Forest carbon sinks are crucial for mitigating urban climate change.Their effectiveness depends on the balance between gross carbon losses and gains.However,quantitative and continuous monitoring of forest change/dist...Forest carbon sinks are crucial for mitigating urban climate change.Their effectiveness depends on the balance between gross carbon losses and gains.However,quantitative and continuous monitoring of forest change/disturbance carbon fluxes is still insufficient.To address this gap,we integrated an improved spatial carbon bookkeeping(SBK)model with the continuous change detection and classification(CCDC)algorithm,long-term Landsat observations,and ground measurements to track carbon emissions,uptakes,and net changes from forest cover changes in the Yangtze River Delta(YRD)of China from 2000 to 2020.The SBK model was refined by incorporating heterogeneous carbon response functions.Our results reveal that carbon emissions(-3.88 Tg C·year^(-1))were four times greater than carbon uptakes(0.93 Tg C·year^(-1))from forest cover changes in the YRD during 2000-2020,despite a net forest cover gain of 10.95×10^(4) ha.These findings indicate that the carbon effect per hectare of forest cover loss is approximately 4.5 times that of forest cover gain.The asymmetric carbon effect suggests that forest cover change may act as a carbon source even with net-zero or net-positive forest cover change.Furthermore,carbon uptakes from forest gains in the YRD during 2000-2020 could only offset 0.28% of energy-related carbon emissions from 2000 to 2019.Urban and agricultural expansions accounted for 37% and 10% of carbon emissions,respectively,while the Grain for Green Project contributed to 45% of carbon uptakes.Our findings underscore the necessity of understanding the asymmetric carbon effects of forest cover loss and gain to accurately assess the capacity of forest carbon sinks.展开更多
精准快速识别小麦-玉米轮作区域对于中国北方地区耕地非粮化动态监测、主粮作物产能保障及农业可持续发展具有重要战略意义。该研究以河南省安阳市滑县为研究区,基于GEE云平台集成2018—2024年关键物候期Sentinel-2时序数据,构建光谱反...精准快速识别小麦-玉米轮作区域对于中国北方地区耕地非粮化动态监测、主粮作物产能保障及农业可持续发展具有重要战略意义。该研究以河南省安阳市滑县为研究区,基于GEE云平台集成2018—2024年关键物候期Sentinel-2时序数据,构建光谱反射率及植被指数时间序列多维特征集,分别使用传统单时相方法和改进的连续变化检测和分类(continuous change detection and classification,CCDC)算法对研究区域内主粮-主粮、主粮-非主粮、非主粮-主粮、非主粮-非主粮等4种轮作模式进行动态分类识别。结果表明:1)传统单时相方法在两个生长季的主粮作物分类总体精度(OA)最高可达96.8%、Kappa系数最高为0.96,两季影像叠加后的轮作模式识别平均OA和Kappa系数分别为71.3%、0.63;2)改进的CCDC-ANN算法对4种轮作模式识别的平均总体精度为91.8%、Kappa系数为0.891,较传统方法提升约20%;3)研究区种植结构呈现出明显的空间异质性,西部丘陵地区以主粮–非主粮轮作为主,东部平原以主粮–主粮、非主粮–主粮为主;4类轮作模式在2018—2024年均呈“先增后降再回升”动态:主粮-非主粮模式波动最剧烈,主粮-主粮模式最为平稳(波动<5%),非主粮-非主粮与非主粮-主粮模式亦表现出明显的阶段性涨落。该研究方法实现了小麦-玉米轮作区域的精准提取,为中国北方地区开展耕地非粮化监测提供了方法支撑。展开更多
基金supported by the Natural Science Foundation of Zhejiang Province(No.ZCLQN25C0301)the National Key Research and Development Program of China(No.2016YFC0502700)the General Program of Education Department of Zhejiang(No.23056209-F).
文摘Forest carbon sinks are crucial for mitigating urban climate change.Their effectiveness depends on the balance between gross carbon losses and gains.However,quantitative and continuous monitoring of forest change/disturbance carbon fluxes is still insufficient.To address this gap,we integrated an improved spatial carbon bookkeeping(SBK)model with the continuous change detection and classification(CCDC)algorithm,long-term Landsat observations,and ground measurements to track carbon emissions,uptakes,and net changes from forest cover changes in the Yangtze River Delta(YRD)of China from 2000 to 2020.The SBK model was refined by incorporating heterogeneous carbon response functions.Our results reveal that carbon emissions(-3.88 Tg C·year^(-1))were four times greater than carbon uptakes(0.93 Tg C·year^(-1))from forest cover changes in the YRD during 2000-2020,despite a net forest cover gain of 10.95×10^(4) ha.These findings indicate that the carbon effect per hectare of forest cover loss is approximately 4.5 times that of forest cover gain.The asymmetric carbon effect suggests that forest cover change may act as a carbon source even with net-zero or net-positive forest cover change.Furthermore,carbon uptakes from forest gains in the YRD during 2000-2020 could only offset 0.28% of energy-related carbon emissions from 2000 to 2019.Urban and agricultural expansions accounted for 37% and 10% of carbon emissions,respectively,while the Grain for Green Project contributed to 45% of carbon uptakes.Our findings underscore the necessity of understanding the asymmetric carbon effects of forest cover loss and gain to accurately assess the capacity of forest carbon sinks.
文摘精准快速识别小麦-玉米轮作区域对于中国北方地区耕地非粮化动态监测、主粮作物产能保障及农业可持续发展具有重要战略意义。该研究以河南省安阳市滑县为研究区,基于GEE云平台集成2018—2024年关键物候期Sentinel-2时序数据,构建光谱反射率及植被指数时间序列多维特征集,分别使用传统单时相方法和改进的连续变化检测和分类(continuous change detection and classification,CCDC)算法对研究区域内主粮-主粮、主粮-非主粮、非主粮-主粮、非主粮-非主粮等4种轮作模式进行动态分类识别。结果表明:1)传统单时相方法在两个生长季的主粮作物分类总体精度(OA)最高可达96.8%、Kappa系数最高为0.96,两季影像叠加后的轮作模式识别平均OA和Kappa系数分别为71.3%、0.63;2)改进的CCDC-ANN算法对4种轮作模式识别的平均总体精度为91.8%、Kappa系数为0.891,较传统方法提升约20%;3)研究区种植结构呈现出明显的空间异质性,西部丘陵地区以主粮–非主粮轮作为主,东部平原以主粮–主粮、非主粮–主粮为主;4类轮作模式在2018—2024年均呈“先增后降再回升”动态:主粮-非主粮模式波动最剧烈,主粮-主粮模式最为平稳(波动<5%),非主粮-非主粮与非主粮-主粮模式亦表现出明显的阶段性涨落。该研究方法实现了小麦-玉米轮作区域的精准提取,为中国北方地区开展耕地非粮化监测提供了方法支撑。
