| Climate warming is one of the biggest challenges to human society development and ecosystem environmental stability.Carbon emission because of the fossil fuel combustion and land use and cover change(LUCC)is considered to be the primary driving force of global warming.LUCC has been recognized as one of the most important human factors that influenced the Earth surface.Present studies indicated that about one third to one half of the Earth's land surface has been changed by human activities.Such rapid changes have exerted significant effects on ecosystem's structure and functions,and ultimately influenced the global carbon cycle.Grasslands,one of the most common vegetation types in the world,account for nearly 20%of the global land surface.The vast land cover and carbon sequestration potential,make it become the important composition of the terrestrial carbon cycle.In China,grasslands mainly locate in the arid and semi-arid areas in the northwest and the Tibetan Plateau alpine climate regions,which make it sensitive and vulnerable to climate change and frequent human intervention.Along with the global warming and population explosion,substantial LUCC has occurred in China due to overgrazing,grassland reclamation and over-mining explorations.All these changes had led to serious ecological problems,such as grassland degradation,desertification,and future changes of the global carbon cycle.In recent decades,Chinese government has initiated several ecosystem restoration programs to mitigate the environment degradation.This study aimed to estimate grassland productivity,explore the spatial-temporal dynamic of grassland carbon source/sink in response to climate change and LUCC,make clear current situation of grassland degradation in China,and quantitatively assess the relative driving forces of climate change and human activities to grassland degradation.The outcomes not only has important theoretical significance for terrestrial ecosystem carbon cycle,but be beneficial to evaluating the performance of ecosystem restoration programs and rational ultilization of grassland resource.Therefore,in the study,by combing the multi-source NDVI data,LUCC,climate data and statistic yearbook,based on light use efficiency model(Camegie-Ames-Stanford approach,CASA)model,we simulated the grassland NPP and analyzed its spatial-temporal dynamic in response to climate change and LUCC.Then,we estimated the grassland net ecosystem productivity(NEP)based on the ecosystem remote sensing model(Boreal Ecosystem Productivity Simulator,BEPS)model.The inter-annual and monthly change dynamic of grassland carbon source or sink was explored,and the correlation between grassland NEP and climate factors was analyzed.In addition,the grassland NPP and coverage was selected as the grassland degradation monitoring indicator to class the degradation level and analyze the grassland degradation situation.Meanwhile,a quantitative assessment method based on potential NPP and the NPP loss because of human activities was established to distinguish the relative roles of climate change and human activities in grassland degradation,and determine the dominant driving factor for different regions.This study accomplished the following tasks:1.Vegetation coverage,an important indicator for evaluating grassland ecosystem condition,is used to monitor grassland vegetation dynamic change.GIMMS NDVI from 1982 to 2006 and MODIS NDVI from 2001 to 2010 were adopted to create a long time serials of NDVI datasets from 1982 to 2010.Then the NDVI datasets were used to estimate the time series of grass coverage,and analyze its spatial distribution and changing trend for the total grasslands and six different grassland types.Meanwhile,grass coverage changes in response to climate factors at inter-annual and monthly time scales were analyzed,combined with the mean annual temperature and precipitation data.Results showed that the nationwide land surface temperature had increased with a rate of 0.04 C/year,while precipitation had decreased with a mean rate of-0.39 mm/year;the warm-wet trend is obvious in northwest China.The spatial pattern of grass coverage showed an increase trend from northwest to southeast across China.During 1982-2010,the mean grass coverage was 34.5%in China,and exhibited obvious spatial heterogeneity,being highest(61.4%)in slope grasslands and lowest(17.1%)in desert grasslands.There was a slightly increase of grass coverage with a rate of 0.17%per year over the study period,and 78.9%of grassland areas in coverage showed an increasing trend.Regionally,the largest increase of the grass coverage was observed in the northwest region and Tibetan Plateau.Increasing trend in slope grassland was as high as 0.27%per year,while in the plain grassland and meadow,the grass coverage increase was the lowest(being 0.11%and 0.1%per year,respectively).Across China,the grass coverage with extremely significant(P<0.01)and significant(P<0.05)increases area accounted for 46.0%and 11.0%of the total