Elsevier

Ecological Economics

Volume 39, Issue 3, December 2001, Pages 333-346
Ecological Economics

ANALYSIS
Change in ecosystem service values in the San Antonio area, Texas

https://doi.org/10.1016/S0921-8009(01)00250-6Get rights and content

Abstract

San Antonio is one of the fastest growing metropolitan areas in the USA. Urban sprawl may significantly impact ecosystem services and functions but such effects are difficult to quantify and watershed-level estimates are seldom attempted. The objective of the study reported here was to determine whether LANDSAT MSS could be used to quantify changes in land-use and ecosystem services due to urban sprawl in Bexar County, TX, in which San Antonio is centered. The size of six land cover categories in the summer of 1976, 1985, and 1991 were estimated in the 141 671 ha of three watersheds in Bexar County. Coefficients published by Costanza and co-workers in 1997 [Nature 387 (1997) 253] were used to value changes in ecosystem services delivered by each land cover category, and a sensitivity analysis was conducted to determine the effect of manipulating these coefficients on the estimated values. Although we estimated that there was a 65% decrease in the area of rangeland and a 29% increase in the area of urbanized land use between 1976 and 1991, there appeared to be only a 4% net decline in the estimated annual value of ecosystem services in the study area (i.e. $5.58 ha−1 per year, with a 15-year cumulative total value of $6.24 million for the whole study area). This relatively small decline could be attributed to the neutralizing effect of the estimated 403% increase in the area of the woodlands, which were assigned the highest ecosystem value coefficient. When we assumed that the shift of rangelands to woodlands produced no net change in the value of ecosystem services per hectare, the estimated annual ecosystem service value declined by 15.4% ($23.22 ha−1 per year) between 1976 and 1991. When conducting time-series studies of ecosystem services, it is important to identify parallel changes in land cover types in order to quantify the potentially neutralizing influence of positive land cover changes on the negative effects of urban sprawl on ecosystem services.

Introduction

Between 1982 and 1997, the amount of urbanized land in the USA increased by 47% to 30.75 million ha, while the population grew by 17% (Fulton et al., 2001). During the same time period, the conversion of land for development was estimated to have increased from about 500 000 ha per year between 1982 and 1992 to 1.3 million ha per year between 1992 and 1997 (NRCS, 1999). In general, urban sprawl in the south has been exacerbated by a decline in population density in urban centers, though to a lesser extent in Texas where urban population densities have decreased less than in other southern metropolitan areas (Fulton et al., 2001).

Population growth has been especially rapid in the states along the USA–Mexico broader (USCB, 1993). In Texas, a border state, the human population is projected to increase from 19 to 33 million by 2030, with over 70% of the growth expected to occur along the central and southern portions of the I-35 highway corridor and in the Lower Rio Grande Valley (Conner and James, 1996). As a result of this growth, San Antonio has become one of the fastest growing metropolitan areas in the USA, experiencing a 25.2% increase in population from 1990 to 1998, reaching approximately 1 million in 1996, and now being the eighth largest city in the country (SAEDF, 1999). This growth can be largely attributed to a steady growth in employment in the San Antonio area during the latter half of the 1990s when several large manufacturers moved into the area in response to the North American Free Trade Agreement (Rylander, 1997).

This population growth is increasingly impacting rural areas, especially those close to major urban centers in the southern part of Texas, by accelerating land subdivision and reducing the average size of land parcels (Conner and James, 1996). In addition, increase in urban sprawl generally leads to greater traffic volumes, increased pressure on local resources, less open space (Holtzclaw, 1999), and such land-use changes often have a significant negative impact on the affected ecosystems and the goods and services that they provide. Ecosystem services represent the benefits that living organisms derive from ecosystem functions that maintain the Earth's life support system, and include nutrient cycling, carbon sequestration, air and water filtration, and flood amelioration, to name a few (Costanza et al., 1997).

