考虑垂向结构的富营养湖泊颗粒有机碳储量遥感

Particulate organic carbon (POC) stock can reflect the magnitude level of biological carbon in lake water and the ability of lake to absorb atmospheric CO2, which plays an important role in the global carbon cycle and has attracted wide attention. At the same time, lake eutrophication leads to algal bloom outbreak and the production of a large deal of POC by absorbing CO2 through photosynthesis, which has important impacts on the carbon cycle of water bodies. Therefore, it is urgent to estimate POC stock in eutrophic lake. Remote sensing has the advantages of large coverage, periodicity and fast recording, and has become one of the most effective tools for monitoring POC. At present, studies on POC in eutrophic lakes mainly focused on surface concentrations. Due to uneven vertical distribution, however, surface POC concentration cannot truly indicate POC stock. By taking the eutrophic Lake Chaohu as the study area and using field observation and other experiment data, this project intends to first clarify vertical structure types and influencing factors for POC in different seasons and weather conditions, and then develop binary decision tree to identify types and empirical equations to parameterize used models which are available to satellite data. Finally, all developed models will be applied to satellite data to estimate long-term POC stock. This project aims at the remote sensing of POC stock in eutrophic lake and expands the remote sensing monitoring from two-dimensional surface to three-dimensional space, which has important scientific significance and research value.

湖泊颗粒有机碳（POC）储量能够反映水体生物碳的量级水平和吸收大气CO2的能力，在全球碳循环中扮演重要角色，引起广泛关注。同时，湖泊富营养化导致藻类大量暴发，并通过光合作用吸收CO2合成大量POC，对水体碳循环产生重要影响。因此，亟需对富营养湖泊POC储量进行估算。遥感技术具有大范围、周期性和快速记录优势，已成为POC监测最有效的手段之一。遥感研究目前主要侧重于表层POC浓度，但由于蓝藻上浮等造成垂向分布不均，使其并不能真实代表POC储量。本项目拟选择富营养巢湖作为研究区，通过野外观测等试验，尝试：理清不同季节和天气条件下POC垂向结构类型和影响因素；确定POC垂向分布概念模型，并研究像元尺度模型参数的定量化表达；结合表层POC遥感模型，实现长时间序列POC储量估算。本项目瞄准富营养湖泊POC储量遥感研究，将遥感监测从二维表层拓展至三维空间，具有重要的科学意义和研究价值。

颗粒有机碳（POC）储量是决定浅水富营养化湖泊碳汇和水质的重要因素。虽然已有研究利用卫星数据观测了表层水体POC含量，但对水柱POC不均匀分布的浅水富营养化湖泊，遥感的表层POC含量无法反映水体POC总储量。本研究基于我国江淮平原19个浅水富营养化湖泊的272个原位观测POC剖面，开发了一种基于OLCI/Sentinel-3A卫星数据的水柱POC储量遥感方法。该方法包含3个关键步骤：遥感估算表层POC含量，通过二叉决策树识别POC剖面类型（均匀、指数衰减或幂函数衰减），以及使用表层POC浓度实现POC剖面参数化。遥感结果显示：所构建的算法对15个湖泊的观测值匹配（N = 88）的偏差为-25.79%；富营养化湖泊POC主要由浮游植物产生(R2 = 0.87)，表层POC含量在夏季呈现单峰特征；除气温调节作用（0.52 ~ 68.53%）外，水位对POC储量的影响也很明显，对六大湖泊的贡献为29.29 ~ 95.51%，未来需要多源卫星数据来遥感反演全球富营养化湖泊的POC储量。本研究首次尝试从空间上三维观测湖泊POC含量，对理解浅水富营养化湖泊的碳循环具有重要意义。.除了富营养湖泊POC储量遥感，研究还构建了富营养湖泊溶解有机碳（DOC）和藻总量遥感算法，利用卫星数据监测了我国大型湖泊水体透明度长时序变化，并深入剖析了中国河流有机碳输运的时空变异特征及影响机制。结果表明：（1）富营养太湖DOC和藻总量呈明显的“夏高冬低”季节变异特征，这与湖泊藻华暴发息息相关，而且藻总量可用于短期藻华预测；（2）中国湖泊透明度整体上表现“西高东低”的空间格局，且在2000-2020年期间有70.15%的湖泊水体变清；（3）受人为活动调控，我国河流有机碳输运正逐渐失去其低DOC/POC比的特征，且强降雨和新冠封锁等短期事件会明显改变河流有机碳输运。研究结果对动态监测湖泊水环境和理清湖泊碳收支具有重要应用价值。