom-up’ through systematic surveys and modelling 3 , 4 . The method circumvents the limitations of using remote sensing alone, which has difficulty in detecting low intensity pressures 5 , such as linear infrastructures 6 and pasture lands 7 , and often confounds natural and anthropogenic land covers in arid and mosaic environments 8 .” | concepts[2] |
| concept | bottom-up — not defined in this paper's own words; systematic survey and modelling inputs — “Human pressures on the environment are the actions taken by humans with the potential to harm nature 1 , 2 . Cumulative pressure mapping measures the breadth of these pressures by coupling top-down remote sensing of land cover change with data on additional human pressures collected ‘bottom-up’ through systematic surveys and modelling 3 , 4 . The method circumvents the limitations of using remote sensing alone, which has difficulty in detecting low intensity pressures 5 , such as linear infrastructures 6 and pasture lands 7 , and often confounds natural and anthropogenic land covers in arid and mosaic environments 8 .” | concepts[3] |
| concept | Human Footprint — not defined in this paper's own words; standardized cumulative terrestrial pressure measure — “Cumulative pressure maps have been developed at regional 9 , 10 and global scales 11 , 12 . The ‘Human Footprint’ was first released in 2002 using data primarily from the early 1990s (approximately 1993) on eight human pressures globally, making it the most complete, highest resolution and globally-consistent terrestrial dataset on cumulative human pressures on the environment 13 . It has been used in a large number of ecological and conservation analyses, and still receives around 100 citations each year, particularly from its data users. However, the Human Footprint is a static and dated view of human pressures on the environment. With many of Earth’s systems experiencing pressures close to or beyond safe levels 14 , there is a strong need for an up-to-date understanding of the spatial and temporal trends in human pressures.” | concepts[4] |
| concept | Human Footprint maps — not defined in this paper's own words; spatial outputs for 1993 and 2009 — “Remotely-sensed and bottom-up survey information were compiled on eight variables measuring the direct and indirect human pressures on the environment globally in 1993 and 2009. This represents not only the most current information of its type, but also the first temporally-consistent set of Human Footprint maps. Data on human pressures were acquired or developed for: 1) built environments, 2) population density, 3) electric infrastructure, 4) crop lands, 5) pasture lands, 6) roads, 7) railways, and 8) navigable waterways. Pressures were then overlaid to create the standardized Human Footprint maps for all non-Antarctic land areas. A validation analysis using scored pressures from 3114×1 km 2 random sample plots revealed strong agreement with the Human Footprint maps. We anticipate that the Human Footprint maps will find a range of uses as proxies for human disturbance of natural systems. The updated maps should provide an increased understanding of the human pressures that drive macro-ecological patterns, as well as for tracking environmental change and informing conservation science and application.” | concepts[5] |
| concept | human disturbance — not defined in this paper's own words; downstream use proxy for pressure on natural systems — “Remotely-sensed and bottom-up survey information were compiled on eight variables measuring the direct and indirect human pressures on the environment globally in 1993 and 2009. This represents not only the most current information of its type, but also the first temporally-consistent set of Human Footprint maps. Data on human pressures were acquired or developed for: 1) built environments, 2) population density, 3) electric infrastructure, 4) crop lands, 5) pasture lands, 6) roads, 7) railways, and 8) navigable waterways. Pressures were then overlaid to create the standardized Human Footprint maps for all non-Antarctic land areas. A validation analysis using scored pressures from 3114×1 km 2 random sample plots revealed strong agreement with the Human Footprint maps. We anticipate that the Human Footprint maps will find a range of uses as proxies for human disturbance of natural systems. The updated maps should provide an increased understanding of the human pressures that drive macro-ecological patterns, as well as for tracking environmental change and informing conservation science and application.” | concepts[6] |
| concept | macro-ecological patterns — not defined in this paper's own words; large-scale ecological outcomes linked to pressure — “Remotely-sensed and bottom-up survey information were compiled on eight variables measuring the direct and indirect human pressures on the environment globally in 1993 and 2009. This represents not only the most current information of its type, but also the first temporally-consistent set of Human Footprint maps. Data on human pressures were acquired or developed for: 1) built environments, 2) population density, 3) electric infrastructure, 4) crop lands, 5) pasture lands, 6) roads, 7) railways, and 8) navigable waterways. Pressures were then overlaid to create the standardized Human Footprint maps for all non-Antarctic land areas. A validation analysis using scored pressures from 3114×1 km 2 random sample plots revealed strong agreement with the Human Footprint maps. We anticipate that the Human Footprint maps will find a range of uses as proxies for human disturbance of natural systems. The updated maps should provide an increased understanding of the human pressures that drive macro-ecological patterns, as well as for tracking environmental change and informing conservation science and application.” | concepts[7] |
| concept | remote sensing — not defined in this paper's own words; top-down observation source — “Human pressures on the environment are the actions taken by humans with the potential to harm nature 1 , 2 . Cumulative pressure mapping measures the breadth of these pressures by coupling top-down remote sensing of land cover change with data on additional human pressures collected ‘bottom-up’ through systematic surveys and modelling 3 , 4 . The method circumvents the limitations of using remote sensing alone, which has difficulty in detecting low intensity pressures 5 , such as linear infrastructures 6 and pasture lands 7 , and often confounds natural and anthropogenic land covers in arid and mosaic environments 8 .” | concepts[8] |
| concept | direct and indirect human pressures — not defined in this paper's own words; eight inputs combined by the map — “Here we use the Human Footprint methodology 13 to compile remotely-sensed and bottom-up survey information on eight variables measuring the direct and indirect human pressures on the environment in 1993 and 2009. This synthesis represents not only the most current information of its type, but also the first temporally-consistent set of Human Footprint maps, allowing for analyses of change over time. We also provide the first validation of a cumulative pressure map by adopting methods from remote sensing 15 to visually interpret human pressures in high resolution (median=0.5 m) imagery from 3114 1 km 2 random sample plots globally ( Supplementary Appendix 1 ). We then determine the level of agreement between these visually interpreted pressures and those mapped by the Human Footprint.” | concepts[9] |
| concept | visually interpret human pressures — not defined in this paper's own words; independent high-resolution plot scoring — “Here we use the Human Footprint methodology 13 to compile remotely-sensed and bottom-up survey information on eight variables measuring the direct and indirect human pressures on the environment in 1993 and 2009. This synthesis represents not only the most current information of its type, but also the first temporally-consistent set of Human Footprint maps, allowing for analyses of change over time. We also provide the first validation of a cumulative pressure map by adopting methods from remote sensing 15 to visually interpret human pressures in high resolution (median=0.5 m) imagery from 3114 1 km 2 random sample plots globally ( Supplementary Appendix 1 ). We then determine the level of agreement between these visually interpreted pressures and those mapped by the Human Footprint.” | concepts[10] |
| concept | Wilderness — not defined in this paper's own words; candidate pressure-free lands — “The Human footprint maps provide information on where humans are exerting pressure on natural systems, altering them from their natural states. They also provide information on where these pressures are absent, and ecosystems are likely to be operating in a more natural state. These pressure-free lands represent candidate sites for consideration as ‘Wilderness’ 16 , 17 . The new Human Footprint maps have already been used to show that recent economic and population growth has far outstripped increases in the Human Footprint, yet the most biologically diverse regions of Earth have been disproportionately impacted 18 . We anticipate that the 1993 and 2009 Human Footprint maps will find a range of additional uses, such as serving as proxies for human disturbance and wilderness, including understanding the role of human pressures in driving macro-ecological patterns 19 , 20 , species extinction risk and distribution analyses 21 , dispersal ecology 22 , conservation science and decision making 23 , and tracking progress toward policy commitments to conservation 23 , among others.” | concepts[11] |
| concept | standardized pressures — not defined in this paper's own words; weighted zero-to-ten input layers — “To create the Human Footprint maps we adopted the methods developed by Sanderson and colleagues 13 . Data on human pressures in 1993 and 2009 were collected or developed for: 1) the extent of built environments, 2) population density, 3) electric infrastructure, 4) crop lands, 5) pasture lands, 6) roads, 7) railways, and 8) navigable waterways, which are described in detail below ( Fig. 1 , step 1). To facilitate comparison across pressures we placed each human pressure within a 0–10 scale ( Fig. 1 , step 2), weighted within that range according to estimates of their relative levels of human pressure following Sanderson et al. 13 . The resulting standardized pressures were then summed together to create the standardized Human Footprint maps for all non-Antarctic land areas ( Fig. 1 , step 3). Pressures are not intended to be mutually exclusive, and many will co-occur in the same location. Three pressures only had data from a single time period, and these are treated as static in the Human Footprint maps.” | concepts[12] |
