基于MODIS数据的渤海海冰厚度反演算法优化

Optimization of the Bohai Sea ice thickness retrieval algorithm based on MODIS data

海冰厚度是监测与研究渤海海冰的重要参数。为了获取更加可靠的渤海海冰厚度数据,本研究基于MODIS数据对海冰厚度反演中的冰水分离环节和冰厚计算方法都进行了改进。对于冰水分离环节,本文在Canny边缘检测算子提取海冰基础上,加入了二值化处理、阈值判别等步骤,实现了较高精度的渤海海冰范围自动化提取。通过试验确定了海冰厚度与反照率指数关系模型中的参数,包括海冰衰减系数和海水反照率参数,使其更加符合渤海海区的物理特征。将改进后算法的海冰厚度反演结果与渤海海上石油平台实测数据进行比较,并分析了误差来源。结果表明,经过对算法的改进,海冰厚度与反照率指数关系模型的反演结果与实测数据之间的平均绝对误差由7.05 cm缩小到2.74 cm,相关系数由0.434提高到0.485。

Sea ice thickness is a crucial parameter for monitoring and studying sea ice in the Bohai Sea.Aiming to get more reliable data conveniently,we improved the ice thickness retrieval algorithm based of MODIS data,including the ice separation process and ice thickness calculation method.In terms of ice-water separation process,some steps like binary processing,threshold discrimination were added based on sea ice extracting with Canny edge detector,which successfully realized the automatic high-precision extraction of sea ice range in the Bohai Sea.Meanwhile,through experiments,we optimize the parameters of the exponential model between sea ice thickness and albedo,including sea ice attenuation coefficient and sea water albedo parameters,to make it more consistent with the physical characteristics of the Bohai Sea area.The sea ice thickness retrieval results of the improved algorithm are compared with the measured data of the Bohai offshore oil platform,and the error reasons are analyzed.The results show that the average absolute error decreases from 7.05 cm to 2.74 cm,and the correlation coefficient increases from 0.434 to 0.485.