【摘要】 有效开展湖泊富营养化监测是准确掌握湖泊生态动态、严格控制湖泊环境污染的重要手段。判别湖泊营养等级、定量湖泊营养状态对湖泊富营养化遥感监测具有重要意义。针对中国湖泊空间异质性高、评价标准难以统一的问题，本研究聚焦营养状态指数TSI（Trophic State Index），结合多类型湖泊全覆盖星地同步数据，全面评估了3种面向全国湖泊开发的营养状态遥感估算方法：水色指数算法FUI（Forel-Ule Index）、吸收系数算法a_tw（Total Non-water Absorption）与藻类生物量算法ABI（Algal Biomass Index）在高浑浊富营养浅水湖泊与贫营养内陆深水湖泊的监测精度，定量了算法在宽波段卫星与多光谱卫星的性能差异。结果表明，FUI算法对富营养水体具有很好的识别精度，对传感器波段设置要求较低（可见光波段），仅能提供定性判断；a_tw算法对中营养水体具有很好的识别精度，同时需要较高的传感器波段设置和精确的大气校正；ABI算法提供了可靠的富营养化评估，对不同等级的卫星遥感产品数据具备较好的稳定性，但依赖于短波红外以实现水体粗分类（浑浊、清洁水体）。ABI算法在Landsat 8与Sentinel-3的总体精度分别为71%与73%，优于a_tw（56%与63%）和FUI（49%与52%）。本研究可推动区域湖泊营养状态监测数据共享，为规范湖泊富营养化遥感监测手段、实现跨境湖泊环境管理提供重要参考价值。更多还原

【Abstract】 Lakes are an important part of terrestrial ecosystems. The eutrophication of lakes has become a major ecological and environmental problem in China and even globally, which is one of the key causes of lake ecosystem degradation. Effective monitoring of lake eutrophication provides an important approach to accurately grasping the ecological dynamics of lakes and strictly controlling environmental pollution in lakes. It is of great significance to determine the trophic level and quantify the trophic state of lakes for remote sensing monitoring of lake eutrophication.In this study, we focused on the Trophic State Index（TSI）, and evaluated the performance and reliability of three remote sensing methods for trophic state estimation developed for national lakes based on simultaneous field data: Forel-Ule index（FUI）, total non-water absorption(a_tw) and Algal Biomass Index（ABI） on Landsat-8 and Sentinel-3 satellites. The fitting coefficients of the empirical models were corrected based on field data, and the accuracy before and after the algorithm corrections were compared to assess stability. The effects of the surface reflectance products and the remote sensing reflectance corrected by the atmospheric correction algorithms on the TSI inversion were further compared to examine the sensitivity of these algorithms.Our results show that the FUI algorithm has good identification accuracy for eutrophic water bodies with lower requirements for sensor band settings（visible bands）. However, it only provides qualitative judgement. The a_tw algorithm has good identification accuracy for mesotrophic lakes with higher requirements for sensor band settings and atmospheric corrections. The ABI algorithm provides a reliable performance of eutrophication, with good stability of satellite remote sensing data at different levels. The ABI algorithm can figure out different degrees of eutrophic waters, but relies on shortwave infrared for optical classification of water bodies. The ABI algorithm shows the best accuracy with an overall accuracy of 71% and 73% for Landsat-8 and Sentinel-3, respectively, which is better than a_tw（56% and 63%） and FUI（49% and 52%） algorithm.To sum up, the FUI algorithm is preferred for the qualitative investigation of lake eutrophication to quickly and roughly grasp the trophic level, and the ABI or a_tw algorithm is further applied for precise quantification. It is recommended to choose one remote sensing evaluation method according to different application scenarios, which can improve the consistency and comparability of lake trophic state assessment. This study can provide an important basis for the future specification of remote sensing monitoring of lake eutrophication.更多还原