| concept | built environments — not defined in this paper's own words; urban and paved direct-pressure settings — “Built environments are human produced areas that provide the setting for human activity. In the context of the human footprint, we take these areas to be primarily urban settings, including buildings, paved land and urban parks. Built environments do not provide viable habitats for many species of conservation concern, nor do they provide high levels of ecosystem services 24–27 . As such, built environments were assigned a pressure score of 10.” | concepts[13] |
| concept | pressure score — not defined in this paper's own words; relative intensity assigned to a pressure layer — “Built environments are human produced areas that provide the setting for human activity. In the context of the human footprint, we take these areas to be primarily urban settings, including buildings, paved land and urban parks. Built environments do not provide viable habitats for many species of conservation concern, nor do they provide high levels of ecosystem services 24–27 . As such, built environments were assigned a pressure score of 10.” | concepts[14] |
| concept | DMSP-OLS — not defined in this paper's own words; night-light satellite sensor — “To map built environments, we used the Defence Meteorological Satellite Program Operational Line Scanner (DMSP-OLS) composite images which gives the annual average brightness of 30 arc second (~1 km at the equator) pixels in units of digital numbers (DN) 28 . These data are provided for each year from 1992 to 2012. We extracted data for the years 1994 (1993 was excluded due to anomalies in the data), and 2009, and both datasets were then inter-calibrated to facilitate comparison 29 . Using the DMSP-OLS datasets, we considered pixels to be ‘built’ if they exhibited a calibrated DN greater than 20. We selected this threshold based on a global analyses of the implications of a range of thresholds for mapped extent of cities 30 , and visual validation against Landsat imagery for 10 cities spread globally.” | concepts[15] |
| concept | digital numbers — not defined in this paper's own words; brightness units in DMSP-OLS imagery — “To map built environments, we used the Defence Meteorological Satellite Program Operational Line Scanner (DMSP-OLS) composite images which gives the annual average brightness of 30 arc second (~1 km at the equator) pixels in units of digital numbers (DN) 28 . These data are provided for each year from 1992 to 2012. We extracted data for the years 1994 (1993 was excluded due to anomalies in the data), and 2009, and both datasets were then inter-calibrated to facilitate comparison 29 . Using the DMSP-OLS datasets, we considered pixels to be ‘built’ if they exhibited a calibrated DN greater than 20. We selected this threshold based on a global analyses of the implications of a range of thresholds for mapped extent of cities 30 , and visual validation against Landsat imagery for 10 cities spread globally.” | concepts[16] |
| concept | over-glow — not defined in this paper's own words; sensor artefact adjacent to built environments — “The DMSP-OLS has limitations for the purpose of mapping human settlements, including hyper sensitivity of the sensors causing detection of over-glow adjacent to built environments 30 and bright lights associated with gas flaring from oil production facilities 29 . However, no other data exist to map built environments in a consistent way globally over our time horizon. While other datasets provide a one year snap shot of urban extent, they cannot be compared across time due to large differences in the methodologies used 31–33 , and the wildly contrasting extents in mapped built environments.” | concepts[17] |
| concept | gas flaring — not defined in this paper's own words; bright oil-production signal confounded with settlement — “The DMSP-OLS has limitations for the purpose of mapping human settlements, including hyper sensitivity of the sensors causing detection of over-glow adjacent to built environments 30 and bright lights associated with gas flaring from oil production facilities 29 . However, no other data exist to map built environments in a consistent way globally over our time horizon. While other datasets provide a one year snap shot of urban extent, they cannot be compared across time due to large differences in the methodologies used 31–33 , and the wildly contrasting extents in mapped built environments.” | concepts[18] |
| concept | human population density — not defined in this paper's own words; local population used as a pressure driver — “Human population density was mapped using the Gridded Population of the World dataset developed by the Centre for International Earth Science Information Network (CIESEN) 37 . The dataset provides a ~4 km×~4 km gridded summary of population census data for the years 1990 and 2010, which we downscaled using bilinear sampling in ArcGIS 10.1 to match the 1 km 2 resolution of the other datasets. For all locations with more than 1000 people·km −2 , we assigned a pressure score of 10 ( Table 2 ). For more sparsely populated areas with densities lower than 1000 people·km −2 , we logarithmically scaled the pressure score using,” | concepts[19] |
| concept | Gridded Population of the World — not defined in this paper's own words; census-derived population surface — “Human population density was mapped using the Gridded Population of the World dataset developed by the Centre for International Earth Science Information Network (CIESEN) 37 . The dataset provides a ~4 km×~4 km gridded summary of population census data for the years 1990 and 2010, which we downscaled using bilinear sampling in ArcGIS 10.1 to match the 1 km 2 resolution of the other datasets. For all locations with more than 1000 people·km −2 , we assigned a pressure score of 10 ( Table 2 ). For more sparsely populated areas with densities lower than 1000 people·km −2 , we logarithmically scaled the pressure score using,” | concepts[20] |
| concept | human pressures — not defined in this paper's own words; direct and indirect pressures represented by the maps — “Remotely-sensed and bottom-up survey information were compiled on eight variables measuring the direct and indirect human pressures on the environment globally in 1993 and 2009. This represents not only the most current information of its type, but also the first temporally-consistent set of Human Footprint maps. Data on human pressures were acquired or developed for: 1) built environments, 2) population density, 3) electric infrastructure, 4) crop lands, 5) pasture lands, 6) roads, 7) railways, and 8) navigable waterways. Pressures were then overlaid to create the standardized Human Footprint maps for all non-Antarctic land areas. A validation analysis using scored pressures from 3114×1 km 2 random sample plots revealed strong agreement with the Human Footprint maps. We anticipate that the Human Footprint maps will find a range of uses as proxies for human disturbance of natural systems. The updated maps should provide an increased understanding of the human pressures that drive macro-ecological patterns, as well as for tracking environmental change and informing conservation science and application.” | concepts[21] |
| concept | pressures people induce on their local natural systems — not defined in this paper's own words; local pressures assumed to rise with density — “Human population density is scored in this way under the assumption that the pressures people induce on their local natural systems increase logarithmically with increasing population density, and saturate at a level of 1000 people per km 2 .” | concepts[22] |
| concept | electric infrastructure — not defined in this paper's own words; rural and suburban pressure mapped by dim night lights — “The high sensitivity of the DMSP-OLS 28 dataset provides a means for mapping the sparser electric infrastructure typical of more rural and suburban areas. In 2009, 79% of the lights registered in the DMSP-OLS dataset had a Digital Number less than 20, and are therefore not included in our ‘built environments’ layers. However, these lower DN values are often important human infrastructures, such as rural housing or working landscapes, with associated pressures on natural environments.” | concepts[23] |
| concept | equal quintile approach — not defined in this paper's own words; night-light rescaling method — “To include these pressures, we used the inter-calibrated DMSP-OLS layers 28 used for the built environments mapping. The equations for intercalibrating across years are second order quadratics trained using data from Sicily, which was chosen as it had negligible infrastructure change over this period and where DN average roughly 14 (ref. 28 ). For our purposes, DN values of six or less where excluded from consideration prior to calibration of data, as the shape of the quadratic function leads to severe distortion of very low DN values. The inter-calibrated DN data from 1994 were then rescaled using an equal quintile approach into a 0–10 scale ( Table 2 ). To scale the data, we divided the calibrated night light data into 10 equal sample bins (each bin with a DN greater than 1 contains the same number of pixels) based on the DN values and then assigned them scores of 1 through 10, starting with the lowest DN bin. DN values of 0 were assigned a score of 0. The thresholds used to bin the 1994 data where then used to convert the 2009 data into a comparable 0–10 scale.” | concepts[24] |
| concept | crop lands — not defined in this paper's own words; intensive agricultural pressure — “Crop lands vary in their structure from intensely managed monocultures receiving high inputs of pesticides and fertilizers, to mosaic agricultures such as slash and burn methods that can support intermediate levels of natural values 38 , 39 . For the purposes of the human footprint, we focused only on intensive agriculture because of its greater direct pressure on the environment, as well as to circumvent the shortcomings of using remotely sensed data to map mosaic agriculture globally, namely the tendency to confound agriculture mosaics with natural woodland and savannah ecosystems 8 .” | concepts[25] |
