An analysis of direct human impacts across Earth's land surface using global satellite images and ground surveys reveals the scale of the 'human footprint' on the world and its changes between 1993 and 2009.
Humanity is causing unprecedented changes to Earth, such that we may be entering a human-dominated geological era termed the Anthropocene1,2 and transgressing the environmental boundaries within which we can live safely3,4. The impact of the growing extent and intensity of human influences on our landscapes is reflected in changes, usually of loss and degradation, in natural habitats and in the species that they contain. We need to understand not only where human pressures occur, but also where they are greatest and how they change over time. Taking advantage of the availability of global data sets on a range of human pressures across 16 years, papers by Venter et al. in Nature Communications5 and Scientific Data6 provide the first analysis of this changing 'human footprint' on the world's terrestrial landscape.
Humans exert pressures on the planet in a great many ways that may lead directly or indirectly to changes in natural systems (Fig. 1). The first step in documenting where human pressures act across the world was taken in 2002 by Sanderson et al.7 with the development of a framework to map the human footprint using eight global data sets of human activities. Constructing such a map presents profound challenges because of the complexity of our impacts on the planet. Therefore, some decisions had to be made about what to include when developing their map of the world's human footprint, and Venter and colleagues have followed the approach taken by Sanderson et al.7.
Venter et al. confined their analysis to the terrestrial landscape because assessing the footprint in the marine environment would require a different approach and data sets. They concentrated on direct, rather than indirect, measures of human influence for which data were available. Only accessible data sets were used that had global coverage and that were easily available and of sufficient quality. Antarctica and many oceanic islands that were absent from these global data sets were excluded. These decisions sought to match current data availability with the ambition of developing a global framework for assessing human impacts on the terrestrial environment.
The authors have built substantially on the work of Sanderson and colleagues and have brought it up to date by analysing the most recently available comprehensive data sets, and by adding an assessment of human footprint changes over time. Furthermore, Venter and colleagues provide a service for the future by clearly describing all data sets6 and how they were used. This allows easy access to the data and methods so that the approach can be developed, and it will enable changes in patterns of human influence to be assessed in the future using data available at the time.
The heart of Venter and colleagues' work lies in combining data sets on several pressures to produce an assessment of how human influences accumulate, an approach that the authors say is more indicative of the totality of direct human pressures than is producing maps of single pressures, some of which are easier to detect than others. The result is a 'cumulative threat map'5 and a human footprint that represents an accumulation of a range of pressures. Venter et al. identified eight data sets representing human population density, land transformation, human transit routes and electrical power infrastructure to serve as proxies for this footprint. Some data were remotely sensed and others collected through ground surveys.
Three data sets (for pasturelands, roads and railways) were not available for the dates needed by the authors to make comparisons over time and were therefore not used when assessing change. This exemplifies the challenge of addressing the fundamental question of how human influence on the planet's terrestrial landscapes has changed. If a study was commissioned to address this from scratch, it would not get off the ground because of the scale of data collection required. Together, the two studies by Venter and colleagues represent a pragmatic approach to these challenges.
The headline findings are that direct impacts of human development can be measured in 75% of the world's terrestrial systems, and that the human footprint increased by 9% between 1993 and 2009, during which time the human population increased by 23% and the global economy increased by 153%. The comparison of footprints reveals intriguing insights, all of which merit further analysis and, potentially, should be brought to the attention of those who make policies and decisions, including governments. For example, 'pressure-free lands' are now restricted to high northern latitudes, some deserts and the most distant parts of the Amazon and Congo rainforests. The change in the footprint over that period varies with geography and habitat. Areas such as the North American tundra, most New Guinea forests and some forests in the Neotropics (the tropical part of the American continent) showed the biggest increases in human impact.
Although these findings provide food for thought, they will also trigger questions regarding caveats and qualifiers about the data that are available and how adequately they reflect human pressures on terrestrial ecosystems. This is inevitable when tackling an issue that is so complex in lots of different ways. Rather than diminishing such work, this should spur us on to improve both its conceptual basis and its technical execution, so that an even better map of human influence across the world's land masses can be developed in the future.
Earth is being changed substantially, and we need ways to both understand and communicate how human pressures on the planet combine. Venter et al. have created a framework that will allow researchers to track a range of direct pressures and, crucially, provide information that could be relevant to those who make high-level policy decisions.
However, we do need to add to this framework. For example, ecologists have no single metric yet for measuring the influence of hunting across terrestrial systems, and, given the huge pressures from over-exploitation of species8 and escalating pressures from the illegal trade in wildlife9, this would be a key step forward. It would be fascinating, and probably alarming, to see how such a metric might change the human footprint map.
About this article