B038-0014
Scaling Up Rice Methane Fluxes in Monsoon Asia Through Data-Driven Model
Wednesday, 9 December 2020
Poster
Zutao Yang1, Robert B Jackson1, Benjamin Runkle2, Dario Papale3, Gavin McNicol4, Etienne Fluet-Chouinard1, Sara Knox5, Maricar C. R. Alberto6, Dennis D Baldocchi7, Alessandro Cescatti8, Chi-Ling Chen9, Jinwei Dong10, Haiqiang Guo11, Geli Zhang12, Hiroki Iwata13, Qingyu Jia14, Weimin Ju15, Minseok Kang16, Hong Li17, Joon Kim18, Hao Lu19, Michele L. Reba20, Debora Regina Roberti21, Youngryel Ryu22, Benjei Tsuang23, Xiangming Xiao24, Wenping Yuan25 and Yongguang Zhang26, (1)Stanford University, Stanford, CA, United States, (2)University of Arkansas, Fayetteville, AR, United States, (3)University of Tuscia, Department for Innovation in Biological, Agro-food and Forest systems (DIBAF), Viterbo, Italy, (4)Stanford University, Earth System Science, Stanford, CA, United States, (5)University of British Columbia, Geography, Vancouver, AB, Canada, (6)International Rice Research Institute, Laguna, Philippines, (7)University of California Berkeley, Department of Environmental Science, Policy, and Management, Berkeley, CA, United States, (8)Joint Research Centre Ispra, Ispra, Italy, (9)Taiwan Agricultural Research Institute, Wufeng, Taichung, Taiwan, (10)Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences, Beijing, China, (11)Fudan University, Ministry of Education Key Laboratory for Biodiversity Science and Ecological Engineering, and Coastal Ecosystems Research Station of the Yangtze River Estuary,, Shanghai, China, (12)China Agricultural University, College of Land Science and Technology, Beijing, China, (13)Shinshu University, Matsumoto, Japan, (14)China Meteorological Administration, Institute of Atmospheric Environment, Shenyang, China, (15)Nanjing University, Nanjing, China, (16)Seoul National University, Seoul, South Korea, (17)Chongqing University, Faculty of Architecture and Urban Planning,, Chongqing, China, (18)Seoul National University, Landscape Architecture & Rural Systems Engineering, Seoul, South Korea, (19)Nanjing University of Information Science and Technology, Jiangsu Key Laboratory of Agricultural Meteorology, Nanjing, China, (20)USDA, Agricultural Research Service, Jonesboro, AR, United States, (21)UFSM Federal University of Santa Maria, Santa Maria, Brazil, (22)Seoul National University, Seoul, Korea, Republic of (South), (23)Chung Hsing Univ, Taichung, Taiwan, (24)Department of Microbiology and Plant Biology, Center for Spatial Analysis, University of Oklahoma, Norman, United States, (25)Sun Yat-sen University, School of Atmospheric Sciences, Guangzhou, China, (26)Nanjing University, International Institute for Earth System Sciences, Nanjing, China
Abstract:
Rice cultivation is an important anthropogenic methane source to the atmosphere, contributing about 8% of total global anthropogenic emissions, and large uncertainties still exist in bottom-up estimates. The geographical distribution of rice emissions has been assessed at regional-to-global scales by bottom-up inventories and land surface models over coarse spatial units. However, high-resolution flux estimates capable of capturing local climatic and management, as well as replicate
in situ data remain challenging to produce. To fill this gap, we use rice methane flux data from 23 global eddy covariance sites (45 growing seasons) and geospatial datasets with machine learning to 1) evaluate data-driven model performance and predictor importance to predict rice CH
4 fluxes; 2) produce gridded up-scaling estimates of rice CH
4 emissions at high resolution (500-1000m) in Monsoon Asia, where >80% of global rice is cultivated.
Our random forest model is trained on CH4 and bioclimatic data at 8-day intervals. We used a leave-one-site-out procedure to reduce over-fitting. Our preliminary model selects 20 best predictors from an initial set of 210. Air temperature and land surface temperature at night are the most important predictor followed by MODIS indices that reflect greenness and surface water conditions (EVI, SRWI, and LSWI). The model reproduces much of the 8-day average flux variation, albeit with substantial error (R2 = 0.47, MAE = 63.8 nmol m-2 s-1, Bias = 1.6 nmol m-2 s-1). The model also predicts well the mean seasonal cycles (R2=0.54) and the site means (R2=0.52).
