Elsevier

Ecological Economics

Volume 41, Issue 3, June 2002, Pages 509-527
Ecological Economics

SPECIAL ISSUE: The Dynamics and Value of Ecosystem Services: Integrating Economic and Ecological Perspectives
Global estimates of market and non-market values derived from nighttime satellite imagery, land cover, and ecosystem service valuation

https://doi.org/10.1016/S0921-8009(02)00097-6Get rights and content

Abstract

We estimated global marketed and non-marketed economic value from two classified satellite images with global coverage at 1 km2 resolution. GDP (a measure of marketed economic output) is correlated with the amount of light energy (LE) emitted by that nation as measured by nighttime satellite images. LE emitted is more spatially explicit than whole country GDP, may (for some nations or regions) be a more accurate indicator of economic activity than GDP itself, can be directly observed, and can be easily updated on an annual basis. As far as we know, this is the first global map of estimated economic activity produced at this high spatial resolution (1 km2). Ecosystem services product (ESP) is an important type of non-marketed value. ESP at 1 km2 resolution was estimated using the IGBP land-cover dataset and unit ecosystem service values estimated by Costanza et al. [Valuing Ecosystem Services with Efficiency, Fairness and Sustainability as Goals. Nature's Services, Island Press, Washington DC, pp. 49–70]. The sum of these two (GDP+ESP)=SEP is a measure of the subtotal ecological–economic product (marketed plus a significant portion of the non-marketed). The ratio: (ESP/SEP)×100=%ESP is a measure of proportion of the SEP from ecosystem services. Both SEP and %ESP were calculated and mapped for each 1 km2 pixel on the earth's surface, and aggregated by country. Results show the detailed spatial patterns of GDP, ESP, and SEP (also available at: http://www.du.edu/∼psutton/esiindexisee/EcolEconESI.htm). Globally, while GDP is concentrated in the northern industrialized countries, ESP is concentrated in tropical regions and in wetlands and other coastal systems. %ESP ranges from 1% for Belgium and Luxembourg to 3% for the Netherlands, 18% for India, 22% for the United States, 49% for Costa Rica, 57% for Chile, 73% for Brazil, and 92% for Russia. While GDP per capita has the usual northern industrialized countries at the top of the list, SEP per capita shows a quite different picture, with a mixture of countries with either high GDP/capita, high ESP/capita, or a combination near the top of the list. Finally, we compare our results with two other indices: (1) The 2001 Environmental Sustainability Index (ESI) derived as an initiative of the Global Leaders of Tomorrow Environment Task Force, World Economic Forum, and (2) Ecological Footprints of Nations: How much Nature do they use? How much Nature do they have? developed by Mathis Wackernagel and others. While both of these indices purport to measure sustainability, the ESI is actually mainly a measure of economic activity (and is correlated with GDP), while the Eco-Footprint index is a measure of environmental impact. The related eco-deficit (national ecological capacity minus national footprint) correlates well with %ESP.

Introduction

Economic activity is fundamentally a spatial phenomenon. Both traditional marketed economic activities (like manufacturing, sales, and final consumption) and ‘non-marketed’ ecosystem services occur at specific spatial locations and are associated with specific natural, agricultural, or urban ecosystems. A necessary step toward better understanding these activities and services is to map their spatial patterns. That is what we have tried to do in this paper, at both the global and national level.

Various measures of ‘economic activity’ and environmental quality are also important as ‘indicators’ for policy decisions. Key questions here revolve around exactly what the indicators measure. Gross Domestic Product (GDP) is the most popular indicator of economic performance. But GDP measures only marketed economic activity or gross income (Costanza et al., 2001). It was never intended as a measure of economic welfare, and it functions very poorly as a welfare measure. Yet it is inappropriately used as a national welfare measure in far too many circumstances.

What are the problems with GDP as a welfare measure? First, lumping all activity or income together does not separate desirable, welfare enhancing activity from undesirable welfare reducing activity. For example, an oil spill increases GDP because someone has to clean it up, but it obviously detracts from welfare. From the perspective of GDP, more crime, more sickness, more war, more pollution, more fires, storms, and pestilence are all good things, since they can increase marketed activity in the economy. Second, GDP leaves out many things that currently do enhance welfare but are outside the market. The unpaid work of mothers caring for their own children at home doesn't show up in GDP, but if they decide to work outside the home to pay for child care, GDP suddenly increases. The non-marketed services of nature in providing clean air and water, food and natural resources do not show up in GDP, but if those services are damaged and we have to pay to fix or replace them, then GDP suddenly increases. Third, GDP takes no account of the distribution of income among individuals. But it is well known that an additional $1 worth of income produces more welfare if one is poor rather than rich (Daly and Cobb, 1989).