| concept | intensive agriculture — not defined in this paper's own words; high-input agricultural form selected for mapping — “Crop lands vary in their structure from intensely managed monocultures receiving high inputs of pesticides and fertilizers, to mosaic agricultures such as slash and burn methods that can support intermediate levels of natural values 38 , 39 . For the purposes of the human footprint, we focused only on intensive agriculture because of its greater direct pressure on the environment, as well as to circumvent the shortcomings of using remotely sensed data to map mosaic agriculture globally, namely the tendency to confound agriculture mosaics with natural woodland and savannah ecosystems 8 .” | concepts[26] |
| concept | pasture lands — not defined in this paper's own words; extensive grazing pressure — “Pasture lands cover 22% of the Earth’s land base or almost twice that of agricultural crops 42 , making them the most extensive direct human pressure on the environment. Land grazed by domesticated herbivores is often degraded through a combination of fencing, intensive browsing, soil compaction, invasive grasses and other species, and altered fire regimes 43 . We mapped grazing lands for the year 2000 using a spatial dataset that combines agricultural census data with satellite derived land cover to map pasture extent 42 . We assigned pasture a pressure score of 4, which was then scaled from 0–4 using the percent pasture for each 1 km 2 pixel.” | concepts[27] |
| concept | direct human pressure — not defined in this paper's own words; direct environmental pressure represented by pasture — “Pasture lands cover 22% of the Earth’s land base or almost twice that of agricultural crops 42 , making them the most extensive direct human pressure on the environment. Land grazed by domesticated herbivores is often degraded through a combination of fencing, intensive browsing, soil compaction, invasive grasses and other species, and altered fire regimes 43 . We mapped grazing lands for the year 2000 using a spatial dataset that combines agricultural census data with satellite derived land cover to map pasture extent 42 . We assigned pasture a pressure score of 4, which was then scaled from 0–4 using the percent pasture for each 1 km 2 pixel.” | concepts[28] |
| concept | roads — not defined in this paper's own words; linear infrastructure and access pressure — “As one of humanity’s most prolific linear infrastructures, roads are an important direct driver of habitat conversion 44 . Beyond simply reducing the extent of suitable habitat, roads can act as population sinks for many species through traffic induced mortality 45 . Roads also fragment otherwise contiguous blocks of habitat, and create edge effects such as reduced humidity 6 and increased fire frequency that reach well beyond the roads immediate footprint 46 . Finally, roads provide conduits for humans to access nature, bringing hunters and nature users into otherwise wilderness locations 47 .” | concepts[29] |
| concept | population sinks — not defined in this paper's own words; road-associated mortality outcome — “As one of humanity’s most prolific linear infrastructures, roads are an important direct driver of habitat conversion 44 . Beyond simply reducing the extent of suitable habitat, roads can act as population sinks for many species through traffic induced mortality 45 . Roads also fragment otherwise contiguous blocks of habitat, and create edge effects such as reduced humidity 6 and increased fire frequency that reach well beyond the roads immediate footprint 46 . Finally, roads provide conduits for humans to access nature, bringing hunters and nature users into otherwise wilderness locations 47 .” | concepts[30] |
| concept | habitat conversion — not defined in this paper's own words; direct road-driven land transformation — “As one of humanity’s most prolific linear infrastructures, roads are an important direct driver of habitat conversion 44 . Beyond simply reducing the extent of suitable habitat, roads can act as population sinks for many species through traffic induced mortality 45 . Roads also fragment otherwise contiguous blocks of habitat, and create edge effects such as reduced humidity 6 and increased fire frequency that reach well beyond the roads immediate footprint 46 . Finally, roads provide conduits for humans to access nature, bringing hunters and nature users into otherwise wilderness locations 47 .” | concepts[31] |
| concept | humidity — not defined in this paper's own words; edge-condition reduced near roads — “As one of humanity’s most prolific linear infrastructures, roads are an important direct driver of habitat conversion 44 . Beyond simply reducing the extent of suitable habitat, roads can act as population sinks for many species through traffic induced mortality 45 . Roads also fragment otherwise contiguous blocks of habitat, and create edge effects such as reduced humidity 6 and increased fire frequency that reach well beyond the roads immediate footprint 46 . Finally, roads provide conduits for humans to access nature, bringing hunters and nature users into otherwise wilderness locations 47 .” | concepts[32] |
| concept | fire frequency — not defined in this paper's own words; edge-effect increased near roads — “As one of humanity’s most prolific linear infrastructures, roads are an important direct driver of habitat conversion 44 . Beyond simply reducing the extent of suitable habitat, roads can act as population sinks for many species through traffic induced mortality 45 . Roads also fragment otherwise contiguous blocks of habitat, and create edge effects such as reduced humidity 6 and increased fire frequency that reach well beyond the roads immediate footprint 46 . Finally, roads provide conduits for humans to access nature, bringing hunters and nature users into otherwise wilderness locations 47 .” | concepts[33] |
| concept | gROADS — not defined in this paper's own words; global roadway source database — “We acquired data on the distribution of roads from gROADS 48 , and excluded all trails and private roads, which were inconsistently mapped, with only a subset of countries mapping their linear infrastructure to this resolution. The dataset is the most comprehensive publicly available database on roads, which compiles nationally mapped road data spanning the period 1980–2000 and has a spatial accuracy of around 500 m. The gROADS data do not include all minor roads, and therefore should be viewed as a map of the major roadways. We mapped the direct and indirect influence of roads by assigning an pressure score of 8 for 0.5 km out for either side of roads, and access pressures were awarded a score of 4 at 0.5 km and decaying exponentially out to 15 km either side of the road ( Table 2 ).” | concepts[34] |
| concept | railways — not defined in this paper's own words; direct linear pressure without diffuse access — “While railways are an important component of our global transport system, their pressure on the environment differs in nature from that of our road networks. By modifying a linear swath of habitat, railways exert direct pressure where they are constructed, similar to roads. However, as passengers seldom disembark from trains in places other than rail stations, railways do not provide a means of accessing the natural environments along their borders. To map railways we used the same dataset as was used in the original footprint 31 , as no update of this dataset or alternate source has been developed. The direct pressure of railways where assigned a pressure score of 8 for a distance of 0.5 km on either side of the railway.” | concepts[35] |
| concept | direct pressure — not defined in this paper's own words; pressure exerted where railways are constructed — “While railways are an important component of our global transport system, their pressure on the environment differs in nature from that of our road networks. By modifying a linear swath of habitat, railways exert direct pressure where they are constructed, similar to roads. However, as passengers seldom disembark from trains in places other than rail stations, railways do not provide a means of accessing the natural environments along their borders. To map railways we used the same dataset as was used in the original footprint 31 , as no update of this dataset or alternate source has been developed. The direct pressure of railways where assigned a pressure score of 8 for a distance of 0.5 km on either side of the railway.” | concepts[36] |
| concept | navigable waterways — not defined in this paper's own words; coast, lake, and river access pressure — “Navigable rivers layers were created for the years 1994 and 2009, and combined with the navigable coasts and inland seas layers to create the final navigable waterways layers. The access pressure from navigable water bodies were awarded a score of 4 adjacent to the water body, decaying exponentially out to 15 km.” | concepts[37] |
| concept | navigable rivers — not defined in this paper's own words; deep, connected waterways near human lights — “Navigable rivers layers were created for the years 1994 and 2009, and combined with the navigable coasts and inland seas layers to create the final navigable waterways layers. The access pressure from navigable water bodies were awarded a score of 4 adjacent to the water body, decaying exponentially out to 15 km.” | concepts[38] |
| concept | Rivers — not defined in this paper's own words; waterways that act as access conduits — “Like roads, coastlines and navigable rivers act as conduits for people to access nature. While all coastlines are theoretically navigable, for the purposes of the human footprint we only considered coasts 31 as navigable for 80 km either direction of signs of a human settlement, which were mapped as a night lights signal with a DN 28 greater than 6 within 4 km of the coast. We chose 80 km as an approximation of the distance a vessel can travel and return during daylight hours. As new settlements can arise to make new sections of coast navigable, coastal layers were generated for the years 1994 and 2009.” | concepts[39] |
| concept | nature — not defined in this paper's own words; natural environments accessed by waterways — “Like roads, coastlines and navigable rivers act as conduits for people to access nature. While all coastlines are theoretically navigable, for the purposes of the human footprint we only considered coasts 31 as navigable for 80 km either direction of signs of a human settlement, which were mapped as a night lights signal with a DN 28 greater than 6 within 4 km of the coast. We chose 80 km as an approximation of the distance a vessel can travel and return during daylight hours. As new settlements can arise to make new sections of coast navigable, coastal layers were generated for the years 1994 and 2009.” | concepts[40] |
