Changes in land use can cause significant changes in the ecosystem structure and process variation of ecosystem services. This study presents a detailed spatial, quantitative assessment of the variation in the value of ecosystem services based on land use change in national nature reserves of the Ningxia autonomous region in China. We used areas of land use types calculated from the remote sensing data and the adjusted value coefficients to assess the value of ecosystem services for the years 2000, 2005, and 2010, analyzing the fluctuations in the valuation of ecosystem services in response to land use change. With increases in the areas of forest land and water bodies, the value of ecosystem services increased from 182.3×107 to 223.8×107 US$ during 2000–2010. Grassland and forest land accounted for 90% of this increase. The values of all ecosystem services increased during this period, especially the value of ecosystem services for biodiversity protection and soil formation and protection. Ecological restoration in the reserves had a positive effect on the value of ecosystem services during 2000–2010.
Citation: Wang Y, Gao J, Wang J, Qiu J (2014) Value Assessment of Ecosystem Services in Nature Reserves in Ningxia, China: A Response to Ecological Restoration. PLoS ONE 9(2): e89174. https://doi.org/10.1371/journal.pone.0089174
Editor: Ben Bond-Lamberty, DOE Pacific Northwest National Laboratory, United States of America
Received: September 16, 2013; Accepted: January 15, 2014; Published: February 19, 2014
Copyright: © 2014 Wang et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: The National Special Scientific Research of Environmental Protection Public Welfare Project (No. 201209027), the National Science and Technology Support Project (No. 2012BAC01B00), and the National Ecological Environment Change Assessment by Remote Sensing Survey Project (2000–2010). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript
Competing interests: The authors have declared that no competing interests exist.
Ecosystem services (ESs) refer to the benefits people obtain from ecosystems or to aspects of ecosystems used (actively and passively) to produce human well-being , . ESs emphasize not only provisioning services (marketable goods), but also supporting (e.g. nutrient cycling), regulating (e.g. soil and water conservation), and cultural services (e.g. aesthetic values). As the Millennium Ecosystem Assessment ,  reminded us, our lives, much less our societies, economies, or well-being, depend on ESs . Ecosystem service values (ESVs) are the values of the contributions of ESs to the sustainability of human well-being . The growing body of literature on ESV includes studies on the value generated by ecosystems , , changes in ESVs in response to changes in land use , , climate change , , approaches and models for assessing ESV , , and other factors , .
Ecosystems have been substantially exploited, degraded, and destroyed in the last century as a consequence of the global increase in economic and societal prosperity . Humans have changed ecosystems more rapidly and extensively in the last 50 years than in any comparable period of human history; some 60% of the ESs studied have been degraded during this period , . The concept of ESs has become a central issue in conservation planning and the assessment of environmental impacts for preventing further abatement of the quality of ecosystems , . Many case studies on ESs have been performed, but too few have paid enough attention to long-term fluctuations of ESs, even though long-term study is necessary for detecting the response of ESs to land use change, climate change, or other factors. Monitoring the fluctuations in ESVs would benefit the management and maintenance of ecosystem sustainability, e.g. the identification and measurement of variations in ESs as a consequence of land use changes appears to be an adequate means of evaluating the environmental costs and benefits of decisions affecting the planning of land use .
The autonomous region of Ningxia is a typical region of western China, and harsh natural conditions and the over-exploitation of resources have deteriorated the already fragile ecological environment in the past few decades. The Chinese government has recently taken measures to improve the ecological environment, such as the conversion of cropland to forest, prohibition of enclosures and grazing, and sand prevention and control. Estimating the effects of these measures is thus essential for restoring ecologically fragile regions. The effects of the restoration efforts during 2000–2010 in western China, the first decade of such efforts, have generated concern. We have assessed the variation of ESVs based on land use change in the national nature reserves in Ningxia to determine the effects of ecological restoration.
