基于激光数据的北极海水二氧化碳分压研究

Arctic sea surface CO_(2)partial pressure based on LiDAR

星载激光雷达作为一种新型的主动遥感技术,为全球海洋昼夜以及极地冬季海水二氧化碳分压的研究提供了可能。研究通过使用云-气溶胶激光雷达与红外探路者卫星(CALIPSO)数据,对全球海表面叶绿素a(Chla)浓度进行了反演,并构建了基于激光数据的前向神经网络模型(FNN-LID),重构了北冰洋海水二氧化碳分压(pCO_(2))昼/夜长时序数据集。在此基础上,对基于主动遥感的极地海水Chla浓度和海水pCO_(2)数据进行了验证与分析。结果显示,基于该算法的反演产品,具有较高的数据质量,不论是和其他被动遥感的产品还是独立浮标观测数据集都有较好的一致性,且能够有效“填充”极地冬季的数据空白。在北冰洋海区,受陆源强烈影响的边缘海都表现为较高的海表Chla浓度。北冰洋海水pCO_(2)的空间格局表现出经向的差异,且pCO_(2)的季节变化十分剧烈,甚至超过80μatm。近20年来,北冰洋稳定地表现为大气二氧化碳的汇,而在东西伯利亚海和喀拉海等海冰显著衰退的地区,海面pCO_(2)的增长率非常显著。

The spaceborne light detection and ranging(LiDAR),as a novel active remote sensing technology,offers possibilities for global diurnal research.In this study,global sea surface chlorophyll-a(Chla)concentrations were inverted using satellite data from Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observations(CALIPSO).A feedforward neural network model based on LiDAR data(FNN-LID)was developed to reconstruct a long-term diurnal dataset of sea surface pCO_(2)in the Arctic Ocean.Subsequently,verification and analysis were conducted on the polar sea surface Chla concentrations and sea surface pCO_(2)based on active remote sensing.The results demonstrated that the inversion products generated by this algorithm exhibit high data quality and exhibit favorable consistency with both other passive remote sensing products and buoy observations.Moreover,these products effectively fill data gaps during polar winters.Along the Arctic Ocean,margin seas significantly influenced by terrestrial sources consistently display high sea surface Chla concentrations.The spatial distribution of sea surface pCO_(2)in the Arctic Ocean manifests meridional variations,with marked seasonal fluctuations,even higher than 80μatm.Over the past two decades,the Arctic Ocean has consistently acted as a carbon dioxide sink,while areas with substantial sea ice decline such as the East Siberian Sea and Kara Sea exhibit pronounced increases in sea surface pCO_(2).