| concept | Dryad Digital Repository — not defined in this paper's own words; repository hosting the maps — “The 1 km 2 resolution, temporally-comparable Human Footprint maps [Data Citation 1] are stored in the Dryad Digital Repository, and may also be freely accessed through the Socioeconomic Data and Applications Center website ( www.worldpop.org/data/ ). From Dryad the files may be downloaded as a single 7-zip file archive (7-Zip.org) which contains individual GeoTIFF datasets, an excel file with the validation data and a PDF with the validation key. The GeoTIFFs include the Human Footprint maps for 1993 and 2009 ( Fig. 2 ), as well 14 additional GeoTIFFs of the processed data for each of the eight pressures ( Fig. 1 , step 2) from each time step ( Tables 1 , 3 (available online only)). The individual pressure layers are provided should data users wish to rework these data to create alternate maps of human pressure for their particular needs or region.” | concepts[41] |
| concept | GeoTIFF datasets — not defined in this paper's own words; distributed raster map layers — “The 1 km 2 resolution, temporally-comparable Human Footprint maps [Data Citation 1] are stored in the Dryad Digital Repository, and may also be freely accessed through the Socioeconomic Data and Applications Center website ( www.worldpop.org/data/ ). From Dryad the files may be downloaded as a single 7-zip file archive (7-Zip.org) which contains individual GeoTIFF datasets, an excel file with the validation data and a PDF with the validation key. The GeoTIFFs include the Human Footprint maps for 1993 and 2009 ( Fig. 2 ), as well 14 additional GeoTIFFs of the processed data for each of the eight pressures ( Fig. 1 , step 2) from each time step ( Tables 1 , 3 (available online only)). The individual pressure layers are provided should data users wish to rework these data to create alternate maps of human pressure for their particular needs or region.” | concepts[42] |
| concept | individual pressure layers — not defined in this paper's own words; reusable components of the composite map — “The 1 km 2 resolution, temporally-comparable Human Footprint maps [Data Citation 1] are stored in the Dryad Digital Repository, and may also be freely accessed through the Socioeconomic Data and Applications Center website ( www.worldpop.org/data/ ). From Dryad the files may be downloaded as a single 7-zip file archive (7-Zip.org) which contains individual GeoTIFF datasets, an excel file with the validation data and a PDF with the validation key. The GeoTIFFs include the Human Footprint maps for 1993 and 2009 ( Fig. 2 ), as well 14 additional GeoTIFFs of the processed data for each of the eight pressures ( Fig. 1 , step 2) from each time step ( Tables 1 , 3 (available online only)). The individual pressure layers are provided should data users wish to rework these data to create alternate maps of human pressure for their particular needs or region.” | concepts[43] |
| concept | high resolution images — not defined in this paper's own words; visual-validation imagery — “High resolution images were used to visually interpret human pressures in 3460×1 km 2 sample plots randomly located across the Earth’s non-Antarctic land areas ( Fig. 3a ). Images for these plots were obtained from World Imagery 52 , which provides one meter or better satellite and aerial imagery in many parts of the world and lower resolution satellite imagery worldwide. The map features 0.3 m resolution imagery across the continental United States and parts of Western Europe, as well as many parts of the world, with concentrations in South America, Eastern Europe, India, Japan, the Middle East and Northern Africa, Southern Africa, Australia, and New Zealand. The imagery used for the validation plots had a median resolution of 0.5 meters and a median acquisition year of 2010. Comparable imagery was not available for the 1993 time period, and therefore only the 2009 map underwent validation.” | concepts[44] |
| concept | visual score — not defined in this paper's own words; reference pressure score — “For the visual interpretation, the extent of built environments, crop lands, pasture lands, roads, human settlements, infrastructures and navigable waterways, were recorded using a standard key for identifying these features ( Supplementary Appendix 1 ). Shape, size, texture and colour of features in the imagery were important characteristics for identifying human pressures on the environment. Interpretations were also marked as ‘certain’ or ‘uncertain’, and the year and resolution of the interpreted image was recorded. The 346 ‘uncertain’ plots were discarded, leaving 3114 validation plots ( Fig. 3a ). In general, plots were classified as ‘uncertain’ for two reasons; either because cloud cover obscured the image, or because only medium resolution (15 m) imagery was available for the plot, preventing accurate interpretation of the image. The human footprint score for each plot was determined in ArcGIS, and the visual and Human Footprint scores were then normalized to a 0–1 scale. As we only retained plots for which visual interpretations of the images were determined to be ‘certain’, we consider the visual score to be the true state of in-situ pressures for the plots.” | concepts[45] |
| concept | uncertain — not defined in this paper's own words; validation label for unusable imagery — “For the visual interpretation, the extent of built environments, crop lands, pasture lands, roads, human settlements, infrastructures and navigable waterways, were recorded using a standard key for identifying these features ( Supplementary Appendix 1 ). Shape, size, texture and colour of features in the imagery were important characteristics for identifying human pressures on the environment. Interpretations were also marked as ‘certain’ or ‘uncertain’, and the year and resolution of the interpreted image was recorded. The 346 ‘uncertain’ plots were discarded, leaving 3114 validation plots ( Fig. 3a ). In general, plots were classified as ‘uncertain’ for two reasons; either because cloud cover obscured the image, or because only medium resolution (15 m) imagery was available for the plot, preventing accurate interpretation of the image. The human footprint score for each plot was determined in ArcGIS, and the visual and Human Footprint scores were then normalized to a 0–1 scale. As we only retained plots for which visual interpretations of the images were determined to be ‘certain’, we consider the visual score to be the true state of in-situ pressures for the plots.” | concepts[46] |
| concept | RMSE — not defined in this paper's own words; numerical prediction error metric — “Two statistics were used to determine Human Footprint performance, root mean squared error (RMSE) 53 and the Cohen kappa statistic of agreement 54 . The RMSE is a dimensioned (expresses average error in the units of variable of interest) error metric for numerical predictions, and tends to heavily punish large errors. The Kappa statistic expresses the agreement between two categorical datasets corrected for the expected agreement, which is based on a random allocation given the relative class sizes. When calculating the kappa statistic, the Human Footprint score was considered as a match to the visual score if they were within 20% of one another on the 0–1 scale.” | concepts[47] |
| concept | error metric — not defined in this paper's own words; class of metric instantiated by RMSE — “Two statistics were used to determine Human Footprint performance, root mean squared error (RMSE) 53 and the Cohen kappa statistic of agreement 54 . The RMSE is a dimensioned (expresses average error in the units of variable of interest) error metric for numerical predictions, and tends to heavily punish large errors. The Kappa statistic expresses the agreement between two categorical datasets corrected for the expected agreement, which is based on a random allocation given the relative class sizes. When calculating the kappa statistic, the Human Footprint score was considered as a match to the visual score if they were within 20% of one another on the 0–1 scale.” | concepts[48] |
| concept | Kappa statistic — not defined in this paper's own words; chance-corrected categorical agreement metric — “Two statistics were used to determine Human Footprint performance, root mean squared error (RMSE) 53 and the Cohen kappa statistic of agreement 54 . The RMSE is a dimensioned (expresses average error in the units of variable of interest) error metric for numerical predictions, and tends to heavily punish large errors. The Kappa statistic expresses the agreement between two categorical datasets corrected for the expected agreement, which is based on a random allocation given the relative class sizes. When calculating the kappa statistic, the Human Footprint score was considered as a match to the visual score if they were within 20% of one another on the 0–1 scale.” | concepts[49] |
| concept | Human Footprint performance — not defined in this paper's own words; agreement with independent visual scores — “Two statistics were used to determine Human Footprint performance, root mean squared error (RMSE) 53 and the Cohen kappa statistic of agreement 54 . The RMSE is a dimensioned (expresses average error in the units of variable of interest) error metric for numerical predictions, and tends to heavily punish large errors. The Kappa statistic expresses the agreement between two categorical datasets corrected for the expected agreement, which is based on a random allocation given the relative class sizes. When calculating the kappa statistic, the Human Footprint score was considered as a match to the visual score if they were within 20% of one another on the 0–1 scale.” | concepts[50] |
| concept | strong agreement — not defined in this paper's own words; validation outcome — “There is strong agreement between the Human Footprint measure of pressure and pressures scored by visual interpretation of high resolution imagery. The RMSE for the 3114 validation plots was 0.125 on the normalized 0–1 scale, indicating an average error of approximately 13%. The Kappa statistic was 0.737 ( P <0.01), also indicating good agreement between the Human Footprint and the validation dataset. Of the 3114×1 km 2 validation plots, the Human Footprint scored 94 of them 20% higher than the visual score and 263 of them 20% lower. The remaining 2757 plots (88.5%) were within 20% agreement. While this represents good agreement, it appears that the Human Footprint is to some extent susceptible to mapping pressures as absent in locations where they are actually present. The maps should therefore be considered as conservative estimates of human pressures on the environment. The Kappa statistic measure of agreement is sensitive to the threshold used to consider plots as a ‘match’. If we apply a more stringent threshold for agreement of within 15% of one another, the Kappa statistic falls to 0.565 (moderate agreement), and if we apply a less stringent threshold of within 25%, the Kappa statistic increases to 0.856 (very high agreement).” | concepts[51] |