The unit prices of ESVs in global biosphere ecosystems were estimated by Costanza et al. . ESVs are now commonly estimated by integrating the use of the Geographic Information System (GIS) and remote sensing (RS) , , . The estimates of Costanza et al.  have been criticized because they overestimate ESVs for wetland and underestimate ESVs for farmland , . A survey of Chinese ecologists led to an improved approach suitable for the situation in China . This approach includes merging some ES functions, as suggested by Costanza et al. , , and extracting the equivalent weighting factors for ESs per hectare for terrestrial ecosystems . Much of the research on ESVs based on the equivalent weighting factors and land use data suggest that land use type can be a proxy for ESs by matching the types to the equivalent biomes, which is convenient and simple for ESVs of large areas , , .
Based on a decade of RS data combined with equivalent weighting factors and the use of ArcGIS, a package of programs for working with GIS data, this study assessed the ESVs of national nature reserves in Ningxia for the years 2000, 2005, and 2010. Our objectives were to: (1) analyze the changes in land use in the reserves during the past 10 years, (2) assess the variation in ESVs in the reserves during this period, and (3) discuss the effects of ESVs on ecological restoration.
Materials and Methods
The study area included six national nature reserves in Ningxia, which is located in western China at 35°25′N–39°22′N and 104°49′E–107°40′E (Figure 1). The area is in a temperate arid/semi-arid region that has a continental monsoon climate, four distinct seasons, and abundant sunshine. The area is geographically diverse with annual average temperatures of −0.9 to 9.6°C and an annual precipitation of 186–800 mm. The multi-type nature reserves in the study site are classified in China as forest, wetland, and desert ecosystems and contain approximately 1000 species of plants and vertebrates in the six nature reserves. The abundant variety of natural resources provides a variety of ESs.
Data Collection and Land Use Classification
An integrated approach utilizing a Geographic Information System (GIS) and Remote Sensing (RS) was used to extract a data set for the changes in land use. The data set was extracted from GIS and RS data from Landsat Thematic Mapper imagery for 2000 and 2005 and from environmental satellite data (http://www.secmep.cn/secPortal/portal/index.faces) for 2010. To maintain consistency of classification with the remote sensing data, all the remote sensing images with good image quality in July were selected for analysis. After converting the data to the unified coordinate system and projection, we used the Krasovsky ellipsoid and the Transverse Mercator projection of ENVI 4.8 to perform RS image radiation correction and geometry correction, respectively.
After completing the pretreatments, including image mosaics, grooming, and data fusion, we used ArcGIS10.0 to consolidate and analyze the land use data with a background of raster images. The maximum likelihood classifier of the supervised classification method was used for classification with ENVI 4.8. According to the confusion matrix for classification accuracy, qualitative errors of precision for deciphering the remote sensing data for different years were controlled at the 90% level. Comparing the results of the interpretation with those of the field survey of typical points, the total classification accuracies were all higher than 90%, and the total Kappa coefficients were all greater than 0.8, which were higher than the minimum acceptable (0.7). The accuracy could thus meet the monitoring accuracy of the demands for land use change. Our data set included seven classified land use types listed in the resource and environmental database established by the Chinese Academy of Sciences (Table 1). ArcGIS 10.0 and SPSS 19.0 were used for the statistical analysis.
We obtained data sets for the normalized difference vegetation index (NDVI) from the Goddard Space Flight Center (NASA) (http://ladsweb.nascom.nasa.gov/data/search.html). We obtained climatic data sets, including the monthly records from 122 radiation stations and 756 ground-based meteorological and automatic stations, from the web of China meteorological data sharing service (Figure 2, http://cdc.cma.gov.cn/index.jsp). Data for vegetation types were obtained from GLC2000 China regional land cover data (http://solargis.info/doc/32).