In this paper we look at the spatial patterns of conventional GDP and also at the value of non-marketed ecosystem services that are not currently included in GDP (de Groot et al., 2002, Costanza et al., 1997a, Costanza et al., 1997b). We do not address the other important shortcomings of GDP (including distribution, unpaid domestic labor, pollution, etc.)—leaving those for future work. In this paper we focus on the subtotal of economic value represented by the sum of conventional marketed economic goods and services (as measured by GDP) and non-marketed ecosystem goods and services. A list of these ecosystem services and their approximate dollar values for a range of ecosystems are given in Costanza et al., 1997a, Costanza et al., 1997b. A more detailed description of these services and their links to ecosystem functions is given in de Groot et al. (2002).

Other indicators at the global and national level are also proliferating. In particular, there are many new proposed indicators of ‘sustainability’. We contend that none of these proposed indicators of sustainability actually measure sustainability. They are generally indicators of economic and/or environmental quality or stress in the present, and the link is tenuous at best as to whether or not these indices say anything at all about the sustainability of these patterns over time.

For example, the ‘Ecological Footprint’ (EF) and the ‘2001 Environmental Sustainability Index’ (ESI) are two distinct, independent attempts specifically aimed at assessing sustainability (Wackernagel et al., 1997, Samuel-Johnson and Esty, 2001). The EF is a composite index involving many variables, which focus on the nature and productivity of land resources, variability of human consumption patterns, and the energy accounting of each nation's international trade. The land variables focus on areal extent, biological productivity, and waste absorption capacity. The consumption variables characterize and account for the differing ecological impact of human consumption throughout the nations of the world. Finally, the ecological footprint index tries to capture the separation of production and consumption by looking at the import and export goods of each nation to see who is actually consuming the energy associated with manufacturing, agriculture, etc. (Chisolm, 1990). Wackernagel's index calculated the following measures for 52 nations of the world: Total Ecological Footprint (a measure of impact), Available Ecological Capacity, and Ecological Deficit (ED, the difference between the two) (Wackernagel and Rees, 1996).

The EF is clearly a measure of environmental impact, not sustainability. The ED is usually interpreted as a measure of sustainability—the higher the ED the lower the sustainability. But while the ED clearly shows whether ecosystem services are being imported across country boundaries, it says nothing (necessarily) about the sustainability of that pattern of imports, or about the capacity of ecosystems to sustain these flows (Folke et al., 1996).

The 2001 ESI was developed as an initiative of the Global Leaders of Tomorrow Environment Task Force of the World Economic Forum. The Yale Center for Environmental Law and Policy (YCELP) and the Center for International Earth Science Information Network (CIESIN) contributed to the development of this index. The ESI attempts to develop a ‘transparent, interactive process that draws on rigorous statistical, environmental, and analytic expertise to quantify environmental sustainability’. According to the main report of the ESI document, the key results are: (1) Environmental Sustainability can be measured. (2) The Index creates benchmarks of environmental conditions that can influence decision-making. (3) Serious ‘data gaps’ for many nations of the world should be filled. (4) Economic conditions affect, but do not determine, environmental conditions; and, policy regarding these conditions are separate choices. The ESI is derived by averaging five key ‘core’ components (parenthetical key: Component [# of Indicators]: Environmental Systems [5], Reducing Stresses [5], Reducing Human Vulnerability [2], Social and Institutional Capacity [7], and Global Stewardship [3]). The ‘Indicator’ variables that constitute the five key components are themselves derived from 67 specifically measurable and nationally aggregate variables. Examples of a few of the 67 fundamental variables are: ‘urban SO2 concentration’, ‘total fertility rate’, ‘scientific and technical articles per million of population’, and ‘number of memberships in environmental inter-governmental organizations’. One of the variables used in the ESI that measured anthropogenic impact on the land was in fact derived from a composite nighttime satellite image and a similar global land cover dataset (Elvidge et al., 1995). How these 67 variables should be weighted is a complex question, which the ESI report reasonably avoids. The ‘weighting’ question is undoubtedly an element of the ‘interactive’ nature of developing indices for which there will be a consensus. One notable result of the arithmetic involved in generating the ESI is the fact that the aggregate national numbers correlate strongly with nationally aggregate measures of GDP/Capita (more on this later).