| concept | conservative estimates — not defined in this paper's own words; interpretation caused by missing present pressures — “There is strong agreement between the Human Footprint measure of pressure and pressures scored by visual interpretation of high resolution imagery. The RMSE for the 3114 validation plots was 0.125 on the normalized 0–1 scale, indicating an average error of approximately 13%. The Kappa statistic was 0.737 ( P <0.01), also indicating good agreement between the Human Footprint and the validation dataset. Of the 3114×1 km 2 validation plots, the Human Footprint scored 94 of them 20% higher than the visual score and 263 of them 20% lower. The remaining 2757 plots (88.5%) were within 20% agreement. While this represents good agreement, it appears that the Human Footprint is to some extent susceptible to mapping pressures as absent in locations where they are actually present. The maps should therefore be considered as conservative estimates of human pressures on the environment. The Kappa statistic measure of agreement is sensitive to the threshold used to consider plots as a ‘match’. If we apply a more stringent threshold for agreement of within 15% of one another, the Kappa statistic falls to 0.565 (moderate agreement), and if we apply a less stringent threshold of within 25%, the Kappa statistic increases to 0.856 (very high agreement).” | concepts[52] |
| concept | geographic variation — not defined in this paper's own words; biome-level variation in RMSE — “While agreement is generally strong, there is some geographic variation in the RMSE results comparing the Human Footprint scores and those derived from visual interpretation ( Fig. 3b ). By calculating RMSE for all biomes that contain at least 100 of the 3114 sample plots, we found that agreement was strongest in the Tundra biome and the Temperate grasslands, savannas and shrublands biomes ( Table 4 ). Agreement was weakest in the Temperate broadleaf and mixed forest biome and the Boreal.” | concepts[53] |
| concept | priority areas — not defined in this paper's own words; protection or restoration targets — “Mapping human pressures to the environment is an essential first step to identifying priority areas for protection or restoration of natural systems. Understanding the spatial distribution of pressures, as well as their change through time, also provides insights for studies on macro-ecological patterns. The Human Footprint maps for 1993 and 2009 represent the first temporally-consistent maps of the human footprint, as well as much more up-to-date information on cumulative pressures than is currently available. Moreover, the 2009 Human Footprint map is the first cumulative pressure map to have undergone an accuracy assessment.” | concepts[54] |
| concept | alternate scoring schemes — not defined in this paper's own words; user-defined regional or species-specific map weights — “The individual pressure maps were developed to be globally consistent, using a scoring approach originally developed by Sanderson and colleagues 13 . However, in some regions and for some species groups, alternate scores may be better suited for reflecting the pressures exerted by humans on nature. We therefore provide the individual pressure layers that compose the Human Footprint maps, thereby allowing data developers to create alternate scoring schemes that better suit their purposes, as well facilitating the addition of new or alternate data sources.” | concepts[55] |
| concept | realized ‘state’ — not defined in this paper's own words; ecosystem condition not measured by Human Footprint — “Moreover, our work is subject to three primary limitations. First, like all attempts to map cumulative pressures we did not fully account for all human pressures. Some of the missing and static pressures, such as invasive species and pollution, may be closely associated with pressures we did consider, and therefore their inclusion may not affect our overall results. Second, a lack of available data resulted in three of our pressures being static through time, which would cause an underestimation of Human Footprint expansion if these pressures expanded at a higher than average rate. Third, the Human Footprint measures the pressure humans place on nature, not the realized ‘state’ or ‘impacts’ on natural systems or their biodiversity. Significant scope exists to determine how natural systems respond to cumulating human pressures, and if non-linearity or thresholds exist where pressures lead to accelerated impacts.” | concepts[56] |
| concept | ‘impacts’ on natural systems — not defined in this paper's own words; realized consequences not measured by the map — “Moreover, our work is subject to three primary limitations. First, like all attempts to map cumulative pressures we did not fully account for all human pressures. Some of the missing and static pressures, such as invasive species and pollution, may be closely associated with pressures we did consider, and therefore their inclusion may not affect our overall results. Second, a lack of available data resulted in three of our pressures being static through time, which would cause an underestimation of Human Footprint expansion if these pressures expanded at a higher than average rate. Third, the Human Footprint measures the pressure humans place on nature, not the realized ‘state’ or ‘impacts’ on natural systems or their biodiversity. Significant scope exists to determine how natural systems respond to cumulating human pressures, and if non-linearity or thresholds exist where pressures lead to accelerated impacts.” | concepts[57] |
| concept | invasive species — not defined in this paper's own words; omitted pressure — “Moreover, our work is subject to three primary limitations. First, like all attempts to map cumulative pressures we did not fully account for all human pressures. Some of the missing and static pressures, such as invasive species and pollution, may be closely associated with pressures we did consider, and therefore their inclusion may not affect our overall results. Second, a lack of available data resulted in three of our pressures being static through time, which would cause an underestimation of Human Footprint expansion if these pressures expanded at a higher than average rate. Third, the Human Footprint measures the pressure humans place on nature, not the realized ‘state’ or ‘impacts’ on natural systems or their biodiversity. Significant scope exists to determine how natural systems respond to cumulating human pressures, and if non-linearity or thresholds exist where pressures lead to accelerated impacts.” | concepts[58] |
| concept | pollution — not defined in this paper's own words; omitted pressure — “Moreover, our work is subject to three primary limitations. First, like all attempts to map cumulative pressures we did not fully account for all human pressures. Some of the missing and static pressures, such as invasive species and pollution, may be closely associated with pressures we did consider, and therefore their inclusion may not affect our overall results. Second, a lack of available data resulted in three of our pressures being static through time, which would cause an underestimation of Human Footprint expansion if these pressures expanded at a higher than average rate. Third, the Human Footprint measures the pressure humans place on nature, not the realized ‘state’ or ‘impacts’ on natural systems or their biodiversity. Significant scope exists to determine how natural systems respond to cumulating human pressures, and if non-linearity or thresholds exist where pressures lead to accelerated impacts.” | concepts[59] |
| concept | non-linearity — not defined in this paper's own words; possible pressure-response form — “Moreover, our work is subject to three primary limitations. First, like all attempts to map cumulative pressures we did not fully account for all human pressures. Some of the missing and static pressures, such as invasive species and pollution, may be closely associated with pressures we did consider, and therefore their inclusion may not affect our overall results. Second, a lack of available data resulted in three of our pressures being static through time, which would cause an underestimation of Human Footprint expansion if these pressures expanded at a higher than average rate. Third, the Human Footprint measures the pressure humans place on nature, not the realized ‘state’ or ‘impacts’ on natural systems or their biodiversity. Significant scope exists to determine how natural systems respond to cumulating human pressures, and if non-linearity or thresholds exist where pressures lead to accelerated impacts.” | concepts[60] |
| concept | thresholds — not defined in this paper's own words; possible points of accelerated impact — “Moreover, our work is subject to three primary limitations. First, like all attempts to map cumulative pressures we did not fully account for all human pressures. Some of the missing and static pressures, such as invasive species and pollution, may be closely associated with pressures we did consider, and therefore their inclusion may not affect our overall results. Second, a lack of available data resulted in three of our pressures being static through time, which would cause an underestimation of Human Footprint expansion if these pressures expanded at a higher than average rate. Third, the Human Footprint measures the pressure humans place on nature, not the realized ‘state’ or ‘impacts’ on natural systems or their biodiversity. Significant scope exists to determine how natural systems respond to cumulating human pressures, and if non-linearity or thresholds exist where pressures lead to accelerated impacts.” | concepts[61] |
| concept | static — not defined in this paper's own words; pressure layers held constant through time — “Moreover, our work is subject to three primary limitations. First, like all attempts to map cumulative pressures we did not fully account for all human pressures. Some of the missing and static pressures, such as invasive species and pollution, may be closely associated with pressures we did consider, and therefore their inclusion may not affect our overall results. Second, a lack of available data resulted in three of our pressures being static through time, which would cause an underestimation of Human Footprint expansion if these pressures expanded at a higher than average rate. Third, the Human Footprint measures the pressure humans place on nature, not the realized ‘state’ or ‘impacts’ on natural systems or their biodiversity. Significant scope exists to determine how natural systems respond to cumulating human pressures, and if non-linearity or thresholds exist where pressures lead to accelerated impacts.” | concepts[62] |
