Estimates of loss in ecosystem service values of Songnen plain from 1980 to 2000

Transcription

1 Journal of Geographical Sciences 15, 1 (2005) ISSN: X Estimates of loss in ecosystem service values of Songnen plain from 1980 to 2000 WANG Zongming 12, ZHANG Bai t, ZHANG Shuqing j, SONG Kaishan L2, DUA N Hongtao ~ (1. Northeast Institute of Geography and Agricultural Ecology, CAS, Changchun , China; 2. Graduate School of Chinese Academy of Sciences, Beijing , China) Abstract: Ecosystem services are premises for human beings to exist and develop in the world, and it is of importance to study effects of land use change on ecosystem functions. Songnen plain, located in Northeast China, is one of the national key bases for agricultural production. Because there were remarkable variations in land use in recent years, it can supply specific foundation for protection and restoration of regional eco-environment to explore changes in its values of ecosystem services. This study aims to qualify long-term changes in land use of Songnen plain from 1980 to 2000, and to explore the impact of land use change on ecosystem services. With 2-period land use data using RS and GIS, this paper analyzed changes in land use and in ecosystem service values using the ecosystem service value coefficients (VC) put forward by Costanza et 01. Results showed that from 1980 to 2000, area of rangeland, water area and wetlands and area of cropland decreased, but urban area and unused land increased. Due to cropland's comparatively low VC and comparatively high VC of wetlands and water area, increment in cultivated land could not offset the total decrement in regional ecosystem service values because of the decrease in area of wetlands and water area. Meanwhile, city sprawl has also led to loss of ecological values. Total ecosystem service values of the study area reduced dramatically, ~om 34, to 31, @$ in the period This represents a 9.11% net decline in annual value of ecosystem services in the study area. Key words: ecological values; land use change; ecosystem service; Songnen plain doi: t0.1360/gs Introduction Ecosystems around the globe create and maintain an environment suitable for the continuation of human life. Ecosystems supply goods such as timber, pharmaceuticals, and seafood, and so provide services including purification of air and water, stabilization of climate, and generation and renewal of soil and soil fertility (Daily, 1997). Ecosystem services have value to society but no relevant market where these values are expressed (Costanza et al., 1997). Changes in land-use and land cover have important consequences for natural resources, significantly affect key aspects of Earth System functioning. Land-use and land-cover changes are so pervasive that, when aggregated globally, they significantly affect key aspects of Earth System functioning (Turner, 2001). They directly impact biotic diversity worldwide (Sala et al., 2000); contribute to local and regional climate change (Chase et al., 1999; Zhong et al., 2003) as well as to global climate warming (Houghton et ol., 1999); are the primary source of soil degradation (Tolba eta/., 1992); and, by altering ecosystem services, affect the ability of biological systems to support human needs (Vitousek et al., 1997). Dramatic land-use changes often have a significant negative impact on ecosystems and the goods and services they provide. Received: Accepted: Foundation: National Natural Science Foundation of China, No ; Knowledge Innovation Project of Chinese Academy of Sciences, No.KZCX1-SW-19; Foundation Item of Key Laboratory of Ecological Restoration and Ecosystem Management of Jilin Province, No.DS Author: Wang Zongming (1976-), Ph.D, and Associate Professor, specialized in ecosystem ecology and remote sensing. zongmingwang@neigae.ac.cn Corresponding author: Zhang Bai (1962-), Ph.D and Professor. zhangbai@neigae.ae.cn

