阿拉斯加北极滨海平原极地湖泊的水深遥感反演

Polar Lake Bathymetry Retrieval from Remote Sensing Data of the Arctic Coastal Plain in Alaska

北极滨海平原地区的湖泊水体深度对理解极地地表过程和环境变化具有重要意义。利用遥感数据反演极地湖泊水体深度具有快速和大范围的优势,但目前在极地湖泊中的研究较少。以遥感探测阿拉斯加极地湖泊深度为研究目的,实施并验证了Stumpf水深反演模型对于极地湖泊深度遥感反演的工作方法,且对该地区16380 km2范围内分布的3 187个湖泊进行了同期深度反演,并利用船载声纳仪实测湖泊水体的深度数据对反演结果进行精度验证。结果显示水深反演结果的平均绝对误差和均方根误差分别为0.37与0.54 m,表明利用遥感技术适用于反演极地湖泊的水体深度。实验结果显示阿拉斯加北部滨海平原极地湖泊平均深度具有沿山麓向滨海平原逐渐下降的趋势。

Bathymetry of lakes in the Arctic coastal plain plays an important role in understanding the regional geographical processes and environmental changes. Inversion of polar lake depth with remote sensing technique owns the advantage of rapidly processing large-scale data, but so far less research was done in Arctic. The aim of this article is to retrieve the bathymetry of shallow thaw lakes in the Arctic Coastal Plain of northern Alaska based on the multispectral remote sensing data. The inversion method proposed by Stumpf was inspected and verified. Then it was applied to retrieve the depth of 3187 lakes distributed in the area of 16 380 km^2. Validation for the model was implemented via field measurements obtained from ship-borne sonar, and the MAE （mean absolute error） and RMSE （root mean square error） were 0.37 m and 0. 54 m respectively. This method is suitable for bathymetrie retrieval from remote sensing data of thousands of lakes in the study area. The spatial pattern of lake depth shows that depth of lake decreases from hill foot of Brooks to the coastline.