While changes in land use may significantly affect ecosystem processes and services, monitoring and projecting the impacts of such land-use changes are difficult for several reasons. Monitoring changes at the regional scale (where the impact of land-use changes on ecosystems often become noticeable) is difficult because of the large volume of data and interpretation required. In addition, accurately quantifying the impacts of urban sprawl on changes in ecosystem services is difficult because of the lack of information about the contribution of alternate landscapes to these services. Finally, in order to facilitate informed decision-making by comparing the impact of anthropogenic land-use changes with the effect of ‘natural’ ecosystem changes requires more explicit measures than simple value indices.

The objectives of this study were: (1) to evaluate the efficacy of using LANDSAT multispectral scanner (MSS) data to quantify land-use change in Bexar County, TX, from 1976 to 1991; and (2) to determine if generalized coefficients can be used to evaluate changes in ecosystem services at the watershed scale.

Potentially adverse ecological impacts of urban sprawl have increasingly prompted attempts to map and characterize urban and suburban growth. The US Geological Survey is developing a geo-referenced database of urban land-use change in selected metropolitan regions by merging information from historical maps, census statistics, commerce records, remotely sensed data, and digital land-use data (Acevedo et al., 1997), but this database is incomplete. As historical satellite imagery has become more readily available and less expensive, LANDSAT imagery has become an important tool for acquiring environmental data at spatial, temporal, and spectral resolutions appropriate for assessing broad land-use changes (Verstraete et al., 1996).

While the relatively low 80×80-m spatial resolution of the LANDSAT MSS data limits the detail that can be extracted from these data, ancillary data, such as maps reflecting land-use at the time that a satellite image was taken, can facilitate classification of coarse-resolution images. If coarse-resolution data and classification levels provide sufficient explanatory power for a given purpose, their use may be advantageous because they are less data intensive and provide better broad-scale uniformity than finer resolution data and classification levels (Bourgeron et al., 1999). Moreover, because LANDSAT MSS images were initially produced as early as 1972, MSS data represent the most comprehensive data set for analyzing large-scale land-use changes during the last 25 years.

While in some instances it is desirable to use high-resolution data to conduct detailed land-use analyses, such data cannot be used to quantify long-term land-use changes. Aerial photographs have been used since the 1940s and thus predate LANDSAT MSS, but such images are generally not available for a specified area at regular intervals. In order to study temporal changes, a time series of images for the location in question must be available. Satellite-based imaging (e.g. LANDSAT MSS, LANDSAT TM, etc.) was the first technology to routinely produce images at regular intervals. Digital land-use maps (based on a wide variety of data including LANDSAT images) can also facilitate analyses of land-use patterns. However, because they are composed of data averaged over some time period, such maps do not represent time-specific data and, therefore, cannot be used for time-series analyses of land-use change. We used LANDSAT MSS data to classify land-use during a 15-year period in Bexar County because: (1) they provided readily available and affordable time-specific digital data obtained at regular intervals since the early 1970s; (2) an objective of this study was to quantify long-term changes in land-use; and (3) the resolution of the data was sufficient for classifying land-use patterns at the watershed scale.

Abramovitz (1998) pointed out that ecosystem services have extensive economic value but that they are not credited for the non-market values they provide until they become depleted. While economic tools can be used to identify trade-offs between known ecological values, it remains challenging to link technical measures of ecosystem services to attributes that can be effectively evaluated by untrained individuals (Schaberg et al., 1999). Despite this and other challenges, several attempts have been made to estimate the worth of natural resources. Most notably, Costanza et al., 1997, Costanza et al., 1998 presented a model for placing an economic value on different biomes and the services that they provided. Based on their model, they estimated that the global biospheric value of 17 identifiable ecosystem services provided by the 16 dominant global biomes is $33 trillion per year, most of which is outside the market. However, because of uncertainties, they stated that this should be considered to be a minimum estimate.

