|其他摘要||As global environmental problems have become increasingly prominent, the effects of climate change and human activities on the terrestrial carbon cycling have been widely concerned. The ecosystem of arid areas is very fragile, and sensitive to climate change and anthropogenic activities. Especially the agricultural ecosystem in the arid oasis has become more active under the oasis expansion, growing agricultural management activities and climate warming, therefore, its carbon dynamics is of great significance for carbon cycling study in the arid regions. However, there are still lack of studies on the quantitative modeling and evaluation of carbon dynamics of arid agricultural ecosystem on a regional scale and its response to climate and agricultural management activities. This is mainly due to the limitations of the research methods for spatio-temporal simulation of the agricultural ecosystem carbon cycle and the difficulty of obtaining the required data.
This study comprehensively evaluated the carbon dynamics from single point to regional scale and its temporal-spatial characteristics as well as its response to climate change and agricultural management in the agricultural ecosystem of Xinjiang, a typical arid zone in the central Asia. A comprehensive ecosystem model (Agro-IBIS) adapted from the dynamic global vegetation model (DGVM) — the Integrated Biosphere Simulator (IBIS) was used for simulation and assessment of the carbon budget in the study area. For this research goal, the long-term (1901-2009) monthly mean observed climate dataset (CRU), 30-year (1961-1990) monthly mean climate anomaly dataset (CRU05) and (NCEP/NCAR) (1948-2009) daily reanalysis dataset（CFSR）were collected and processed to the required format for Agro-IBIS model. A spatial climate, soil and vegetation dataset was set up for the arid zone in Xinjiang. The Agro-IBIS model was calibrated and evaluated, and applied to temporally and spatially estimate the carbon dynamics of oasis agricultural ecosystem in Xinjiang. The relationships between the carbon dynamics and agricultural management, climate change as well as the CO2 concentration were analyzed through scenario modeling approach to put forward rational agricultural land use and management system for sustainable development of agricultural ecosystems. The main results are:
(1) The Agro-IBIS model can be applied to simulate the carbon dynamics in the arid oasis agricultural ecosystems in Xinjiang according to the results from the model evaluation (crop LAI, Biomass and NPP) at Wulanwusu agricultural climate observation site.
(2) The regional simulation results for the period of 1979-2009 shows that the agricultural ecosystems in Xinjiang acted as a net carbon sink, with a NEE mean value of -63.36 g C m-2 yr-1, and NPP value of 723.78 gC m-2year-1. The mean carbon stock in Xinjiang farmland over 33 years was about 237.15 TgC with the vegetation and soil carbon stock of 19.5TgC and 217.65TgC, respectively. There was stronger relationship between the carbon dynamics and the temperature; however, the weak relationship was found between the carbon dynamics and precipitation anomaly in the last 30 years. This indicates that there is closer relationship between agricultural ecosystem NPP and temperature. This is mainly due to irrigation agriculture dominated in Xinjiang, in which crop growth mainly relies on the irrigation, and overall effects of rainfall on it is very small. However, the trend of warmer and wetter climate in Xinjiang during the last 30 years accelerated the promotion of agricultural ecosystem NPP.
(3) The appropriate fertilization and irrigation increase NPP of cropland in arid area, especially agricultural ecosystem NPP increases as increasing fertilization and irrigation amount under the non-changing climate condition. It was found that there was strong correlation between the agricultural ecosystem NPP and the fertilization as well as irrigation amount. During the period of 1979-2009, the amount of fertilization in Xinjiang cropland increased by near 7-times, and crop yield also increased at the rate of 78.22kg/ha year-1. The contribution of irrigation to NPP was about 78.37%, and the reliance of cropland of arid area on the irrigation was greater than that on the fertilization.
(4) The human impact on the crop growth in the arid oasis was far greater than the impact of natural factors. The arid oasis agriculture was irrigation-based with assisted by other agricultural management such as fertilization. Total Crop NPP simulated by Agro-IBIS model under climatic vacillation was generally low, of which NPP for sprig maize was 166 ~ 502 g C m-2, and NPP for winter wheat was 163 ~ 332 g C m-2. It increased by about 20% with adding fertilization, while it increased significantly fertilization + irrigation, which was reached to 753 ~ 1294 g C m-2 for spring maize and 493~632 g C m-2 for winter wheat, respectively. This indicates that crop NPP simulated under climate with irrigation was significantly higher than that with fertilization. Crop NPP also increases under elevated CO2 concentration in the atmosphere. Total simulated NPP in Xinjiang cropland was 41.49 TgC year-1 under the current CO2 level (380ppm), and which reached up to 47.13 TgC year-1 when CO2 elevated to 550ppm.
(5) This study estimated the ideal fertilization level by scenario modeling analysis with different fertilization amounts that can effectively promote the crop yield. Excessive fertilization is not able to increase crop yield, but is likely to accelerate the soil N leaching loss. Irrigation was the main method that can make up the part rainfall was not enough for crop growth in that area, where the effects of rainfall was lower during the wet years and higher during the dry years. Adaptive management for the cropland could help maintain the steady and sustainable development of agricultural ecosystem in the arid oases, as well as strengthen its ability to sequester carbon.
This study implemented the application of Agro-IBIS model in Xinjiang oasis croplands, which made it possible to evaluate the carbon-water cycling of the agricultural ecosystem temporally and spatially despite the limited numbers of experimental sites and the varieties of crop types. The results obtained in this study could help assess the carbon budget spatially and temporally, as well as its biochemical mechanism in the arid areas. It could also help evaluate the effects of climate change and agricultural management on the carbon dynamics. Overall, it contributes to make sustainable development strategy for arid agricultural ecosystem and provide scientific support to make regional carbon management plan for climate adaptation and mitigation.|