| concept | Human Footprint map — not defined in this paper's own words; name reserved for unaltered distributed data — “While we welcome the opportunity to contribute intellectually and as co-authors to research projects that incorporate our datasets into their work, we make the data freely available without restriction for non-commercial use and redistribution. The data may be altered from their original form, and redistributed if done so free of charge and with a full description of any alterations to the original data. We do however ask that term ‘Human Footprint map’ be used only when referring to the unaltered data in the Human Footprint 7-zip file, and not to alternative versions of the data created by data users, and that the data be cited following the template at the end of this manuscript.” | concepts[63] |
| concept | Sanderson and colleagues — not defined in this paper's own words; source of the adopted mapping method — “To create the Human Footprint maps we adopted the methods developed by Sanderson and colleagues 13 . Data on human pressures in 1993 and 2009 were collected or developed for: 1) the extent of built environments, 2) population density, 3) electric infrastructure, 4) crop lands, 5) pasture lands, 6) roads, 7) railways, and 8) navigable waterways, which are described in detail below ( Fig. 1 , step 1). To facilitate comparison across pressures we placed each human pressure within a 0–10 scale ( Fig. 1 , step 2), weighted within that range according to estimates of their relative levels of human pressure following Sanderson et al. 13 . The resulting standardized pressures were then summed together to create the standardized Human Footprint maps for all non-Antarctic land areas ( Fig. 1 , step 3). Pressures are not intended to be mutually exclusive, and many will co-occur in the same location. Three pressures only had data from a single time period, and these are treated as static in the Human Footprint maps.” | concepts[64] |
| concept | cumulative pressure map — not defined in this paper's own words; map class receiving first validation — “Here we use the Human Footprint methodology 13 to compile remotely-sensed and bottom-up survey information on eight variables measuring the direct and indirect human pressures on the environment in 1993 and 2009. This synthesis represents not only the most current information of its type, but also the first temporally-consistent set of Human Footprint maps, allowing for analyses of change over time. We also provide the first validation of a cumulative pressure map by adopting methods from remote sensing 15 to visually interpret human pressures in high resolution (median=0.5 m) imagery from 3114 1 km 2 random sample plots globally ( Supplementary Appendix 1 ). We then determine the level of agreement between these visually interpreted pressures and those mapped by the Human Footprint.” | concepts[65] |
Full text — tagged
Global terrestrial Human Footprint mapsterm for 1993date and 2009date Abstract Remotely-sensed and bottom-upterm survey information were compiled on eight variables measuring the direct and indirect human pressuresterm on the environment globally in 1993date and 2009date. This represents not only the most current information of its type, but also the first temporally-consistent set of Human Footprint mapsterm. Data on human pressuresterm were acquired or developed for: 1) built environmentsterm, 2) population densityterm·var, 3) electric infrastructureterm, 4) crop landsterm, 5) pasture landsterm, 6) roadsterm, 7) railwaysterm, and 8) navigable waterwaysterm. Pressures were then overlaid to create the standardized Human Footprint mapsterm for all non-Antarctic land areas. A validation analysis using scored pressures from 3114×1 km 2 random sample plots revealed strong agreementterm with the Human Footprint mapsterm. We anticipate that the Human Footprint mapsterm will find a range of uses as proxies for human disturbanceterm of natural systems. The updated maps should provide an increased understanding of the human pressuresterm that drive macro-ecological patternsterm, as well as for tracking environmental change and informing conservation science and application.
Subject terms: Conservation biology, Environmental chemistry Background & Summary Human pressures on the environmentterm are the actions taken by humans with the potential to harm natureterm 1 , 2 . Cumulative pressure mappingterm measures the breadth of these pressures by coupling top-down remote sensingterm of land cover change with data on additional human pressuresterm collected ‘bottom-upterm’ through systematic surveys and modelling 3 , 4 . The method circumvents the limitations of using remote sensingterm alone, which has difficulty in detecting low intensity pressures 5 , such as linear infrastructures 6 and pasture landsterm 7 , and often confounds natural and anthropogenic land covers in arid and mosaic environments 8 .
Cumulative pressure mapsterm have been developed at regional 9 , 10 and global scales 11 , 12 . The ‘Human Footprintterm’ was first released in 2002 using data primarily from the early 1990s (approximately 1993date) on eight human pressuresterm globally, making it the most complete, highest resolution and globally-consistent terrestrial dataset on cumulative human pressures on the environmentterm 13 . It has been used in a large number of ecological and conservation analyses, and still receives around 100 citations each year, particularly from its data users. However, the Human Footprintterm is a staticterm and dated view of human pressures on the environmentterm. With many of Earth’s systems experiencing pressures close to or beyond safe levels 14 , there is a strong need for an up-to-date understanding of the spatial and temporal trends in human pressuresterm.
Here we use the Human Footprintterm methodology 13 to compile remotely-sensed and bottom-upterm survey information on eight variables measuring the direct and indirect human pressuresterm on the environment in 1993date and 2009date. This synthesis represents not only the most current information of its type, but also the first temporally-consistent set of Human Footprint mapsterm, allowing for analyses of change over time. We also provide the first validation of a cumulative pressure mapterm by adopting methods from remote sensingterm 15 to visually interpret human pressuresterm in high resolution (median=0.5 m) imagery from 3114 1 km 2 random sample plots globally ( Supplementary Appendix 1 ). We then determine the level of agreement between these visually interpreted pressures and those mapped by the Human Footprintterm.
The Human footprint mapsterm provide information on where humans are exerting pressure on natural systems, altering them from their natural states. They also provide information on where these pressures are absent, and ecosystems are likely to be operating in a more natural state. These pressure-free lands represent candidate sites for consideration as ‘Wildernessterm’ 16 , 17 . The new Human Footprint mapsterm have already been used to show that recent economic and population growth has far outstripped increases in the Human Footprintterm, yet the most biologically diverse regions of Earth have been disproportionately impacted 18 . We anticipate that the 1993date and 2009date Human Footprint mapsterm will find a range of additional uses, such as serving as proxies for human disturbanceterm and wildernessterm, including understanding the role of human pressuresterm in driving macro-ecological patternsterm 19 , 20 , species extinction risk and distribution analyses 21 , dispersal ecology 22 , conservation science and decision making 23 , and tracking progress toward policy commitments to conservation 23 , among others.
Methods Overview of methods for mapping the Human Footprintterm To create the Human Footprint mapsterm we adopted the methods developed by Sanderson and colleaguesterm 13 . Data on human pressuresterm in 1993date and 2009date were collected or developed for: 1) the extent of built environmentsterm, 2) population densityterm·var, 3) electric infrastructureterm, 4) crop landsterm, 5) pasture landsterm, 6) roadsterm, 7) railwaysterm, and 8) navigable waterwaysterm, which are described in detail below ( Fig. 1 , step 1). To facilitate comparison across pressures we placed each human pressureterm within a 0–10 scale ( Fig. 1 , step 2), weighted within that range according to estimates of their relative levels of human pressureterm following SandersonpersName et al. 13 . The resulting standardized pressuresterm were then summed together to create the standardized Human Footprint mapsterm for all non-Antarctic land areas ( Fig. 1 , step 3). Pressures are not intended to be mutually exclusive, and many will co-occur in the same location. Three pressures only had data from a single time period, and these are treated as staticterm in the Human Footprint mapsterm.
Workflow of the Human Footprintterm approach to mapping cumulative human pressures on the environmentterm.
We used ArcGIS 10.1 to integrate spatial data on human pressuresterm. Analyses were conducted in Mollowedie equal area projection at the 1 km 2 resolution, yielding ~134.1 million pixels for Earth’s non-Antarctic terrestrial surface. For any grid cell, the Human Footprintterm can range between 0–50. The following sections and Table 1 (available online only) describe in detail the source data for each pressure, the processing steps applied, and the rationale behind the pressure weighting, and the output datasets created.
Summary of data inputs, manipulations and outputs in the Human Footprintterm workflow Built environmentsterm Built environmentsterm are human produced areas that provide the setting for human activity. In the context of the human footprintterm, we take these areas to be primarily urban settings, including buildings, paved land and urban parks. Built environmentsterm do not provide viable habitats for many species of conservation concern, nor do they provide high levels of ecosystem services 24–27 . As such, built environmentsterm were assigned a pressure scoreterm of 10.
