High Mountain Asia(HMA),recognized as a third pole,needs regular and intense studies as it is susceptible to climate change.An accurate and high-resolution Digital Elevation Model(DEM)for this region enables us to ana...High Mountain Asia(HMA),recognized as a third pole,needs regular and intense studies as it is susceptible to climate change.An accurate and high-resolution Digital Elevation Model(DEM)for this region enables us to analyze it in a 3D environment and understand its intricate role as the Water Tower of Asia.The science teams of NASA realized an 8-m DEM using satellite stereo imagery for HMA,termed HMA 8-m DEM.In this research,we assessed the vertical accuracy of HMA 8-m DEM using reference elevations from ICESat-2 geolocated photons at three test sites of varied topography and land covers.Inferences were made from statistical quantifiers and elevation profiles.For the world’s highest mountain,Mount Everest,and its surroundings,Root Mean Squared Error(RMSE)and Mean Absolute Error(MAE)resulted in 1.94 m and 1.66 m,respectively;however,a uniform positive bias observed in the elevation profiles indicates the seasonal snow cover change will dent the accurate estimation of the elevation in this sort of test sites.The second test site containing gentle slopes with forest patches has exhibited the Digital Surface Model(DSM)features with RMSE and MAE of 0.58 m and 0.52 m,respectively.The third test site,situated in the Zanda County of the Qinghai-Tibet,is a relatively flat terrain bed,mostly bare earth with sudden river cuts,and has minimal errors with RMSE and MAE of 0.32 m and 0.29 m,respectively,and with a negligible bias.Additionally,in one more test site,the feasibility of detecting the glacial lakes was tested,which resulted in exhibiting a flat surface over the surface of the lakes,indicating the potential of HMA 8-m DEM for deriving the hydrological parameters.The results accrued in this investigation confirm that the HMA 8-m DEM has the best vertical accuracy and should be of high use for analyzing natural hazards and monitoring glacier surfaces.展开更多
ICESat-2(Ice,Cloud and Land Elevation Satellite-2)是世界首颗采用光子计数模式的激光测高卫星,可快速获得高精度、大尺度地面三维数据。光子探测机制使得数据中除了地面信号外,还包含大气散射等背景信号,需要通过滤波才能获得地形...ICESat-2(Ice,Cloud and Land Elevation Satellite-2)是世界首颗采用光子计数模式的激光测高卫星,可快速获得高精度、大尺度地面三维数据。光子探测机制使得数据中除了地面信号外,还包含大气散射等背景信号,需要通过滤波才能获得地形等信息。为分析ICESat-2背景和信号光子的分布特点及点云滤波算法的效果和适用性,本文首先选取了六种地表覆盖类型(城市、海冰、沙漠、植被、海洋及冰盖/冰川)及不同观测条件的数据,对其背景光子率进行统计分析。分析结果表明:白天观测数据的背景光子率平均为106(点/秒)数量级,远高于夜晚观测数据的背景光子率——104(点/秒)数量级,弱波束的背景光子率与强波束背景光子率相当,六种地表覆盖类型中,冰盖/冰川的背景光子率最高。然后,根据统计结果筛选出21组测高数据,并选取七种具有代表性的点云滤波对其进行去噪实验,分析精度后得出结论:改进局部密度法的去噪效果最佳,算法召回率、精准度和F值均大于0.90,算法较为稳定。最后,对所选取各滤波算法的精度、特点与适用性等性质进行了总结与分析,可为后续该数据的使用和滤波算法的选择提供参考。展开更多
基金The authors gratefully acknowledge the science teams of NASA High Mountain Asia 8-meter DEM and NASA ICESat-2 for providing access to the data.This work was conducted with the infrastructure provided by the National Remote Sensing Centre(NRSC),for which the authors were indebted to the Director,NRSC,Hyderabad.We acknowledge the continued support and scientific insights from Mr.Rakesh Fararoda,Mr.Sagar S Salunkhe,Mr.Hansraj Meena,Mr.Ashish K.Jain and other staff members of Regional Remote Sensing Centre-West,NRSC/ISRO,Jodhpur.The authors want to acknowledge Dr.Kamal Pandey,Scientist,IIRS,Dehradun,for sharing field-level information about the Auli-Joshimath.This research did not receive any specific grant from funding agencies in the public,commercial,or not-for-profit sectors.
文摘High Mountain Asia(HMA),recognized as a third pole,needs regular and intense studies as it is susceptible to climate change.An accurate and high-resolution Digital Elevation Model(DEM)for this region enables us to analyze it in a 3D environment and understand its intricate role as the Water Tower of Asia.The science teams of NASA realized an 8-m DEM using satellite stereo imagery for HMA,termed HMA 8-m DEM.In this research,we assessed the vertical accuracy of HMA 8-m DEM using reference elevations from ICESat-2 geolocated photons at three test sites of varied topography and land covers.Inferences were made from statistical quantifiers and elevation profiles.For the world’s highest mountain,Mount Everest,and its surroundings,Root Mean Squared Error(RMSE)and Mean Absolute Error(MAE)resulted in 1.94 m and 1.66 m,respectively;however,a uniform positive bias observed in the elevation profiles indicates the seasonal snow cover change will dent the accurate estimation of the elevation in this sort of test sites.The second test site containing gentle slopes with forest patches has exhibited the Digital Surface Model(DSM)features with RMSE and MAE of 0.58 m and 0.52 m,respectively.The third test site,situated in the Zanda County of the Qinghai-Tibet,is a relatively flat terrain bed,mostly bare earth with sudden river cuts,and has minimal errors with RMSE and MAE of 0.32 m and 0.29 m,respectively,and with a negligible bias.Additionally,in one more test site,the feasibility of detecting the glacial lakes was tested,which resulted in exhibiting a flat surface over the surface of the lakes,indicating the potential of HMA 8-m DEM for deriving the hydrological parameters.The results accrued in this investigation confirm that the HMA 8-m DEM has the best vertical accuracy and should be of high use for analyzing natural hazards and monitoring glacier surfaces.
文摘ICESat-2(Ice,Cloud and Land Elevation Satellite-2)是世界首颗采用光子计数模式的激光测高卫星,可快速获得高精度、大尺度地面三维数据。光子探测机制使得数据中除了地面信号外,还包含大气散射等背景信号,需要通过滤波才能获得地形等信息。为分析ICESat-2背景和信号光子的分布特点及点云滤波算法的效果和适用性,本文首先选取了六种地表覆盖类型(城市、海冰、沙漠、植被、海洋及冰盖/冰川)及不同观测条件的数据,对其背景光子率进行统计分析。分析结果表明:白天观测数据的背景光子率平均为106(点/秒)数量级,远高于夜晚观测数据的背景光子率——104(点/秒)数量级,弱波束的背景光子率与强波束背景光子率相当,六种地表覆盖类型中,冰盖/冰川的背景光子率最高。然后,根据统计结果筛选出21组测高数据,并选取七种具有代表性的点云滤波对其进行去噪实验,分析精度后得出结论:改进局部密度法的去噪效果最佳,算法召回率、精准度和F值均大于0.90,算法较为稳定。最后,对所选取各滤波算法的精度、特点与适用性等性质进行了总结与分析,可为后续该数据的使用和滤波算法的选择提供参考。