Land Use Dynamics
ESVs were determined by the value coefficients and areas of land use types. The changes of areas directly caused the variation in ESVs. The areas were statistical calculations of the areas of various land use types from the remote sensing data, so the remote sensing data were the basis of the calculated ESV data and had a direct relation with the ESV calculation. Images of the national nature reserves in Ningxia from 2000, 2005, and 2010 were used to estimate the land use changes in the past decade. The Map Algebra function of ArcGIS10.0 was used to calculate the figures of land use type for 2000, 2005, and 2010 and the dynamics of land use. The rate (R) of change of land use was calculated as:(1)where A0 and A1 represent the initial and final areas of a given land use, respectively.
Costanza's  theory and the survey of 500 Chinese ecologists  suggested that the equivalent value per unit area (Table 2) was practicable in China, and it has been widely used to assess ES , , . The ESV of one equivalent weight factor was calculated as :(2)where VC0 is the economic value of one equivalent weight factor (Yuan·ha−1·yr−1)(1 Yuan = 0.16 US$), qi is the average grain price nationwide (Yuan·kg−1), pi is the yield of per unit area of crops in year i (kg·ha−1·yr−1), and i is the year.
Species resources, especially for rare species, are much more abundant inside than outside nature reserves, so adjusting the equivalent value per unit area of biodiversity protection is essential. A database of 3337 rare and endangered species in China was established from lists of national key protected wildlife species, the CITES (Convention on International Trade in Endangered Species) appendix, IUCN (the International Union for Conservation of Nature) endangered species level 3.1, a list of China's endemic species, and the IUCN Red List of Threatened Species. Based on information of species protection in 861 nature reserves in China, the distributions of 2157 rare and endangered species were determined for the nature reserves. Information about protection in national nature reserves accounted for 96.2% of the objects and 99.7% of the area, indicating that the calculated value per unit area was better representative of the important species. We then used the density of important species as a parameter for correction, calculated as:(3)where Vb is the equivalent value per unit area of biodiversity protection in nature reserves, d is the density of important species in nature reserves (species·ha−1), D is the average distribution density of important species on a national scale (species·ha−1), and Vb0 is the equivalent value per unit area of biodiversity protection in China.
The equivalent value per unit area of ES was based on the national average, so we calibrated for regional differences when assessing the ESV of a local area. Because ESV is closely related to ecosystem productivity, we used a correction factor based on ecosystem productivity to adjust the calculation of ESV. The regional correction factor reflected the difference in net primary production (NPP) caused by the variations in climate between local areas and the country as a whole. The ESV of one unit area for a year was calculated as:(4)where bi and Bi are the average NPPs of ecosystems in the study areas and the country in year I, respectively, and i is 2000, 2005, or 2010.
The CASA (Carnegie-Ames-Stanford Approach) model is based on the principle that plant productivity is correlated with the amount of photosynthetically active radiation absorbed or intercepted by green foliage , . Transformation to regional scales is easily achieved by the model, which is valuable for estimating the annual dynamic change in NPP on regional scales. RS data can provide information on many parameters of the vegetation. We obtained the parameter FPAR (see below) required for calculating NPP from time-series data for NDVI from the MODIS spectroradiometer aboard the EOS satellites. The climatic data, including total solar radiation, average temperature, and duration of sunshine, were obtained from 122 radiation stations and 756 ground-based meteorological and automatic stations in China. The equations we used were:(5)(6)(7)where APAR is photosynthetically active radiation (mJ·m−2); ε is the actual light use efficiency of vegetation (g·mJ−1); x and t refer to location and time, respectively; PAR is the total incident photosynthetically active radiation (mJ·m−2); FPAR is the fraction of PAR absorbed by the vegetation canopy; εmax is the maximum light use efficiency under ideal conditions (g·mJ−1); Tε1 and Tε2 refer to stress effects of low and high temperatures on the use efficiency of light energy, respectively; and Wε is the water stress influence coefficient, which represents the influence of moisture conditions.
The ESVs of each land use type and service function and the total ESV were then calculated as:(8)(9)(10)where ESVk, ESVf, and ESV refer to the ESVs of land use type k, service function f, and the ecosystem (Yuan·ha−1), respectively; Ak is the area of land use type k (ha); and VCikf is the value coefficient for land use type k with ES function type f (Yuan·ha−1) in year i.