To map built environmentsterm, we used the Defence Meteorological Satellite Program Operational Line Scanner (DMSP-OLSterm) composite images which gives the annual average brightness of 30 arc second (~1 km at the equator) pixels in units of digital numbersterm (DN) 28 . These data are provided for each year from 1992 to 2012. We extracted data for the years 1994 (1993date was excluded due to anomalies in the data), and 2009date, and both datasets were then inter-calibrated to facilitate comparison 29 . Using the DMSP-OLSterm datasets, we considered pixels to be ‘built’ if they exhibited a calibrated DN greater than 20. We selected this thresholdterm based on a global analyses of the implications of a range of thresholdsterm for mapped extent of cities 30 , and visual validation against Landsat imagery for 10 cities spread globally.
The DMSP-OLSterm has limitations for the purpose of mapping human settlements, including hyper sensitivity of the sensors causing detection of over-glowterm adjacent to built environmentsterm 30 and bright lights associated with gas flaringterm from oil production facilities 29 . However, no other data exist to map built environmentsterm in a consistent way globally over our time horizon. While other datasets provide a one year snap shot of urban extent, they cannot be compared across time due to large differences in the methodologies used 31–33 , and the wildly contrasting extents in mapped built environmentsterm.
Population densityterm·var Many of the pressures humans impose on the environment are proximate to their location, such as human disturbanceterm, hunting and the persecution of non-desired species 34 . Moreover, even low-density human populations with limited technology and infrastructure developments can have significant impacts on biodiversity, as evidenced by the widespread loss of various taxa, particularly mega fauna, following human colonization of previously unpopulated areas 35 , 36 .
Human population densityterm was mapped using the Gridded Population of the Worldterm dataset developed by the Centre for International Earth Science Information Network (CIESEN) 37 . The dataset provides a ~4 km×~4 km gridded summary of population census data for the years 1990 and 2010, which we downscaled using bilinear sampling in ArcGIS 10.1 to match the 1 km 2 resolution of the other datasets. For all locations with more than 1000 people·km −2 , we assigned a pressure scoreterm of 10 ( Table 2 ). For more sparsely populated areas with densities lower than 1000 people·km −2 , we logarithmically scaled the pressure scoreterm using, Pressure scheme used to assign weights to the eight individual pressures in the Human Footprint mapsterm.
Human population densityterm is scored in this way under the assumption that the pressures people induce on their local natural systemsterm increase logarithmically with increasing population densityterm·var, and saturate at a level of 1000 people per km 2 .
Night-time lightsterm·def The high sensitivity of the DMSP-OLSterm 28 dataset provides a means for mapping the sparser electric infrastructureterm typical of more rural and suburban areas. In 2009date, 79% of the lights registered in the DMSP-OLSterm dataset had a Digital Numberterm less than 20, and are therefore not included in our ‘built environmentsterm’ layers. However, these lower DN values are often important human infrastructures, such as rural housing or working landscapes, with associated pressures on natural environments.
To include these pressures, we used the inter-calibrated DMSP-OLSterm layers 28 used for the built environmentsterm mapping. The equations for intercalibrating across years are second order quadratics trained using data from Sicily, which was chosen as it had negligible infrastructure change over this period and where DN average roughly 14 (ref. 28 ). For our purposes, DN values of six or less where excluded from consideration prior to calibration of data, as the shape of the quadratic function leads to severe distortion of very low DN values. The inter-calibrated DN data from 1994 were then rescaled using an equal quintile approachterm into a 0–10 scale ( Table 2 ). To scale the data, we divided the calibrated night light data into 10 equal sample bins (each bin with a DN greater than 1 contains the same number of pixels) based on the DN values and then assigned them scores of 1 through 10, starting with the lowest DN bin. DN values of 0 were assigned a score of 0. The thresholdsterm used to bin the 1994 data where then used to convert the 2009date data into a comparable 0–10 scale.
Crop and pasture landsterm Crop landsterm vary in their structure from intensely managed monocultures receiving high inputs of pesticides and fertilizers, to mosaic agricultures such as slash and burn methods that can support intermediate levels of natural values 38 , 39 . For the purposes of the human footprintterm, we focused only on intensive agricultureterm because of its greater direct pressureterm on the environment, as well as to circumvent the shortcomings of using remotely sensed data to map mosaic agriculture globally, namely the tendency to confound agriculture mosaics with natural woodland and savannah ecosystems 8 .
Spatial data on remotely sensed agriculture extent in 1992 were extracted from the UMD Land Cover Classification 40 , and for 2009date from GlobCover 41 . Although intensive agricultureterm often results in whole-scale ecosystem conversion, we gave it a pressure scoreterm of 7 ( Table 2 ), which is lower than built environmentsterm because of their less impervious cover.
Pasture landsterm cover 22% of the Earth’s land base or almost twice that of agricultural crops 42 , making them the most extensive direct human pressureterm on the environment. Land grazed by domesticated herbivores is often degraded through a combination of fencing, intensive browsing, soil compaction, invasive grasses and other species, and altered fire regimes 43 . We mapped grazing lands for the year 2000 using a spatial dataset that combines agricultural census data with satellite derived land cover to map pasture extent 42 . We assigned pasture a pressure scoreterm of 4, which was then scaled from 0–4 using the percent pasture for each 1 km 2 pixel.
Roadsterm and railwaysterm As one of humanity’s most prolific linear infrastructures, roadsterm are an important direct driver of habitat conversionterm 44 . Beyond simply reducing the extent of suitable habitat, roadsterm can act as population sinksterm for many species through traffic induced mortality 45 . Roadsterm also fragment otherwise contiguous blocks of habitat, and create edge effects such as reduced humidityterm 6 and increased fire frequencyterm that reach well beyond the roadsterm immediate footprint 46 . Finally, roadsterm provide conduits for humans to access natureterm, bringing hunters and natureterm users into otherwise wildernessterm locations 47 .
We acquired data on the distribution of roadsterm from gROADSterm 48 , and excluded all trails and private roadsterm, which were inconsistently mapped, with only a subset of countries mapping their linear infrastructure to this resolution. The dataset is the most comprehensive publicly available database on roadsterm, which compiles nationally mapped roadterm data spanning the period 1980–2000 and has a spatial accuracy of around 500 m. The gROADSterm data do not include all minor roadsterm, and therefore should be viewed as a map of the major roadways. We mapped the direct and indirect influence of roadsterm by assigning an pressure scoreterm of 8 for 0.5 km out for either side of roadsterm, and access pressures were awarded a score of 4 at 0.5 km and decaying exponentially out to 15 km either side of the roadterm ( Table 2 ).
While railwaysterm are an important component of our global transport system, their pressure on the environment differs in natureterm from that of our roadterm networks. By modifying a linear swath of habitat, railwaysterm exert direct pressureterm where they are constructed, similar to roadsterm. However, as passengers seldom disembark from trains in places other than rail stations, railwaysterm do not provide a means of accessing the natural environments along their borders. To map railwaysterm we used the same dataset as was used in the original footprint 31 , as no update of this dataset or alternate source has been developed. The direct pressureterm of railwaysterm where assigned a pressure scoreterm of 8 for a distance of 0.5 km on either side of the railwayterm.
Navigable waterwaysterm Like roadsterm, coastlines and navigable riversterm act as conduits for people to access natureterm. While all coastlines are theoretically navigable, for the purposes of the human footprintterm we only considered coasts 31 as navigable for 80 km either direction of signs of a human settlement, which were mapped as a night lights signal with a DN 28 greater than 6 within 4 km of the coast. We chose 80 km as an approximation of the distance a vessel can travel and return during daylight hours. As new settlements can arise to make new sections of coast navigable, coastal layers were generated for the years 1994 and 2009date.
Large lakes can act essentially as inland seas, with their coasts frequently plied by trade and harvest vessels. Based on their size and visually identified shipping traffic and shore side settlements, we treated the great lakes of North America, Lake Nicaragua, Lake Titicaca in South America, Lakes Onega and Peipus in Russia, Lakes Balkash and Issyk Kul in Kazakhstan, and Lakes Victoria, Tanganyika and Malawi in Africa as we did navigable marine coasts.
Riversterm were considered as navigable if their depth was greater than 2 m and there were signs of nighttime lightsterm·def (DN>=6) within 4km of their banks, or if contiguous with a navigable coast or large inland lake, and then for a distance of 80 km or until stream depth is likely to prevent boat traffic ( Table 2 ). To map riversterm and their depth we used the hydrosheds (hydrological data and maps based on shuttle elevation derivatives at multiple scales) 49 dataset on stream discharge, and the following formulae from 50 , 51 : and and and Assuming second order parabola as channel shape.