grassland area,respectively,while those with extremely significant and significant decrease encompassed only 4.1%and 3.2%,respectively.At the annual time-scale,there are no significant correlations between grass coverage and annual temperature and precipitation for the total grassland area.However,the grass coverage was more affected by temperature than precipitation in alpine and sub-alpine grassland,alpine and sub-alpine meadow,slope grassland and meadow,while grass coverage in desert grassland and plain grassland was more affected by precipitation.At the monthly time-scale,there are significant positive correlations between grass coverage with both temperature and precipitation,indicating that the grass coverage is more affected by seasonal fluctuations of hydrothermal factors.Additionally,there is an obvious time-lag effect between grass coverage and climate factors for each grassland type:correlations are highest between the grass coverage and former one month's temperature and precipitation.2.Based on the time serials of NDVI datasets from 1982 to 2010,land cover data and meteorological data,the grasslands NPP of China during 1982-2010 were simulated by the CASA model.The spatial-temporal dynamic of grasslands NPP and the frequency distribution characteristic of different grassland type's NPP were analyzed;meanwhile,the relationships between grasslands NPP and annual temperature and precipitation were analyzed.The results showed that the grasslands mean NPP was 279.4 gC/ml/yr,showing an obvious decreasing trend from southeast to northwest with a changing rate of 228.6 gC/m2/yr/100;and the total NPP of grassland was 918.4 Tg C/yr(Tg=1012g).The grasslands NPP exhibited an increasing trend during the 29 years with a rate of 0.56 gC/m2/yr,and the areas showing an increasing trend accounted for 68.75%of the total grasslands area.Spatially,the extremely significant increase regions of NPP mainly distributed in south and southeast of Tibetan Plateau,south slope of Tianshan Mountains,Mu Us Sandy Land;and the obvious decreasing trend of NPP mainly located in Hulun Buir and Hunshadake Sandy Land of Inner Mongolia.There are non-significant positive correlation between grasslands NPP and temperature(R=0.12)and precipitation(R=0.2),and precipitation was the main influence factors for the grasslands NPP.In addition,the response of grasslands NPP to temperature and precipitation varied greatly among the different grassland types.Alpine and sub-alpine meadow was more affected by temperature than precipitation with a correlation of 0.15 and 0.03 respectively,and the pixels numbers with positive correlation between NPP and temperature accounted for 68.5%of the total pixels numbers.There was positive correlation between Slope grassland NPP and precipitation,but negatively correlated with temperature,and the pixels numbers with positive correlation between NPP and precipitation accounted for 81.6%of the total pixels numbers.However,NPP of the following grassland types,such as plain grassland,desert grassland and meadow,has more positive correlation with precipitation than temperature,especially for desert grassland,there was significant positive relation with precipitation;Statistically showed that the percentage of pixel numbers with positive correlation between NPP and precipitation was 84.7%,93.5%and 92.1%,respectively.For Alpine and sub-alpine grassland,the correlation between NPP and temperature and precipitation was nearly equal.3.Based on multi-source remote sensing image,leaf area index(LAI),climate data and physiological parameter data,the grassland NEP was simulated using the ecosystem remote sensing model(BEPS model).And the annual and monthly dynamic of NEP and its response to climate factors was analyzed.Besides,the inter-annual change tendency of the grassland carbon transformation efficiency(NPP/GPP),water stress index(evaporation/precipitation)and water use efficiency(NPP/precipitation)was also explored.Results showed that the grassland NEP was 13.6 gC/m2/yr during 1979-2008,indicating that China's grassland was a carbon sink.The area with NEP greater than zero accounted for 73.1%of the total grassland areas.The overall distribution pattern represented a descending trend from southeast to northwest in China and the regions with NEP greater than zero corresponded with the precipitation larger than 200mm.The net carbon sequestration of grassland was 26.6 Tg C per year.At annual scales,the grassland NEP was greater than zero from 1979 to 2008;and at monthly scales,the changing pattern of NEP could be expressed as inverse U curves,with positive value from June to September showing as a carbon sink,the rest of months,the grasslands NEP was negative showing as a carbon source.During the thirty years,the grassland carbon transformation efficience and Rain use efficience showed significant increase.There was positive correlation between grasslands NEP and precipitation,and weak positive correlation with temperature;meanwhile,the area percentage