| Land soil carbon pool is the largest carbon pool on the surface of the earth, and the carbon storage in soil is about 2500Pg on the global. Soil carbon pool is about 3.3 times as the atmospheric carbon pool (760Pg), and is 4.5 times as plant carbon pool (560Pg). Carbon storage in farmland soil is 8-10 percent of soil carbon pool。Generally, soil organic carbon density will be reduced by agricultural use, and has been reduced 30-50 percent of soil organic carbon. Most of carbon released into the atmosphere in the form of CO2, the contribution rate of agro-ecosystem to atmospheric greenhouse gases has get 20%. In the carbon pool of global land ecosystem, the carbon pool of farmland ecosystem is affected strongly by human disturbance and can be adjusted in a short term. Currently, the achievement of carbon storage of farmland ecosystem in small-scale is still few. Therefore, to study the organic carbon storage in small-scale is positive significance on the sustainable agricultural development, maintain and improve the capacity of agricultural soil carbon sequestration, food security and mitigation of climate change trends.In this paper, Mapo town which is in Weishan County, Shandong province was taken as the study area, the conventional indicators of surface soil in farmland ecosystem and vegetation information was measured. Combine remote sensing image and basic map such as administrative map, 1:50000 topographic map, land use status maps and so on, we try to apply the remote sensing to soil organic carbon storage. Farmland and four vegetation indices were extracted, and remote sensing estimation model of aboveground biomass based on vegetation index was build. Based on the measured data of soil and vegetation, the relationship between various ecological factors and soil organic carbon was analyzed. Impact factor assembly and simulation model of surface soil organic carbon was build. At last, Combine surface soil organic carbon storage and vegetation carbon storage, organic carbon of farmland ecosystem was estimated. The main results obtained are summarized as follows: (1) Applied supervised classification and unsupervised classification of human intervention, the remote sensing image was classified. The result of unsupervised classification is better, and its process is more sample and fast. Classification test results show that crop classification result with high accuracy can be acquired by using the high spatial resolution remote sensing image and extracting directly crop area.(2) By analyzing the correlation between soil organic carbon and ecological factors, we obtain that the ecological factor which is significant correlation with soil organic carbon include sand, silt, clay, alkali-hydrolyzed N, available potassium, soil water content, soil bulk density. Regression model of soil organic carbon was established by multiple linear regression analysis, and the best model was 8-factor model (aboveground biomass, clay, sand, surface soil temperature, straw K, soil water content, soil alkali-hydrolyzed N, soil available Potassium) and 9-factor model (aboveground biomass, clay, sand, surface soil temperature, straw K, soil water content, straw N, soil alkali-hydrolyzed N, soil available Potassium).(3)The correlation between aboveground biomass and vegetation index is well. It is possible to build the remote sensing estimation model of farmland crop biomass based on the vegetation index. However, there are some differences between different remote sensing data sources, vegetation indices and types of models. In terms of different remote sensing data sources, correlation coefficient between SPOT-5 vegetation indices and aboveground corn biomass is higher than that of ETM vegetation indices, and its value were all more than 0.8. In addition, among the vegetation indices extracted from ETM image, in addition to SAVI, correlation coefficient of the others were all more than 0.8. In terms of different vegetation indices, the correlation between aboveground biomass and four vegetation indices is well, and the significance level gets to 0.01. But the correlation of NDVI is highest, and the value is 0.889 and 0.907 partly. RVI and RDVI is followed, and the value of SAVI is lowest with 0.792 and 0.841. Compared with the ETM data, the band information of the SPOT-5 data can meet the requirements. The SPOT-5 data can provide the higher spatial resolution, and have better correspondence with the location of measured aboveground biomass samples, and is more suitable for small-scale remote sensing of biomass.(4)Combined the measured surface soil organic carbon content and biomass carbon storage, we estimate the surface soil organic carbon storage of farmland ecosystem. The result of carbon pool storage as follow: the organic carbon storage of aboveground biomass is 3.18×10~7kg; the organic carbon storage of underground root is 3.18×10~6kg; the organic carbon storage of surface soil is 1.52×10~8kg.
|
PDF Full Text Request