[M-IS01] Environmental, Socio-Economic and Climatic Changes in Northern Eurasia

[MIS01-06] Annual maps of surface water bodies and wetlands in Northeast China during 2015-2020

*Xiangming Xiao¹、Chenchen Zhang¹、Xinxin Wang²、Yuanwei Qin¹ (1.University of Oklahoma, Norman, Oklahoma 73019, USA、2.Fudan University, Shanghai, China)

キーワード：land use and land cover change, surface water bodies, paddy rice, wetlands, remote sensing

Annual maps of surface water bodies and wetlands in Northeast China are important for the studies of biodiversity, climate change, food security, and zoonotic infectious diseases. However, the current maps of surface water bodies and wetlands in Northeast China have moderate accuracy and large uncertainty. In this presentation, we report the results from our recent study that integrated optical images (e.g., Landsat, Sentinel-2), microwave images (Sentinel-1, ALOS PALSAR-2), and thermal images (MODIS). We used time series Landsat, Sentinel-2, and Sentinel-1 images to identify and generate annual maps of year-long, seasonal, and ephemeral surface water bodies. We used time series Landsat, Sentinel-2, and Sentinel-1 images to identify and generate annual maps of paddy rice, open-canopy and closed-canopy marshes. We use PALSAR-2, and time series Landsat and Sentinel-2 images to identify and generate annual maps of forest, evergreen forest, deciduous forest, and deciduous forest swamp. We assessed the resultant annual map of surface water bodies, paddy rice, and natural wetlands in 2020 at 10-m spatial resolution with in-situ reference data, and the accuracy assessment shows a high overall accuracy of 95%. According to the annual map in 2020, the Northeast China had total of 21,019 km² year-long surface water bodies, 83,365 km² paddy rice, 37,852 km² seasonal open-canopy marsh, 63,606 km² yearlong closed-canopy marsh, and 37,400 km² deciduous forest swamp. Our study has demonstrated the potential of integrated image datasets and knowledge-based algorithms for identifying and generating annual maps of surface water bodies and wetlands in Northeast China. The algorithms developed and evaluated over the Northeast China could be readily adapted to other regions in the Northeastern Asia.