of the two kinds of correlation was 74.2%and 50.2%.And the rising precipitation promoted the increase of NEP,especially in the northeast and middle parts of Inner Mongolia,west of Xinjiang,Southwest of Tibet,and regions of Qinghai,and there were significant positive correlation between NEP and precipitation.4.LUCC in the Shiyanghe River Basin and its impacts on the grasslands NPP were analyzed by combining the land use data of 2001 and 2010 and the CASA model.The relative roles of LUCC,land management measures and climate change on the NPP variaitons were explored using the actual NPP and potential NPP.During the 10 years,the net increase of grassland area was 5105.5 km2,and 80.4%of the newly developed grasslands were attributed to desert-to-grassland conversion.And the conversion from grassland to farmland was the main reason that cause the decrease of grassland area,the conversion area was 1119 km2,accounting for 64.9%of the total grassland area loss.The total NPP of grasslands in 2001 was 2072.6 Gg C,2732.29 Gg C in 2010,which increased by 659.62 Gg C(Gg=109).The contributions of human activities(Land use cover change and land management measures)and climate change to total NPP net increment were 133%and-33%,respectively.For the land cover change region,when the influence of climate factor on NPP change was extracted,the land use change led to an increase of 654.82 Gg C,the conversion from desert and farmland to grassland led to an increase by 674.46 Gg C and 156.85 Gg C,while the conversion from grassland to farmland led to a decrease of 235.64 Gg C.For the land cover unchanged region,the grasslands NPP increased by 219.97 Gg C as a result of improved land management measures.Therefore,land conversion and improved management measures directly lead to the increase of grassland NPP.These factors are dominant positive driving forces for NPP increase;whereas warm and dry climate imposed adverse effects on grasslands NPP in the study areas which led to a decrease by 215.17 Gg C.Lastly,the quantitative assessment method in this paper,can be effectively assess the performance of the ecosystem restoration program.5.The grassland area,landscape index and NPP in 7 provinces of northwest China in 1985,1995,2000 and 2010 was analyzed based on NDVI remote sensing image,Land cover data such as NRED(National resource and environment data)and MCD12Q1,climate data and statistical yearbooks.The grassland area in 1985,1995,2000 and 2010 were 2.479,2.435,2.462 and 2.318 million km2,respectively.The grassland area decreased by 161 500 km2 from 1985 to 2010 and accounted for 6.52%of that in 1985,which mainly resulted from the decrease of grassland area in Xinjiang and Tibet,however,the grassland area significantly increased in Inner Mongolia and Qinghai.Compared with the period of 1985-1995 and 1995-2000,the landscape heterogeneity and fragmentation of the study area's grassland decreased in the period of 2000-2010,and the dominance of grassland increased,the shape of grassland patches became simplification and regulation along with the decrease of AWMSI(Area weighted mean shape index)from 2000 to 2010.And the changing trend of AWMFD(Area weighted mean fractal dimension)showed that the human intervention decreased from 1995 to 2010;and the grassland patches connectivity and cohesion increased with the IJI(Interspersion juxtaposition index)decreased from 2001 to 2010.There is obviously spatial-temporal difference for the mean grasslands NPP.During 1985-1995,2000-2010 and 1985-2010,the mean grasslands NPP in the study area increased by 14.34%,25.82%and 40.95%,respectively.But during 1995-2000,grasslands NPP decreased by 4.82 gC/m2/yr,accounting for 2.03%of that in 1995.The total grasslands NPP for the whole study area reached the highest value in 2010 and increased by 31.89%of that in 1985 and 18.4%of that in 2000.These changes in grassland area,landscape index and NPP were resulted from the interaction of climate change and human activities,and the effects of the two factors were different among the seven provinces.The implementation intensity of ecosystem restoration programs was largest in Inner Mongolia,and its degraded grassland indeed obtained good restoration.6.The grassland degradation remote sensing monitor was conducted based on grasslands NPP and vegetation coverage.And the driving factors of grassland degradation were quantitatively assessed based on potential NPP and the diffidence between potential NPP and actual NPP.In addition,the spatial distribution and contribution of the dominant driving factors of grassland degradation was determined.Results showed that the mean grasslands NPP and coverage was 281.7 gC/m2/yr and 46%in 2001-2010.During the period of 1982-2010,22.73%of grassland areas underwent degradation,whereas only 31.65%exhibited restoration trend,the rest of 45.62%remained unchanged.For the grassland degradation,the contribution of climate change and human activities nearly reached equilibrium(44.86%43.36%).By contrast,78.08%of