Navigable riversterm layers were created for the years 1994 and 2009date, and combined with the navigable coasts and inland seas layers to create the final navigable waterwaysterm layers. The access pressure from navigable water bodies were awarded a score of 4 adjacent to the water body, decaying exponentially out to 15 km.
Data Records The 1 km 2 resolution, temporally-comparable Human Footprint mapsterm [Data Citation 1] are stored in the Dryad Digital Repositoryterm, and may also be freely accessed through the Socioeconomic Data and Applications Center website ( www.worldpop.org/data/ ). From Dryad the files may be downloaded as a single 7-zip file archive (7-Zip.org) which contains individual GeoTIFF datasetsterm, an excel file with the validation data and a PDF with the validation key. The GeoTIFFs include the Human Footprint mapsterm for 1993date and 2009date ( Fig. 2 ), as well 14 additional GeoTIFFs of the processed data for each of the eight pressures ( Fig. 1 , step 2) from each time step ( Tables 1 , 3 (available online only)). The individual pressure layersterm are provided should data users wish to rework these data to create alternate maps of human pressureterm for their particular needs or region.
The Human Footprint mapterm for 2009date, with panels showing regional overlays with the results of the validation plots.
The name, description and type of data included in the HumanFooptrint.zip file Technical Validation High resolution imagesterm were used to visually interpret human pressuresterm in 3460×1 km 2 sample plots randomly located across the Earth’s non-Antarctic land areas ( Fig. 3a ). Images for these plots were obtained from World Imagery 52 , which provides one meter or better satellite and aerial imagery in many parts of the world and lower resolution satellite imagery worldwide. The map features 0.3 m resolution imagery across the continental United States and parts of Western Europe, as well as many parts of the world, with concentrations in South America, Eastern Europe, India, Japan, the Middle East and Northern Africa, Southern Africa, Australia, and New Zealand. The imagery used for the validation plots had a median resolution of 0.5 meters and a median acquisition year of 2010. Comparable imagery was not available for the 1993date time period, and therefore only the 2009date map underwent validation.
Results from 3114×1 km 2 validation plots interpreted and scored following Supplementary Appendix 1 . ( a ) The location and visually interpreted pressure scoreterm for plots, and ( b ) the disagreement between the Human Footprintterm score and the visual validation score on a normalized 0–1 scale.
For the visual interpretation, the extent of built environmentsterm, crop landsterm, pasture landsterm, roadsterm, human settlements, infrastructures and navigable waterwaysterm, were recorded using a standard key for identifying these features ( Supplementary Appendix 1 ). Shape, size, texture and colour of features in the imagery were important characteristics for identifying human pressures on the environmentterm. Interpretations were also marked as ‘certain’ or ‘uncertainterm’, and the year and resolution of the interpreted image was recorded. The 346 ‘uncertainterm’ plots were discarded, leaving 3114 validation plots ( Fig. 3a ). In general, plots were classified as ‘uncertainterm’ for two reasons; either because cloud cover obscured the image, or because only medium resolution (15 m) imagery was available for the plot, preventing accurate interpretation of the image. The human footprintterm score for each plot was determined in ArcGIS, and the visual and Human Footprintterm scores were then normalized to a 0–1 scale. As we only retained plots for which visual interpretations of the images were determined to be ‘certain’, we consider the visual scoreterm to be the true state of in-situ pressures for the plots.
Two statistics were used to determine Human Footprint performanceterm, root mean squared error (RMSEterm) 53 and the Cohen kappa statisticterm of agreement 54 . The RMSEterm is a dimensioned (expresses average error in the units of variable of interest) error metricterm for numerical predictions, and tends to heavily punish large errors. The Kappa statisticterm expresses the agreement between two categorical datasets corrected for the expected agreement, which is based on a random allocation given the relative class sizes. When calculating the kappa statisticterm, the Human Footprintterm score was considered as a match to the visual scoreterm if they were within 20% of one another on the 0–1 scale.
There is strong agreementterm between the Human Footprintterm measure of pressure and pressures scored by visual interpretation of high resolution imagery. The RMSEterm for the 3114 validation plots was 0.125 on the normalized 0–1 scale, indicating an average error of approximately 13%. The Kappa statisticterm was 0.737 ( P <0.01), also indicating good agreement between the Human Footprintterm and the validation dataset. Of the 3114×1 km 2 validation plots, the Human Footprintterm scored 94 of them 20% higher than the visual scoreterm and 263 of them 20% lower. The remaining 2757 plots (88.5%) were within 20% agreement. While this represents good agreement, it appears that the Human Footprintterm is to some extent susceptible to mapping pressures as absent in locations where they are actually present. The maps should therefore be considered as conservative estimatesterm of human pressures on the environmentterm. The Kappa statisticterm measure of agreement is sensitive to the thresholdterm used to consider plots as a ‘match’. If we apply a more stringent thresholdterm for agreement of within 15% of one another, the Kappa statisticterm falls to 0.565 (moderate agreement), and if we apply a less stringent thresholdterm of within 25%, the Kappa statisticterm increases to 0.856 (very high agreement).
While agreement is generally strong, there is some geographic variationterm in the RMSEterm results comparing the Human Footprintterm scores and those derived from visual interpretation ( Fig. 3b ). By calculating RMSEterm for all biomes that contain at least 100 of the 3114 sample plots, we found that agreement was strongest in the Tundra biome and the Temperate grasslands, savannas and shrublands biomes ( Table 4 ). Agreement was weakest in the Temperate broadleaf and mixed forest biome and the Boreal.
Root Mean Square Errors results comparing the Human Footprintterm scores with 3114 validation plots globally, and for biomes with at least 100 plots within them.
Usage Notes Mapping human pressuresterm to the environment is an essential first step to identifying priority areasterm for protection or restoration of natural systems. Understanding the spatial distribution of pressures, as well as their change through time, also provides insights for studies on macro-ecological patternsterm. The Human Footprint mapsterm for 1993date and 2009date represent the first temporally-consistent maps of the human footprintterm, as well as much more up-to-date information on cumulative pressures than is currently available. Moreover, the 2009date Human Footprint mapterm is the first cumulative pressure mapterm to have undergone an accuracy assessment.
The individual pressure maps were developed to be globally consistent, using a scoring approach originally developed by Sanderson and colleaguesterm 13 . However, in some regions and for some species groups, alternate scores may be better suited for reflecting the pressures exerted by humans on natureterm. We therefore provide the individual pressure layersterm that compose the Human Footprint mapsterm, thereby allowing data developers to create alternate scoring schemesterm that better suit their purposes, as well facilitating the addition of new or alternate data sources.
Moreover, our work is subject to three primary limitations. First, like all attempts to map cumulative pressures we did not fully account for all human pressuresterm. Some of the missing and staticterm pressures, such as invasive speciesterm and pollutionterm, may be closely associated with pressures we did consider, and therefore their inclusion may not affect our overall results. Second, a lack of available data resulted in three of our pressures being staticterm through time, which would cause an underestimation of Human Footprintterm expansion if these pressures expanded at a higher than average rate. Third, the Human Footprintterm measures the pressure humans place on natureterm, not the realized ‘state’term or ‘impacts’ on natural systemsterm or their biodiversity. Significant scope exists to determine how natural systems respond to cumulating human pressuresterm, and if non-linearityterm or thresholdsterm exist where pressures lead to accelerated impacts.
While we welcome the opportunity to contribute intellectually and as co-authors to research projects that incorporate our datasets into their work, we make the data freely available without restriction for non-commercial use and redistribution. The data may be altered from their original form, and redistributed if done so free of charge and with a full description of any alterations to the original data. We do however ask that term ‘Human Footprint mapterm’ be used only when referring to the unaltered data in the Human Footprintterm 7-zip file, and not to alternative versions of the data created by data users, and that the data be cited following the template at the end of this manuscript.
Additional information How to cite this article: VenterpersName, O. et al. Global terrestrial Human Footprint mapsterm for 1993date and 2009date. Sci. Data 3:160067 doi: 10.1038/sdata.2016.67 (2016).
Supplementary Material Acknowledgments This work was supported by grants from the Wildlife Conservation Society, James Cook University and the Australian Research Council DP110102872 and DE140101624. We thank Neil Burgess, Jonas Geldman, Michelle VenterpersName, Todd Stevens, Dan Segan, Joe Walston, John Robinson, Kim Fisher for their thoughtful feedback on this work. We thank Christopher Small for help with the nightlights data, and Alex De Sherbinin for help with roadsterm and population data.
Footnotes Data Citations References Associated Data Data Citations Supplementary Materials