The contribution rate used to assess the effect of ESV variation on land use change was calculated as :(11)where Skt is the proportion of the absolute value of ESV variation of land use type k in period t to the total amount of ESV variation of land use type k in period t.
Sensitivity Analysis of ESV
The coefficient of sensitivity (CS) validates the land use types representative of ecosystem type and certainties in the value coefficients , , . CS takes the response of ESV to the ecological value of changes in unit price as a measure of the degree of sensitivity of ESV to a coefficient. CS was calculated as:(12)where ESVi and ESVj are the initial and adjusted total ESVs, respectively, and VCik and VCjk are the initial and adjusted value coefficients, respectively. ESV is considered to be unaffected by the coefficient, and the results will be reliable when CS<1, and ESV is considered elastic relative to the coefficient when CS>1. Larger values of CS will define VCs more accurately. Regardless of how the value coefficients change, the sensitivity of ESV to changes in the value coefficients must be kept low to ensure the reliability of our results. To verify CS, a 50% adjustment in the value coefficients was made to estimate the percent changes in the calculated total ESV and the CSs.
Changes of Land Use
Table 3 and Figure 3 show the land use changes in the national nature reserves in Ningxia during 2000–2010. The area of grassland was an important factor in the reserves, accounting for approximately 60% of the total land area. The areas of wetland and unused land decreased during this period by 49.7% and 5.7%, respectively (Figure 4). The area of grassland increased during 2000–2005, decreased during 2005–2010, and had decreased by 1.1% by the end of the decade. The area of farmland decreased during 2000–2005, increased during 2005–2010, and had eventually decreased by 0.58% by the end of the decade. The areas of forest land, water bodies, and construction land continuously increased throughout the decade, by 4.3, 35.8, and 48.8%, respectively. The amount of wetland in the reserves declined more sharply than did the other land use types. The amount of construction land rose sharply and had the highest rate of increase due to the increasing encroachment of human activities.
Changes of ESV
The range of annual mean NPP was determined based on the observed NPP data of different vegetation types, including evergreen broad-leaved forest, evergreen coniferous forest, deciduous broad-leaved forest, deciduous coniferous forest, mixed forest, grassland and farmland , , . The modeled NPP values were all within the range of observed values (Figure 5 A), indicating that the modeled result was consistent with the actual NPP. The modeled NPP values between 2000 and 2010 from other studies were compared with this study through correlation analysis , . The correlation coefficient was R2 = 0.81 (Figure 5 B), showing that the modeled NPP in this study was in agreement with other studies. Through the above validation, we concluded that the NPP calculation by CASA in this study was reliable. The NPP values of terrestrial ecosystems in China calculated with the CASA model were 689, 711, and 692 gC·m−2·yr−1 in 2000, 2005, and 2010, respectively. The NPP values of terrestrial ecosystems in Ningxia were 384, 441, and 468 gC·m−2·yr−1 in the three years, respectively. The ESVs of each land use type in 2000, 2005, and 2010 are shown in Table 4 and Figure 6 (calculated with Eqs. 1–4, 8–10). The total ESV of the reserves in Ningxia increased throughout 2000–2010 by 22.7%. Total ESV increased at a higher percentage during 2000–2005 than during 2005–2010, by 12.3 and 9.2%, respectively.
The ESV of grassland, with such a large area, contributed most to the total ESV, accounting for approximately 57%. The ESV of forest land, accounting for approximately 33% of the total, continually increased during the decade. The ESVs of forest land and grassland thus constituted a substantial portion of the total ESV. The contribution rates used to assess the effect of ESV variation on land use change are shown in Figure 7. Grassland contributed most in nearly every period, indicating that changes in the area of grassland had the strongest influence on the variation in ESV. The areas and influences of forest land and farmland were also relatively large. The area of water bodies was much lower than the area of unused land, but the ESV of water bodies was nearly that of unused land, with a high value coefficient. The contribution rates of water bodies were higher than those of unused land in every period.
