基于Landsat8遥感影像的青海湖BOD含量反演模型研究

来源:用户上传      作者:张怡然 牛海林 冯昶瑞 谢朝阳 宋玉庭

　　摘要为了解青海湖BOD含量反演的最适模型，根据Landsat8数据特点及反演技术要求，利用波段比值法对青海湖BOD含量进行反演模型研究。首先利用经过预处理的Landsat8影像中7个波段分别与BOD含量建立反演关系，选择拟合度较高的B5、B6、B7波段，按照B5+B6、B5+B7、B6+B7、B5+B6+B7、B5×B6、B5×B7、B6×B7、B5×B6×B7的波段组合策略进行波段组合，对各个组合波段与BOD含量分别建立指数函数、线性函数、幂函数和对数函数4种类型的反演模型，结果表明，拟合度最高的模型是y=0.743 3e0.000 2x（R2=0.636 7），最后利用该模型对青海湖8个点BOD含量实测值进行验证时发现反演得到的预测值与实测值的决定系数（R2）为0.574 1，且8个验证点中实测值与预测值相对误差小于30%的验证点有6个，故认为利用该模型进行青海湖BOD含量反演是可行的。
　　关键词Landsat8遥感影像;BOD含量;反演模型;青海湖流域
　　中图分类号X171.1文献标识码A文章编号0517-6611（2022）15-0055-04
　　doi：10.3969/j.issn.0517-6611.2022.15.015
　　开放科学（资源服务）标识码（OSID）：
　　Research on the Inversion Model of BOD Content in Qinghai Lake Based on Landsat8 Remote Sensing Images
　　ZHANG Yi-ran，NIU Hai-lin，FENG Chang-rui et al
　　（College of Eco-environmental Engineering of Qinghai University，Xining，Qinghai 810016）
　　AbstractIn order to understand the optimal model for the inversion of the BOD content in Qinghai Lake，this paper used the band ratio method to study the inversion model of the BOD content in Qinghai Lake according to the characteristics of Landsat8 data and the technical requirements of inversion.First，the seven bands in the preprocessed Landsat8 image were used to establish an inversion relationship with the respective BOD contents，and select the three bands with higher fitting degree：B5，B6 and B7.The band combination strategy of B5+B6，B5+B7，B6+B7，B5×B6，B5×B7，B6×B7，B5×B6×B7 performed band combination，and established exponential function，linear function，power function and logarithm for each combined band and BOD content.There were four types of inversion models for the function，and the results showed that the model with the highest fitting degree was y=0.743 3e0.000 2x （R2=0.636 7）.Finally，the model was used to verify the measured BOD content of 8 points in Qinghai Lake.It was found that the coefficient of determination between the predicted value obtained by inversion and the measured value was R2=0.574 1，and the relative error between the measured value and the predicted value in the 8 verification points was less than 30%.There were 6 verification points，so it was feasible to use this model to invert the BOD content of Qinghai Lake.
　　Key wordsLandsat8 remote sensing image;BOD content;Inversion model;Qinghai Lake Basin
　　湖泊在人繁衍生息过程中发挥着极其重要的作用。随着人类活动加剧及经济社会发展，全球范围内湖泊水质受到极大威胁。面对日益严峻的生态环境问题，需要进一步去探索更加高效且精准的监测方法。传统水质监测方法局限于实验室分析，尽管该方法精度高，但因其工作量大、耗时长等弊端，难以实现对湖泊水质的连续监测及早期预警。而遥感技术则利用不同传感器进行电磁波信息的收集及处理，通过收集到的遥感影像进行水质监测。较于传统监测手段，遥感监测在应用过程中体现出成本低、信息量大、高效、数据的强实时性、广覆盖面及可同步显示地物特征等优越性，可在一定程度上弥补传统监测方法的缺陷。基于以上特点，遥感技术在生态环境监测中的应用越来越广泛[1-3]。

nlc202208171502

转载注明来源:https://www.xzbu.com/1/view-15437804.htm