|
| 无人机高光谱遥感技术在太浦河水质监测中的应用初探 |
| Preliminary Application of UAV Hyperspectral Remote Sensing Technology in Water Quality Monitoring of the Taipu River |
| |
| DOI: |
| 中文关键词: 无人机 高光谱遥感 反演模型 水质监测 太浦河 |
| 英文关键词: UAV Hyperspectral 〖GK!87mm〗remote sensing Inversion model Water quality monitoring Taipu River |
| 基金项目:国家自然科学青年基金资助项目(32001162);上海市科委科研基金资助项目(22dz1202605) |
|
| 摘要点击次数: 168 |
| 全文下载次数: 54 |
| 中文摘要: |
| 以太浦河为研究对象,利用无人机高光谱遥感影像结合监测船在线监测数据和地面实测数据,构建空地一体化水质监测方法,并利用卷积神经网络(CNN)和长短期记忆网络(LSTM)构建CNN-LSTM模型,对叶绿素a、总氮、总磷和氨氮进行遥感反演。结果表明:叶绿素a、总氮、总磷和氨氮反演精度分别为8828%、9098%、8753%和7884%;2024年3月太浦河水质整体较好,局部区域叶绿素a、总氮和总磷测定值偏高,存在一定富营养化风险,氨氮值整体较低;各水质指标总体呈现上游高于下游、南岸高于北岸的空间分布特征,以Ⅱ类和Ⅲ类水质为主,部分区域为Ⅰ类水质。 |
| 英文摘要: |
| Taking the Taipu River as the study area, an air ground integrated water quality monitoring approach was developed by combining unmanned aerial vehicle hyperspectral imagery with online monitoring data acquired from a monitoring vessel and in situ measurements. A hybrid convolutional neural network long short term memory(CNN LSTM) model was established to retrieve chlorophyll a(Chl a), total nitrogen(TN), total phosphorus(TP), and ammonia nitrogen(NH3 N). The results showed that the retrieval accuracies of Chl a, TN, TP, and NH3 N reached 8828%, 9098%, 8753%, and 7884%, respectively. In March 2024, the overall water quality of the Taipu River was relatively good. Elevated concentrations of Chl a, TN, and TP were observed in some local areas, indicating a potential risk of eutrophication, whereas NH3 N remained at relatively low levels. The spatial distribution of all water quality parameters exhibited higher values in the upstream reaches than in the downstream reaches and on the southern bank than on the northern bank. Water quality was dominated by Class Ⅱ and Class Ⅲ, with some areas reaching Class Ⅰ. |
|
查看全文
查看/发表评论 下载PDF阅读器 |
| 关闭 |