The ESVs of the ESs in the reserves are shown in Figure 8. The ESVs of each ES increased throughout 2000–2010. Biodiversity protection contributed most to the total ESV, accounting for more than 50% of the total value in the decade, indicating the effect of nature reserves on biodiversity protection. The value of food production was the lowest, accounting for less than 2% of the total, and demonstrated the importance of the policies to protect nature reserves from commercial exploitation. The value of biodiversity protection increased most during the period. The areas of forest land and water bodies increased sharply with high value coefficients, showing good effects on habitat supply. The ESV of soil formation and protection increased during this period, indicating that conditions of soil desertification in the reserves in Ningxia had improved. The Wilcoxon Signed Ranks Test of SPSS 19.0 showed that the ESVs of various land use types increased significantly in different periods, and the ESVs of each ecosystem service increased quite remarkably in different periods (Table 5). These increases indicated that the ESVs of the study area increased significantly during 2000–2010.
Ecosystem Sensitivity Analysis
As is shown in Table 6, the CS in each case was less than 1, indicating that the estimated ESV was not affected by changes in the value coefficients. The CSs of grassland and forest land, with large areas and high value coefficients, were higher than those of the other land use types, with values of approximately 0.6 and 0.3, respectively. The areas of farmland and unused land were relatively large, but these types had low CSs and value coefficients. We can conclude from the estimated CSs that the calculated ESVs were responsible for the uncertainties in the value coefficients.
The recent enthusiasm for analyzing the concepts and methods for ES valuation appears to have been mostly initiated by the needs of conservationists to recommend broader and better-founded policies for protecting natural resources . The responses of ecosystem quality to changes in land use and other factors are apparent in ESV variation. Economic analysis of ESs is an adequate framework for timely and effectively improving decisions involving various aspects of nature conservation and ecological restoration. Valuation may be a first step toward a “commodification” of nature and is not an end in itself but rather a conceptual and methodological framework for organizing information as a guide to making decisions and for managing ecological restoration and nature conservation.
Ecological restoration has been practiced in western China since the turn of the century. Through a wide range of comprehensive measures including policy, projects, technology, and capital, the trends of ecological degradation in the western region have been relieved. Ecological function is gradually being restored through the efforts of ecological restoration in weatern China. Variations in ESVs can reflect changes to the health of ecosystems and can thus evaluate the effects of these efforts at restoration. The evaluation of ESVs conducted in this study is a fast and effective way to assess the results of ecological restoration in western China. The increase of ESVs of the national nature reserves in Ningxia reflected the effects of ecological restoration to some extent. For example, the ESVs of forest land and water bodies increased with expanded areas and improved ecosystem quality (e.g. ecosystem productivity), and the ESVs of grassland, farmland, wetland, and unused land increased with improved ecosystem quality, even when the areas decreased. Water bodies, with a high value coefficient, had great potential for ESs. We thus recommend that more attention be paid to these land use types (e.g. water bodies) for ecological restoration and construction in the near future.
The method of deriving ESVs by multiplying the area of land use types and the value coefficients was proposed by Costanza et al. . Their value coefficients were based on the average of global ecosystems, not tallied from the Chinese situation. Xie et al.  modified the value coefficients to apply to China. The accuracy and reliability of the evaluated results are mainly determined by the accuracy of the value coefficients. More accurate value coefficients are thus necessary. ESs have considerable spatial heterogeneity, so RS and GIS technology must be used to conduct ESs assessments to improve the reliability of ESVs at the regional scale. For our study, the protection of ecosystems in nature reserves confers advantages in ecosystem productivity and biodiversity, so parameter corrections to the value coefficients are needed for accurate estimation of the ESVs in nature reserves. With NPP and biodiversity parameter corrections on value coefficients, the ESV of biodiversity protection was the highest, and the ESV of food production was the lowest, in agreement with the actual situation of nature reserves in China.
