Quantification of lake clarity in China using Landsat OLI imagery data
Quantification of lake clarity in China using Landsat OLI imagery data
复制标题
使用 Landsat OLI 图像数据量化中国湖泊透明度
DOI：
10.1016/j.rse.2020.111800
复制
发表时间：
2020-06
影响因子：
13.5
通讯作者：
Chong Fang
中科院分区：
文献类型：
--
作者：
Kaishan Song;Ge Liu;Qiang Wang;Zhidan Wen;Lili Lyu;Yunxia Du;Linwei Sha;Chong Fang
湖泊、水库（以下简称湖泊）的富营养化问题已引起公众和政府的关注。水体透明度与叶绿素a、总悬浮物和营养盐密切相关，是衡量水体富营养化状况的可靠指标。传统上，水的透明度是用塞奇圆盘深度（SD）来测量的。通过将来自水面的光谱信号与现场测量的SD相联系，遥感提供了一种有用的工具，以重复的方式在大尺度上估计SD。在许多地区，通过建立不同卫星立交桥的特定模型，并结合现场测量的SD，报告了遥感获得的水体透明度，但在中国，全国水体透明度仍然未知。在这项研究中，从2013-2018年的34次实地活动中收集了2152个样本，其中1792个样本是在Landsat OLI立交桥± 7天内收集的。我们使用Landsat 8 OLI波段1-4来开发回归模型（n= 1016），并使用782个样本来验证模型性能。我们进一步收集了三个额外的原位SD数据集来验证最佳性能模型，并最终使用它来绘制全国范围的SD，主要是在2016年获得的OLI图像。我们的研究结果表明，整个数据集的SD与Landsat反射率有很强的关联，在中国湖泊的测量和估计SD（RMSE = 63 cm）之间的均方根误差较低。2016年全国水体平均净度为176 cm，空间变异性较大（S.D：216 cm），东部平原地区沃茨与青藏高原地区清水差异显著。中国东北部（75厘米）和东部（84厘米）的湖泊由于水深浅，悬浮物和藻类丰度高，透明度低。云贵高原（91厘米）、内蒙古和新疆自治区（114厘米）的湖泊透明度居中;而西藏高原（294厘米）的湖泊透明度最高。这项调查表明，与瑞利散射校正反射率结合在现场观测的陆地卫星图像可以提供定量信息的湖泊清晰度与表面积>8公顷。此外，这种方法有可能检索SD与归档Landsat图像，以确定SD的时间变化在国家或大陆尺度，可用于支持内陆水管理和决策者改善水质。
The eutrophication of lake and reservoir (hereafter referred to as lakes) has attracted concerns from the public and government in China. Water clarity is a reliable indicator for quantifying eutrophic status because of its strong association with chlorophyll-a, total suspended matter, and nutrients. Traditionally, water clarity is measured using Secchi disk depth (SD). By linking the spectral signal from water surface with in situ measured SD, remote sensing provides a useful tool for SD estimation at a large scale in a repetitive manner. Remote sensing derived water clarity has been reported in many regions with specific models established for different satellite overpasses concurrent with in situ measured SD, but national water clarity remained unknown in China. In this study, 2152 samples were collected from 34 field campaigns in 2013–2018, of which 1792 samples were gathered within ± 7 days of Landsat OLI overpasses. We used Landsat 8 OLI bands 1–4 to develop regression models (n= 1016), and 782 samples to validate model performances. We further collected three additional in situ SD datasets to validate the best performance model, and eventually used it to map SD at a national scale with OLI images mainly acquired in 2016. Our results indicated that the entire dataset of SD has a strong association with Landsat reflectance, yielding low root mean square error between measured and estimated SD (RMSE = 63 cm) for lakes in China. The national water clarity was averaged to 176 cm in 2016 with large spatial variability (S.D: 216 cm) due to the marked variation between turbid waters in the east plain area and clean water across the Tibet Plateau. Lakes in the northeastern (75 cm) and eastern (84 cm) China had low clarity due to shallow water depth combined with high suspended matter and algal abundance. Lakes in the Yungui Plateau (91 cm), Inner Mongolia and Xinjiang autonomous regions (114 cm) exhibited intermediate clarity; while lakes in the Tibet Plateau (294 cm) displayed the highest clarity. This investigation demonstrated that Landsat imagery with Rayleigh scattering correction reflectance combined with in situ observation can provide quantitative information about the lake clarity with surface area >8 ha. Moreover, this method has the potential to retrieve SD with archived Landsat imagery to determine the temporal variation of SD at national or continental scale which can be used to support inland water management and decision-maker for improving water quality.
登录
查看更多内容
影响因子：
13.5
作者：
K. Song;Lin Li;L. Tedesco;Shuai Li;H. Duan;Dianwei Liu;Bob Hall;Jia Du;Zuchuan Li;Kun Shi
通讯作者：
K. Song;Lin Li;L. Tedesco;Shuai Li;H. Duan;Dianwei Liu;Bob Hall;Jia Du;Zuchuan Li;Kun Shi
影响因子：
13.5
作者：
Doron, Maeva;Babin, Marcel;Garnesson, Philippe
通讯作者：
Garnesson, Philippe
影响因子：
4.5
作者：
Ruddick, KG;De Cauwer, V;Moore, G
通讯作者：
Moore, G
影响因子：
4.4
作者：
Gordon, HR
通讯作者：
Gordon, HR
影响因子：
9.8
作者：
Wang, Jing-Han;Yang, Cheng;Hu, Hong-Ying
通讯作者：
Hu, Hong-Ying