grassland restoration was caused by human activities,whereas 21.14%was caused by climatic factors.There were significant regional differences of the contribution of climate and human factors to grassland degradation and restoration.For the grassland degradation,regions with the contribution of climate change larger than human activities included Inner Mongolia,Shaanxi,Tibet,Gansu,Qinghai and Sichuan;while in Yunnan,Ningxia and Xinjiang,human activities was the dominant driving factors for degradation,especially in Xinjiang,85.13%of the degradation was attributed to human activities.For the grassland restoration,the contribution of climate factors larger than human activities only included Tibet,and its percentage was 59.69%;for the rest of eight provinces,human activities were the dominant driving factors,especially in Sichuan and Shaanxi,more than 99%of restoration was attributed to human activities.According to the above results,we found that there were significant difference of ecosystem restoration program implementation regions and intensity;it mainly focused on the serious degraded grassland or desertification region.For example,the implementation intensity of program was largest in Inner Mongolia,and also the contribution of human activity to grassland restoration was 94.2%;however,the implementation intensity was least in Tibet and Xinjiang,and its contribution to grassland restoration was 40.25%and 64.31%.There are several innovations of this study:Firstly,the regression equation between GIMMS-NDVI and MODIS-NDVI was build in order to extend the NDVI time series from 1982 to 2010.In order to improve the simulation accuracy of NPP by CASA model,we use the different max light use efficiency for different vegetation types.The spatial-temporal dynamic of grasslands NPP and its response to climate change and LUCC were analyzed at different spatial and time scales.The research indicators in this study included grassland coverage,LUCC,landscape index,NPP,and statistical yearbook data.Secondly,at present,the observation and simulation of grassland NEP mainly foucused on the homogenious samples,lacking of large scale NEP simulation.In this paper,we simulated the grassland NEP using the BEPS model in China through parameter optimization accoriding to the grassland physiological ecology parameter.Then the spatial-temporal dynamic of grasslands NEP and its response to climate hydrothermal factors change were analyzed;then the carbon source or carbon sink function of grassland ecosystem at nationswide scale were determined.Thirdly,study on the grassland degradation monitoring;previously mainly foucused on the field observation lacking of remote sensing monitoring on the large scales.Therefore,in this paper,the grassland degradation remote sensing monitoring was conducted in nationwide scales based on NPP and grass coverage.Meanwhile,using the potential NPP and the loss of NPP because of human activities,the relative contribution of climate change and human activities to grassland degradation was quantitatively assessed.These will help us to determine the driving contribution of each factor at different regions and make clear the dominant factors of grassland degradation.The outcomes will provide important suggestions for reasonable adjustment of ecosystem restoration programs;meanwhile,research conclusions also reflected the performance of the restoration programs.There were huge spatial heterogeneity of the grassland NPP and NEP in China,because of the vast distribution area of grassland and complex terrain and climate conditions.Along with the global warming and human interference intensification,grasslands in China have been seriously degraded.Based on the carbon accounting results of China's grassland ecosystem,we found that grasslands in China has been an obvious carbon sink in 1998-2010,and showed great spatial heterogeneity;there existed positive correlation between annual precipitation and grassland NPP,and NEP.Based on the quantitatively analysis of driving factors that lead to grassland degradation,the roles of climate and human factors in grassland degradation are quite similar,whereas human activities dominate the grassland restoration process.The implementation of ecosystem restoration programs indeed promoted changes in grass coverage,productivity and carbon sequestration potential,and obtained approving restoration effectiveness.However,there also existed some problems of these programs,such as the grassland degradation governance investment was not enough and had great regional differences,these measures mainly implemented in these serious degraded and desertification regions,which lead to the ecological environment construction showed a situation in China:local improvement,but overall deterioration.Therefore,the ecosystem restoration programs need reasonable adjustment and increase investment,increased grassland carbon sink function and give full play to grassland ecological service values. |
PDF Full Text Request