Land use can be used as a proxy measure of ESs through matching land use types with equivalent biomasses and ESVs can be easily conducted based on land use data. Using the average value coefficients, however, may not be precise enough, because the land use classification was only applied to the first-level classification (e.g. forest land and grassland). The structural and functional differences of different ecosystems at the same level, e.g. the forest land type includes broad-leaved forests, coniferous forests, bush forests, etc., may lead to uncertain ESVs. Further studies, then, must apply the detailed land use classification and value coefficients to improve the understanding of the distributed characteristics of ESs.
Different methods of evaluation can provide different results. For example, both Li et al.  and Peng et al.  evaluated the ESV of Shenzhen for the same year, giving estimates of 2.9 billion and 126.5 billion Yuan, respectively. Absolute numbers of ESVs have less meaning, and the dynamics of ESVs are commonly indicating ecological problems. Despite the residual uncertainties due to the complex, dynamic, and nonlinear nature of ecosystems , , , accurate coefficients are often less critical for time-series than for cross-sectional analyses because value coefficients tend to affect estimates of directional change less than estimates of ecosystem values at specific points in time . Our evaluation is thus valid for calculating ESVs over extended periods as a means of assessing the changes in ESVs in response to changes in land use.
We assessed ESVs and their changes for national nature reserves in Ningxia from land use data obtained during 2000–2010. The areas of forest land, water bodies, and construction land increased, while the areas of grassland, farmland, wetland, and unused land decreased. Ecological restoration helped to increase the total ESV in the reserves during 2000–2010 from 182.3 million to 223.8 million US$, an increase of 22.7%. Grassland and forest land contributed approximately 90% of the total ESV. The ESVs of all ESs increased throughout the decade, especially biodiversity protection and soil formation and protection. We can thus conclude that ecological restoration in the national nature reserves in Ningxia during 2000–2010 had achieved good results.
By matching the land use types to equivalent biomes, ESV can be estimated by the land use data and the value coefficients using GIS and RS data. The value coefficients are the essential issue for the accuracy and reliability of ESV estimation. The coefficients of sensitivity indicated that the estimated ESVs were relatively rigid relative to the changes in the value coefficients. Our analysis of the variation in the ESVs in the national nature reserves in Ningxia will be able to serve as a reference for future analyses.
We are grateful to the staff of the report of the key techniques for monitoring and conserving important biological species and for demonstrating its application in China (No. 2008BAC439B00).
Conceived and designed the experiments: YW JG JW. Performed the experiments: YW. Analyzed the data: YW JQ. Contributed reagents/materials/analysis tools: YW JG. Wrote the paper: YW.
- 1. MEA (2005) Millennium Ecosystem Assessment - Ecosystems and wellbeing: A framework for assessment. Washington, DC: Island Press.
- 2. Fisher B, Turner RK, Morling P (2009) Defining and classifying ecosystem services for decision making. Ecological Economics 68: 643–653.
- 3. MEA (2003) Millennium Ecosystem Assessment - Ecosystems and wellbeing: A framework for assessment. Washington, DC: World Resources Institute.
- 4. Summers JK, Smith LM, Case JL, Linthurst RA (2012) A review of the elements of human well-being with an emphasis on the contribution of ecosystem services. Ambio 41: 327–340.
- 5. Costanza R, Folke C (1997) Valuing ecosystem services with efficiency, fairness, and sustainability as goals. In: Daily GC, editor. Nature's Services: Societal Dependence on Natural Ecosystems. Washington D.C: Island Press. pp. 49–69.
- 6. Kubiszewski I, Costanza R, Dorji L, Thoennes P, Tshering K (2013) An initial estimate of the value of ecosystem services in Bhutan. Ecosystem Services 3: e11–e21.