While Costanza et al.'s article did focus debate on the importance of ecosystem services that are generally undervalued in standard economic analyses, their cross-sectional estimate based on average, often local, per-unit values, was widely criticized by economists for both theoretical and empirical reasons (Pimm, 1997, Toman, 1998, Masood and Garwin, 1998, Norgaard et al., 1998, Pearce, 1998). For example, because the last hectare of an ecosystem to disappear is likely to be worth much more than the first, simple multiplication of selected average values by all the units in the biosphere underestimated a potentially infinite social value of ecosystem services. Pearce's (1998) greatest concern was that Costanza et al.'s estimated $33 trillion ‘value of everything’ is larger than the world GNP which is around $18 trillion per year. Since 1997, additional studies conducted to quantify the value of ecosystem services have produced lower estimates. For example, Alexander et al. (1998) estimated that ecosystem services are 44–88% of global GNP and concluded that while this estimate is lower than Costanza et al.'s estimate, it nevertheless indicates that accounting for ecosystem service values would greatly alter current GNP estimates. In a regional study using locally derived data, Seidl and Moraes (2000) re-estimated the ecosystem contribution of the Pantanal sub-region Nhecolandia to global production and derived a value of $15.5 billion per year, approximately 50% of Costanza et al.'s corresponding estimate.

Although Costanza et al.'s estimates of the value of ecosystem services are imperfect, and we lay no claim to their veracity, they do represent the most comprehensive set of first-approximations available for quantifying the change in the value of services provided by a wide array of ecosystems. Since one objective of our study was to determine the effectiveness of using generalized value coefficients to estimate watershed-level changes in ecosystem services, and because the scope of our project did not allow us to obtain area specific value coefficients, we used Costanza et al. (1997) estimates in our study despite their limitations.

Section snippets

Study area and estimation approach

San Antonio is centered in Bexar County (29°27′ N, 98°31′ W) near the head of the San Antonio River Basin, which traverses the Edwards Plateau, the Texas Blackland Prairies, and the Western Gulf Coastal Plain eco-regions (Fig. 1). Bexar County was chosen for the study in order to maximize the probability of detecting changes in ecosystem services due to urban sprawl. The three major streams running through the county are Salado Creek, the Upper San Antonio River, and Leon Creek. Since

Land-use change estimates

It is important to emphasize that, due to a lack of reference data, a limitation of retrospective land-use classification is uncertainty about the accuracy of the estimated size of land-use categories (Congalton and Green, 1999). Therefore, observation of changes in the size of land-use categories must be treated with caution. However, if the magnitude of the estimated changes in land use is substantial, it may still be possible to draw general inferences about the effect of perceived changes

Discussion

Remote sensing from satellites may be the only economically feasible way to regularly gather information with high spatial, spectral, and temporal resolution over large areas (Verstraete et al., 1996). This advantage will increase as the cost of obtaining such data declines and computational power to cope with larger data sets from higher resolution sensors increases. However, one limitation for conducting time series analyses of land-use changes using remotely sensed data is that satellite

Acknowledgements

We thank Ben Wu, Richard Conner and Fred Smeins, Department of Rangeland Ecology and Management, Texas A&M University, and three anonymous reviewers for their helpful comments and suggestions on earlier drafts of this manuscript.

References (36)

  • P.S Bourgeron et al.

    Impact of broad- and fine-scale patterns on regional landscape characterization using AVHRR-derived land cover data

    Ecosystem Health

    (1999)
  • T.W Boutton et al.

    Structure, function and dynamics of a subtropical savanna

    Rapid Commun. Mass Spectrom.

    (1999)
  • R Congalton et al.

    Assessing the Accuracy of Remotely Sensed Data: Principles and Practices

    (1999)
  • Conner, J.R., James, L., 1996. Environment and natural resources: trends and implications. Texas Agricultural and...
  • R Costanza et al.

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

    Nature

    (1997)
  • R Costanza et al.

    Auditing the earth: the value of the world's ecosystem services and natural capital, Costanza and his co-authors reply

    Environment

    (1998)
  • E.J Dyksterhuis

    Condition and management of rangeland based on quantitative ecology

    J. Range Manage.

    (1949)
  • ERDAS, 1997. ERDAS Field Guide. ERDAS Inc. Atlanta,...
  • Cited by (507)

    View all citing articles on Scopus

    Additional information: this paper was in part prepared by Heather Harris for a graduate course in Ecological Economics offered in the Department of Rangeland Ecology and Management, Texas A&M University, College Station, TX, USA.

    View full text