2 Estimates of loss in ecosystem service values of Songnen plain from 1980 to Songnen plain is one of the important agricultural bases, and has suffered dramatic land use changes as a result of activities of human beings. These changes led to aggravation of soil and water loss, decrease in soil fertility in the eastern black soil zone, and wetland loss, desertification and grassland degradation in the western part. Many efforts have been made to analyze climate change, landscape change, and effects of agricultural activities on local environment in Songnen plain (Li et ol., 1998; Liu, 2001; Meng and Zhang, 2001; Huang et 01., 2002; Luo and Zhu, 2002; Luo et 01., 2003; Meng, 2004). Yet there is no quantitative knowledge on changes in ecosystem services at regional level for the whole area (Liu, 2001). The objectives of this study are: (1) to qualify long-term changes in land use of Songnen plain, from 1980 to 2000; (2) to explore the impact of land use change on changes in ecosystem services of the study area; and (3) to detect the spatial differences of change in ecological values among counties of Songnen plain. Inner Mongolia Autonomous Region. '~ Liaoning Province Jilin Province Heilongjiang Province ~Northeast China Figure 1 Location of Songnen plain in Northeast China 2 Materials and methods 2.1 Study area Songnen plain is located in the central part of Northeast China (42 30'-51 20'N, ' 'E), and covers an area of about 239,184 km z (Figure 1). It includes 35 counties of Heilongjiang province and 20 counties of Jilin province. As one of the important agricultural regions of the nation, Songnen plain accounts for 12% of the grain yield of China. The climate is typical temperate continental monsoon climate between maritime humid and continental arid zones. The climate is characterized by cold, arid winter, and warm, rainy summer. The mean temperature in January is about (2, and the mean temperature in July is C. The average annual precipitation is mm and decreases from the east to the west. 2.2 Land-use change estimates using LANDSAT MSS and LANDSAT TM data Geographical information system technique was used in this paper to analyze the characteristics of the region in question spatially. Data sets incorporating Songnen plain during relatively cloud-flee days in September 1980 (MSS data) and August 2000 (TM data) were obtained. After the unsupervised classifications of the 1980 and 2000 data sets were completed, the resulting classes in each image were Table 1 Costanza et al. (1997) biome equivalents for the seven land use categories, and the corresponding ecosystem values Land cover categories Equivalent biome Ecosystem service coefficient ($ha -t per year) Forest land Temperate/boreal forest 302 Grassland Grass/rangelands 232 Farmland Cropland 92 Water area Lakes/rivers 8498 Marshland Wetlands Residential land Urban area 0 Unused land Desert 0

3 82 Wang Zongming, Zhang Bai, Zhang Shuqing et 01. grouped into seven land cover categories, i.e., forest land, grassland, farmland, water area, marshland, residential land, and unused land. 2.3 Assignments of ecosystem services values Despite many challenges, a few studies in the past few decades have tried to estimate the worth of natural resources. Most notably, Costanza et oi. (1997) presented a model for placing an economic value on different biomes and the services that they provided. The model we used in our analysis has been challenged. However, they represented the most reprehensive set of valuation coefficients available to us. In our study, the most representative biome was used as a proxy for each land cover category, including grass/rangeland for grasslands, temperate/boreal forest for forest land, cropland for farmland, lakes/rivers for water area, wetlands for marshland, urban area for residential land, and desert for unused land (Table 1). The total value of ecosystem services in Song/ten plain in 1980 and 2000 was estimated as: ESV : ~(A k xvc k) (1) where ESV is the estimated ecosystem services value, Ak is the area (km 2) and VCk O.e. the ecosystem service coefficient, as shown in Table 1) is the value coefficient ($ ha -~ per year) for land use category k. Due to uncertainties, we conducted sensitivity analysis to determine the dependence of our estimates of changes in ecosystem service values on the applied coefficients. The coefficient of sensitivity (CS) was calculated as: CS = (ESVj - ESV~)/ESV~ (2) (v G - vg )/vg where ESV is the estimated ecosystem service value, VC is the value coefficient, i and j represent the initial and adjusted values, respectively, and k is the land use category. The ecosystem value coefficients of all land use categories were individually adjusted by 50%. The impact of these adjustments on our estimates of the total value of ecosystem services and the associated Coefficients of Sensitivity (CS) are presented. 3 Results and discussion 3.1 Land use change detection Land use change of Songnen plain from 1980 to 2000 was shown in Figure 2 and Table 2. In the period from 1980 to 2000, farmland, residential land and unused land increased, but forest land, grassland, water area and marshland decreased correspondingly, among which grassland and water area decreased most dramatically. Songnen plain had 126,850 km 2 of farmland in 1980, which increased by 137,823 km 2 in During the 20 years, farmland increased by 10,973 km 2 and a change ratio of 8.65%. In contrast, the area of grassland decreased from 24,914 km 2 in 1980 to 18,594 km 2 in The size of this land cover category decreased by an estimated 25.37% between 1980 and During the 20 years from 1980 to 2000, water area decreased by 935 km 2 and a change ratio of Table 2 Area of each land use category in Songnen plain and changes in land use in 1980 and in 2000 Land use categories Farmland Forest Grassland Water Marsh- Residential Unused land area land Land land Area (kin 2) Area change (km ~) Change ratio (%) Change ratio per year (%)