- 7. Yang W, Li F, Wang R, Hu D (2011) Ecological benefits assessment and spatial modeling of urban ecosystem for controlling urban sprawl in Eastern Beijing, China. Ecological Complexity 8: 153–160.
- 8. Kreuter UP, Harris HG, Matlock MD, Lacey RE (2001) Change in ecosystem service values in the San Antonio area, Texas. Ecological Economics 39: 333–346.
- 9. Sawut M, Eziz M, Tiyip T (2013) The effects of land-use change on ecosystem service value of desert oasis: a case study in Ugan-Kuqa River Delta Oasis, China. Canadian Journal of Soil Science 93: 99–108.
- 10. Lorencová E, Frélichová J, Nelson E, Vačkář D (2013) Past and future impacts of land use and climate change on agricultural ecosystem services in the Czech Republic. Land Use Policy 33: 183–194.
- 11. Su C, Fu B (2013) Evolution of ecosystem services in the Chinese Loess Plateau under climatic and land use changes. Global and Planetary Change 101: 119–128.
- 12. Kaplowitz MD (2000) Identifying ecosystem services using multiple methods: Lessons from the mangrove wetlands of Yucatan, Mexico. Agriculture and Human Values 17: 169–179.
- 13. Koschke L, Fürst C, Frank S, Makeschin F (2012) A multi-criteria approach for an integrated land-cover-based assessment of ecosystem services provision to support landscape planning. Ecological Indicators 21: 54–66.
- 14. Su CH, Fu BJ, He CS, Lü YH (2012) Variation of ecosystem services and human activities: A case study in the Yanhe Watershed of China. Acta Oecologica 44: 46–57.
- 15. Wang X, Chen W, Zhang L, Jin D, Lu C (2010) Estimating the ecosystem service losses from proposed land reclamation projects: A case study in Xiamen. Ecological Economics 69: 2549–2556.
- 16. Daily GC (1997) Nature's Services: Societal Dependence on Natural Ecosystems. Washington, DC: Island Press.
- 17. Burkhard B, Petrosillo I, Costanza R (2010) Ecosystem services - Bridging ecology, economy and social sciences. Ecological Complexity 7: 257–259.
- 18. Fisher B, Kerry Turner R (2008) Ecosystem services: Classification for valuation. Biological Conservation 141: 1167–1169.
- 19. Barral MP, Oscar MN (2012) Land-use planning based on ecosystem service assessment: A case study in the Southeast Pampas of Argentina. Agriculture, Ecosystems & Environment 154: 34–43.
- 20. Costanza R, d'Arge R, Groot Rd, Farber S, Grasso M, et al. (1997) The value of the world's ecosystem services and natural capital. Nature 387: 253–260.
- 21. Troy A, Wilson MA (2006) Mapping ecosystem services: Practical challenges and opportunities in linking GIS and value transfer. Ecological Economics 60: 435–449.
- 22. Wang ZM, Zhang B, Zhang SQ (2004) Study on the effects of land use change on ecosystem service values of Jilin province. Journal of Natural Resources 19: 55–61.
- 23. Zhao J, Wei L, Chen S (2010) Dynamics of the ecosystem service values along the upper reaches of Shiyanghe River Basin. Journal of Arid Resources and Environment 1: 36–40.
- 24. Heal G (2000) Valuing ecosystem services. Ecosystems 3: 24–30.
- 25. Wilson MA, Howarth RB (2002) Discourse-based valuation of ecosystem services: establishing fair outcomes through group deliberation. Ecological Economics 41: 441–443.
- 26. Xie GD, Zhen L, Lu CX, Xiao Y, Chen C (2008) Expert knowledge based valuation method of ecosystem services in China. Journal of Natural Resources 23: 911–919.
- 27. Costanza R, Cumberland J, Daly H, Goodland R, Norgaard R (1997) An introduction to ecological economics. Delray Beach Fla USA: St. Lucie Press.
