摘要
内陆水体中浮游植物的存在对悬浮物(TSM)遥感反演模型精度具有一定的影响,藻类丰度会导致水体遥感反射率降低。实验基于中国、澳大利亚和美国内陆水体的372个采样点(4个数据集)水质分析和光谱实测数据,构建内陆水体遥感反射率与TSM的相关关系,建立最优波段比模型(OBR),并分析了藻类颗粒物存在对该模型精度的影响。由于水质的不均一性,不同区域的水质参数敏感波段存在差异,因此各数据集用于建模的最优波段比值不同。结果表明,OBR模型精度较高,误差较小,中国水体模型验证均具有较好效果(石头口门水库:R^(2)=0.87,RMSE=14.1 mg/L;查干湖:R^(2)=0.82,RMSE=23.6 mg/L),澳大利亚水体模型验证效果最佳,R^(2)值高达0.95(RMSE=4.2 mg/L),美国水体模型精度较低(R^(2)=0.78,RMSE=3.7 mg/L)。研究发现,模型精度受水体叶绿素(Chla)浓度和Chla/TSM比率影响,当水体以TSM浓度较高的非藻类颗粒物为主时(如中国石头口门水库和南澳洲地区水体数据集),最优波段比值模型表现更好;而当水体以浮游植物为主时,水体中的浮游植物的丰度会使光谱信号复杂化,从而限制或降低TSM浓度遥感算法的精度(如美国印第安纳州中部水库数据集)。
The presence of inland phytoplankton in inland water has a certain impact on the accuracy of Total Solid Matter(TSM)inversed remote sensing model,and the abundance of algae will influence the decreasing of remote sensing reflectance of water.Based on in situ dataset of water quality and spectrum of 372 sampling points across China,Australia and the United States,the relationship between the reflectance of water body and TSM is established and the Optimal Band Ratio(OBR)model is applied to analyze the influence of algae particles on the accuracy of the model.Due to the unbalance of water quality parameters in various regions,the sensitive bands of water quality parameters varies in different regions,therefore the optimal band ratio used for modeling in each dataset is different.The results show that the high accuracy of OBR model is observed with low errors.The model validation for two water bodies in China have a good performance(STKM:R^(2)=0.87,RMSE=14.1;CGH:R^(2)=0.82,RMSE=23.6).The validation of Australian waterbody achieves high coefficient of0.95(R^(2)=0.95,RMSE=4.2),while the model applied to American water body has a lower accuracy(R^(2)=0.78,RMSE=3.7).It is found that the model accuracy is affected by Chla concentration and Chla/TSM ratio.When the water is dominated by non-algal particles with high TSM concentration(such as water datasets from Shitoukoumen,China and South Australia),the model of optimal band ratio performs better.However,when the water body is dominated by phytoplankton,the water body will complicate the spectral signal,thus limiting or reducing the accuracy of remote sensing algorithm(such as the water body data set in central Indiana,USA).
作者
赵方睿
王强
温志丹
刘晓静
尚盈辛
陶慧
杜云霞
宋开山
Zhao Fangrui;Wang Qiang;Wen Zhidan;Liu Xiaojing;Shang Yingxin;Tao Hui;Du Yunxia;Song Kaishan(College of Tourism and Geography,Jilin Normal University,Siping 136000,China;Northeast Institute of Geography and Agroecology,Chinese Academy of Sciences,Changchun 130102,China)
出处
《遥感技术与应用》
CSCD
北大核心
2022年第4期993-1002,1011,共11页
Remote Sensing Technology and Application
基金
国家自然科学基金重点项目(41730104)
吉林省自然科学基金学科布局项目(20200201054JC)
关键词
遥感反演
模型构建
悬浮物
叶绿素
最优波段比值模型
Remote sensing inversion
Model construction
Total suspended solids
Chlorophyll-a
OBR