4 Estimates of loss in ecosystem service values of Songnen plain from 1980 to Land use type Farmland 1 Forest land Grassland 1 Water area Matshland Imm Residential land Unused land Figure 2 Land use of Songnen plain in 1980 and in %. Meanwhile, there was an 8.58% decrease for marshland in the period. It was concluded from interpretation, the intensive ground-truth studies during the summer of 2002, and the interviews with relevant experts and farmers, the sizable increase in farmland resulted from the irrational reclamation from grassland, water area, even from marshland. In addition, during the past 20 years, elevated air temperature in this region accounts for partial incentives for reclamation. In the study '~ o) decreased from 33,367 krn 2, and has a 6.69% change ratio. Residential land including commercial and residential areas and their associated transportation surfaces has a 1.98% increase. 3.2 Changes in ecosystem services Using land use change data and the ecosystem service value coefficients O Area Beolo~i eal va lue~ munused land Residential land mmarshland QWater area NGrassland l~forest land mfarmland Figure 3 Area and value contribution of 7 land use categories in Songnen plain area, forest land Table 3 Percentage change in estimated total ecosystem 35,758 km 2 to service value and coefficient sensitivity (CS) resulting from adjustment of ecosystem valuation coefficients (VC) Change in valuation coefficient % CS % CS Farmland VC±50% _ _ Woodland VC_+ 50% _ _ Grassland VC±50%,1, ,1, Water area VC±50% -t ,1, Marshland VC_+_ 50% _ _

5 84 Wang Zongming, Zhang Bai, Zhang Shuqing et el. reported by Costanza et ol. (1997), we found that land use changes in our study area resulted in an average net decline of $ million per year in ecosystem services between 1980 and This means a 9.11% net decline in annual value of ecosystem services. Assuming a linear decline in ecosystem services, this indicates a cumulative loss of $31,815.6 million in ecosystem services over the 20-year period of the study. The shifts show the relative proportions of each land category and corresponding ecosystem service values in 1980 and 2000 (Figure 3). 3.3 Ecosystem service sensitivity analysis The effects of using altemative coefficients to estimate total ecosystem service values in the study area in 1980 and 2000 are shown in Table 3. The CS of these analyses was less than unity in all cases. Adjustment to the value coefficient for farmland, woodland, grassland and water area categories had very little effect on the estimates of total ecosystem service values (< 2% change for a 50% change in value coefficient). The coefficient of sensitivity ranged from a low of to for grassland and farmland categories to a high of when the coefficient for water area category was adjusted. The relatively low coefficients of sensitivity (percentage change in total ecosystem service value per unit change in a land-cover value coefficient) reflect that total ecosystem values estimated for the study area are relatively inelastic, with respect to the ecosystem service coefficients. 3.4 Detection of spatial variation in change of ecosystem service values The spatial distribution of ecosystem service values of Songnen plain in 2000 for 55 counties was illustrated in Figure 4. Spatial variation of change of ecosystem values in Songnen plain from 1980 to 2000 is showed in Figure 5 and statistics of change in ecosystem service values in different counties are listed in Table 4. We can find that three counties' ~osystemser~ce lues (million $) ~<200 ~ ~-~ ~T]-111ooo-15oo ~ ~ ~>3000 O ~130 Km I1,1 Figure 4 Spatial distribution of ESV of Songnen plain in 2000 ecosystem.lues (%) ~ O0 Km Figure 5 Spatial variation of changes in ecosystem values among counties of Songnen plain

