摘要
火烧迹地是全球变化和碳循环等研究领域的一个重要参量。在全球高精度样本库基础上,基于Landsat 8时序卫星数据和火烧迹地敏感光谱参量,利用机器学习算法生产并发布全球30米分辨率火烧迹地产品。全球30米分辨率火烧迹地产品能够有效反映面积较小的燃烧斑块,同时在火烧迹地位置确定和面积量算上具有优势,可应用于全球火灾监测和评估、碳排放计算、生态环境保护等领域。
Burned area(BA)is an important research parameter in the field of global change and carbon cycle.On the basis of high-precision global sample database,we input Landsat 8 time series satellite data and several sensitive spectral parameters of burned areas to the machine learning algorithm,produced and released the 30-meter resolution global burned area products.The 30-meter resolution global BA products can effectively detect small burned patches,and excel in location spotting and area measurement of the burned patches,which can be applied to global fire monitoring and disaster assessment,carbon emission calculation,ecological environment protection and other fields.
作者
张兆明
唐朝
何国金
龙腾飞
魏明月
Zhang Zhaoming;Tang Chao;He Guojin;Long Tengfei;Wei Mingyue(Aerospace Information Research Institute,Chinese Academy of Sciences,Beijing 100094,P.R.China;School of Electronic,Electrical and Communication Engineering,University of Chinese Academy of Sciences,Beijing 100049,P.R.China;Hainan Key Laboratory for Earth Observation,Sanya 572029,P.R.China)
基金
国家自然科学基金(61731022)
国家重点研发计划课题(2016YFA0600302)
中国科学院A类战略性先导科技专项(XDA19090300)。
