Quantifying methane emissions from rice paddies in Northeast China by integrating remote sensing mapping with a biogeochemical model

摘要

The Sanjiang Plain located in Northeastern China is one of the major riceproducing regions in the country. However, differing from the majority riceregions in Southern China, the Sanjinag Plain possesses a much coolerclimate. Could the rice paddies in this domain be an important source ofglobal methane? To answer this question, methane (CH₄) emissions fromthe region were calculated by integrating remote sensing mapping with aprocess-based biogeochemistry model, Denitrification and Decomposition orDNDC. To quantify regional CH₄ emissions from the plain, the model wasfirst tested against a two-year dataset of CH₄ fluxes measured at atypical rice field within the domain. A sensitivity test was conducted tofind out the most sensitive factors affecting CH₄ emissions in theregion. Based on the understanding gained from the validation andsensitivity tests, a geographic information system (GIS) database wasconstructed to hold the spatially differentiated input information to driveDNDC for its regional simulations. The GIS database included a rice mapderived from the Landsat TM images acquired in 2006, which provided crucialinformation about the spatial distribution of the rice fields within thedomain of 10.93 million ha. The modeled results showed that the total1.44 million ha of rice paddies in the plain emitted 0.48–0.58 Tg CH₄-Cin 2006 with spatially differentiated annual emission rates ranging between38.6–943.9 kg CH₄-C ha?1, which are comparable with that observed inSouthern China. The modeled data indicated that the high SOC contents, longcrop season and high rice biomass enhanced CH₄ production in the coolpaddies. The modeled results proved that the northern wetland agroecosystemscould make important contributions to global greenhouse gas inventory.