- 28. Liu Y, Li J, Zhang H (2012) An ecosystem service valuation of land use change in Taiyuan City, China. Ecological Modelling 225: 127–132.
- 29. Li H, Wang S, Ji G, Zhang L (2011) Changes in land use and ecosystem service values in Jinan, China. Energy Procedia 5: 1109–1115.
- 30. Li J, Wang W, Hu G, Wei Z (2010) Changes in ecosystem service values in Zoige Plateau, China. Agriculture, Ecosystems & Environment 139: 766–770.
- 31. Song G, Fu C, E Y (2011) The Analysis of Ecosystem Service Value's Change in Yueqing Bay Wetland Based on RS and GIS. Procedia Environmental Sciences 11: 1365–1370.
- 32. Li T, Li W, Qian Z (2010) Variations in ecosystem service value in response to land use changes in Shenzhen. Ecological Economics 69: 1427–1435.
- 33. Wu KY, Ye XY, Qi ZF, Zhang H (2013) Impacts of land use/land cover change and socioeconomic development on regional ecosystem services: The case of fast-growing Hangzhou metropolitan area, China. Cities 31: 276–284.
- 34. Xie GD, Lu CX, Leng YF, Zheng D (2003) Ecological assets valuation of the Tibetan Plateau. Journal of Natural Resources 18: 189–196.
- 35. Monteith JL, Moss CJ (1977) Climate and the Efficiency of Crop Production in Britain [and Discussion]. Philosophical Transactions of the Royal Society of London B, Biological Sciences 281: 277–294.
- 36. Potter CS, Randerson JT, Field CB, Matson PA, Vitousek PM, et al. (1993) Terrestrial ecosystem production: A process model based on global satellite and surface data. Global Biogeochemical Cycles 7: 811–841.
- 37. Li RQ, Dong M, Cui JY, Zhang L, Cui QG, et al. (2007) Quantification of the Impact of Land-Use Changes on Ecosystem Services: A Case Study in Pingbian County, China. Environmental Monitoring & Assessment 128: 503–510.
- 38. Wang ZM, Zhang SQ, Zhang B (2004) Effects of land use change on value of ecosystem services of Sanjiang Plain. China Environmental Science 24: 125–128.
- 39. Gao ZQ, Liu JY (2008) A comparative study of Chinese vegetation net productivity. Chinese Science Bulletin 53: 317–326.
- 40. Chen YM, Zhang WQ, Yang TX, Zhao G, Wang SB (2012) The change characteristics of net primary production in different vegetation types in China. Journal of Fudan University 51: 377–381.
- 41. Zhang FM, Ju WM, Chen JM, Wang SQ, Yu GR, et al. (2012) Characteristics of terrestrial ecosystem primary productivity in East Asia based on remote sensing and process-based model. Chinese Journal of Applied Ecology 23: 307–318.
- 42. Gao YN, Yu GR, Zhang L, Liu M, Huang M (2012) The changes of net primary productivity in Chinese terrestrial ecosystem: based on process and parameter models. Progress in Geography 31: 109–117.
- 43. Balmford A, Bruner A, Cooper P, Costanza R, Farber S, et al. (2002) Economic reasons for conserving wild nature. Science 297: 950–953.
- 44. Peng J, Wang YL, Chen YF, Li WF, Jiang YY (2005) Economic value of urban ecosystem services: a case study in Shenzhen. Acta Scientiarum Naturalium Universitatis Pekinensis 41: 594–604.
- 45. Hein L, van Koppen K, de Groot RS, van Ierland EC (2006) Spatial scales, stakeholders and the valuation of ecosystem services. Ecological Economics 57: 209–228.
- 46. Limburg KE, O'Neill RV, Costanza R, Farber S (2002) Complex systems and valuation. Ecological Economics 41: 409–420.
- 47. Turner RK, Paavola J, Cooper P, Farber S, Jessamy V, et al. (2003) Valuing nature: lessons learned and future research directions. Ecological Economics 46: 493–510.