6 Estimates of loss in ecosystem service values of Songnen plain from 1980 to Table 4 Statistics of change in ecosystem service values in different counties of Songnen plain Change Number Number of Number of Name of counties ratio of counties in counties in counties Heilongjiang Jilin Province Province 10 /~34% %-10% c7~5% %--~% V ---10% V~-20% %--30% Zhaodong Lanxi, Acheng Changling, V~chang, Yian, Tailai, Beian, Keshan, Lishu, Daqing, Anda, Songyuan, Qianan, Hailun, Yushu, Zhaoyuan, Mingshui, Wangkui, Qinggang, Qiqihar, Baiquan, Jiutai, Kedong, Shuangyang, Suihua, Qingan, Shuangcheng, Suileng, Lindian, Gongzhuling, Hulan, Changchun, Yitong, Wudalianchi, Mulan, Tongyu, Zhenlai, Bayan Harbin, Binxian, 2~haozhou, Fuyu, Nongan Dehui, Fuyu, Nehe, Dorbod, Longjiang, Gannan, Shuangliao Qianguo, Taonan Baicheng, Daan ecosystem service values increased by more than 5%. They are Zhaodong, Lanxi and Acheng, all affiliated to Heilongjiang province. For Zhaodong county, from 1980 to 2000, the ecological values increased by 33.34% due to conversion from cultivated land to grassland and water area. Meanwhile, 16.4% of unused land has been converted to grassland. For Lanxi county, the increase of 6.64% in ecological values is the result that 2.06% of cultivated land and 13.8% of unused land were converted to grassland. As for Acheng county, we found there was a 5.54% increase in ecosystem service values, which resulted from the land use change similar to that of Lanxi county. All these shifts are contributed to the "returning farmland to grassland" policy adopted by Heilongjiang province since the end of the 1990s. There were 36 counties whose ecosystem services hardly changed (with a change of -5% to 5%), among which 24 counties are affiliated to Heilongjiang province and 12 counties to Jilin province. Five counties have a -5%--10% decrease in ecosystem services. They are Harbin city, Binxian and Zhaozhou counties of Heilongjiang province, and Fuyu and Nongan counties of Jilin province. Five counties of Heilongjiang province, i.e., Fuyu, Nehe, Dorbod, Longjiang and Gannan, and two counties of Jilin province, Dehui and Shuangliao counties have a -20%--10% decrease. The obvious phenomenon is that the four counties with a dramatic -57%--20% of ecological value decrease are all located in the western part of Jilin province. These counties had been considered as the areas with fragile ecological environment. Our study results confirmed that due to irrational utilization of water and soil in this region, the ecological quality degraded to a large extent from 1980 to 2000, compared with other counties in Songnen plain. 4 Conclusions Based on the estimated size of seven land cover categories and Costanza et a/.'s ecosystem services values for corresponding biomes, we determined that the total annual ecosystem service values in Songnen plain, Northeast China declined from $34, to $31, million per year from 1980 to This about 9% decrease resulted from the conversion of grassland, water area, and marshland into farmland and urban area due to increasing demand for food and residence. As for spatial variation, the northern and eastern parts of Songnen plain have a comparatively less loss of ecological value, even in Zhaodong county, there is a 30% increase of ecological values. In contrast, ecological values of four counties in the western part of Jilin province decreased from 20% to 57%. There was a process from natural ecosystem, semi-manmade ecosystem to artificial

7 86 Wang Zongming, Zhang Bai, Zhang Shuqing et ol. ecosystem nowadays in Songnen plain, during which human activities played an important role. Land cover modification and conversion are driven by the interaction in space and time between biophysical and human dimensions. The increase of size of farmland and residential land in Songnen plain from 1980 to 2000 was related to rapid population growth and improved quality of life since the reform and opening up policy was adopted in China in the late 1970s. Increased human activities led to the acceleration of deforestation and cultivation of grassland and wetlands. As for policy formulation of land use plan, we argue that, in future, conservation of such important ecosystems as grassland, water body, and marshland should take precedence over the single-minded reclamation of these areas for economic purposes. References Costanza R, d'arge R, Groot R et d., The value of the world's ecosystem services and natural capital. Nature, 387(15): Daily G, Nature's Services: Societal Dependence on Natural Ecosystems. Washington DC. Houghton R A, The U.S. carbon budget: contribution from land-use change. Science, 285: Huang Fang, Liu Xiangnan, Ye Baoying et al., Land use change of the ecotone in the west part of Songnen Plain. Journal of Northeast Normal University, 34(1): (in Chinese) Li Qusheng, Qiu Shanwen, Deng Wei, Study on the secondary saline-alkalization of land in Songnen Plain. Scientia Geographica Sinica, 18(3): (in Chinese) Liu Xingtu, Management on Degraded Land and Agricultural Development in the Songnen Plain. Beijing: Science Press. (in Chinese) Luo Xinzheng, Zhu Tan, Sun Guangyou, Influence of human activities on ecological environment of Songnen Plain. China Population, Resources and Environment, 12(4): (in Chinese) Luo Xinzheng, Zhu Tan, Sun Guangyou et al., Cause and countermeasure of wetland desertification in Songnen plain. Journal of Desert Research, 23(4): (in Chinese) Meng Kai, Zhang Hongyan, Analysis on successional law of agfi-ecological system in black soil region of Songnen plain. System Sciences and Comprehensive Studies in Agriculture, 17(4): (in Chinese) Meng Kai, The sustainable development of the black soil arable crops agro-ecology system in Songnen plain. System Sciences and Comprehensive Studies in Agriculture, 20(1): (in Chinese) Sala O E, Biodiversity: global biodiversity scenarios for the year Science, 287: Tolba M K, The World Environment : Two Decades of Challenge. London: Chapman & Hall. Turner IIB L, Skole D, Sanderson S et al., Land-Use and Land-Cover Change. Science/Research Plan (IGBP Report No. 35, HDP Report No. 7). IGBP of the ICSU and HDP of the ISSC, Stockholm and Geneva. Vitousek P M, Mooney H A, Human domination of earth's ecosystems. Science, 277: Zhong Linsheng, Zhang Yangmin, Zhao Shidong et al., Land use change and its ecological effect in Qian'an County of Jilin Province. Journal of Geographical Sciences, 13(4):