The point is that there is a strong need for defensible, measurable indicators of economic performance and environmental quality and these indicators do not yet exist. We propose to remedy this deficit in this paper. We also compare our results with the two existing indicators mentioned above (ED and ESI), noting that neither of them is actually measuring sustainability. Sustainability is much more difficult to assess and requires (at minimum) a dynamic modeling approach to even begin to frame the appropriate questions (c.f. Boumans et al., 2002). It also requires a shift in thinking from a view of nature as stable and in equilibrium to one of complex and adaptive social–ecological systems (Costanza et al., 1993).

Section snippets

Data and methods

The indices described in this paper use two datasets to measure the ‘Market’ and ‘Non-Market’ economy of nations. The proxy measure of ‘Market’ economy is the amount of LE emitted from each nation as measured by a nighttime satellite image of the world. Nighttime imagery has been shown to be a strong proxy measure of GDP in earlier studies (Elvidge et al., 1997). The proxy measure of ‘Non-Market’ economy is a measure of the total value of the ecosystem services of the lands and waters of each

Results

Fig. 2, Fig. 3, Fig. 4, Fig. 5, Fig. 6, Fig. 7, Fig. 8 represent a new and more comprehensive picture of marketed and non-marketed economic value at a global scale. The 1 km2 resolution of the basic data sets and Fig. 2, Fig. 3, Fig. 4 allow for an unprecedented view of how these values are spatially distributed around the planet.

Fig. 2 is the highest spatial resolution estimate of marketed economic activity (GDP) we know of to date. The concentration of GDP in the US, the EU, and Japan is

Discussion

The data and maps we have assembled allow for some interesting new pictures of the world to be constructed and for some interesting new questions to be posed. One can ask, for example, which countries have the highest percentage of support of human welfare arising from ecosystem services relative to marketed goods and services? Fig. 8 (%ESP) shows an interesting pattern as an answer to that question, with countries as disparate as Canada, Russia, Nicaragua, and Botswana ranking high, and most

Conclusion

This paper presents a spatially explicit map (1 km2 resolution) of marketed economic activity derived from nighttime satellite imagery and nationally aggregate measures of GDP (Fig. 2), non-marketed economic activity derived from ecosystem service valuation and a global landcover dataset (ESP, Fig. 2), in addition to several nationally aggregated measures and other manipulations of these maps (Fig. 5, Fig. 6, Fig. 7). One particular measure derived from these datasets was percent of economy

Acknowledgements

This work was conducted as part of the Working Group on the Value of the World's Ecosystem Services and Natural Capital; Toward a Dynamic, Integrated Approach supported by the National Center for Ecological Analysis and Synthesis, a Center funded by NSF (Grant #DEB-0072909), the University of California, and the Santa Barbara campus. Additional support was also provided for the Postdoctoral Associate, Matthew A. Wilson, in the Group. We thank Matt Wilson and anonymous peer reviewers for

References (24)

  • Belward, A.S., 1996. The IGBP-DIS global 1 km land cover data set (DISCover)—proposal and implementation plans,...
  • Boumans, R., Costanza, R., Farley, J., Wilson, M.A., Rotmans, J., Villa, F., Portela, R., Grasso, M., 2002. Modeling...
  • M. Chisolm

    The increasing separation of production and consumption

  • R. Costanza et al.

    Modeling complex ecological economic systems: toward an evolutionary dynamic understanding of people and nature

    BioScience

    (1993)
  • Costanza, R., d'Arge, R., deGroot, R., Farber, S., Grasso, M., Hannon, B., Naeem, S., Limburg, K., Paruelo, J.,...
  • R. Costanza et al.

    The value of the world's ecosystem services and natural capital

    Nature

    (1997)
  • R. Costanza et al.

    The value of the world's ecosystem services: putting the issues in perspective

    Ecological Economics

    (1998)
  • R. Costanza et al.

    Green national accounting: goals and methods

  • H.E. Daly et al.
  • de Groot, R., Wilson, M., Boumans, R., 2002. A Typology for the description, classification, and valuation of ecosystem...
  • C.N.H. Doll et al.

    Night-time imagery as a tool for global mapping of socio-economic parameters and greenhouse gas emissions

    Ambio

    (2000)
  • C.D. Elvidge et al.

    Mapping city lights with nighttime data from the DMSP OLS

    Photogrammetric Engineering and Remote Sensing

    (1995)
  • Cited by (0)

    1

    Tel.: +1-410-326-7263.

    2

    As of 09/01/2002, this author can be reached at the Gund Institute for Ecological Economics, University of Vermont, School of Natural Resources, George D. Aiken Center, Burlington VT 05405–0088, USA.

    View full text