Quantify water quality variation of urban river from hyperspectral images through ripple propagation network with spatially inconsecutive sampling

IF 5 2区地球科学 Q1 WATER RESOURCES

Journal of Hydrology-Regional Studies Pub Date : 2025-09-23 DOI:10.1016/j.ejrh.2025.102765

Yishan Zhang , Lun Wu

{"title":"Quantify water quality variation of urban river from hyperspectral images through ripple propagation network with spatially inconsecutive sampling","authors":"Yishan Zhang , Lun Wu","doi":"10.1016/j.ejrh.2025.102765","DOIUrl":null,"url":null,"abstract":"<div><div>Study region</div><div>Huangyang River, Zhuhai city, Guangdong province, China</div><div>Study focus</div><div>High-quality water environment is indispensable to industrial and agricultural advancement. Present methods predicted concentrations of various water quality parameters (WQPs) using their individual models, which suffered calculation instability on various WQPs and needed spatially successive water sampling covering the entire monitoring area to keep calculation accuracy. This study proposed a feature interaction ripple network (FIRN) to retrieve concentrations of WQPs including total phosphorus (TP), total nitrogen (TN), chlorophyll a (Chl-a), chemical oxygen demand (COD), biochemical oxygen demand (BOD), and total suspended solids (TSS). FIRN reduced dependence of prediction accuracy on spatial continuity of water sampling and the amount of training samples. All above WQPs were quantified in a unified framework from unmanned aerial vehicle (UAV) hyperspectral images, where sampled and unsampled regions were correlated through information sharing and delivery.</div><div>New hydrological insights for the region</div><div>Spatial distributions of various WQPs concentrations of Huangyang River in 2024 were visualized by FIRN, where the water quality degraded from 03/2024 to 06/2024 and improved from 06/2024 to 09/2024 with respect to the average WQPs concentrations. Multi-aspect analysis of impact by environmental factors including temperature, daylength, precipitation, and PM2.5 upon water quality was conducted, elucidating the correlation among environmental factors and spatial distributions of WQPs. Spatial distribution visualization of WQPs concentrations indicated locations of potential contamination sources in different time periods. The proposed method was applied to monitor variation of water quality of Huangyang River over time, laying theoretical and technical foundation to formulate water environment management scheme of urban rivers. Experimental results showed that the best mean absolute percent error (MAPE) and coefficient of determination (<span><math><msup><mrow><mi>R</mi></mrow><mrow><mn>2</mn></mrow></msup></math></span>) of FIRN were 7.85% and 0.96 respectively.</div></div>","PeriodicalId":48620,"journal":{"name":"Journal of Hydrology-Regional Studies","volume":"62 ","pages":"Article 102765"},"PeriodicalIF":5.0000,"publicationDate":"2025-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Hydrology-Regional Studies","FirstCategoryId":"89","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2214581825005944","RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"WATER RESOURCES","Score":null,"Total":0}

引用次数: 0

Abstract

Study region

Huangyang River, Zhuhai city, Guangdong province, China

Study focus

High-quality water environment is indispensable to industrial and agricultural advancement. Present methods predicted concentrations of various water quality parameters (WQPs) using their individual models, which suffered calculation instability on various WQPs and needed spatially successive water sampling covering the entire monitoring area to keep calculation accuracy. This study proposed a feature interaction ripple network (FIRN) to retrieve concentrations of WQPs including total phosphorus (TP), total nitrogen (TN), chlorophyll a (Chl-a), chemical oxygen demand (COD), biochemical oxygen demand (BOD), and total suspended solids (TSS). FIRN reduced dependence of prediction accuracy on spatial continuity of water sampling and the amount of training samples. All above WQPs were quantified in a unified framework from unmanned aerial vehicle (UAV) hyperspectral images, where sampled and unsampled regions were correlated through information sharing and delivery.

New hydrological insights for the region

Spatial distributions of various WQPs concentrations of Huangyang River in 2024 were visualized by FIRN, where the water quality degraded from 03/2024 to 06/2024 and improved from 06/2024 to 09/2024 with respect to the average WQPs concentrations. Multi-aspect analysis of impact by environmental factors including temperature, daylength, precipitation, and PM2.5 upon water quality was conducted, elucidating the correlation among environmental factors and spatial distributions of WQPs. Spatial distribution visualization of WQPs concentrations indicated locations of potential contamination sources in different time periods. The proposed method was applied to monitor variation of water quality of Huangyang River over time, laying theoretical and technical foundation to formulate water environment management scheme of urban rivers. Experimental results showed that the best mean absolute percent error (MAPE) and coefficient of determination (

R^{2}

) of FIRN were 7.85% and 0.96 respectively.

Abstract Image

查看原文本刊更多论文

利用空间非连续采样的波纹传播网络，从高光谱图像中量化城市河流水质变化

研究区域：中国广东省珠海市黄阳河。研究重点：高质量的水环境是工农业发展不可或缺的。现有方法采用各自的模型预测各种水质参数（WQPs）浓度，存在各种水质参数计算不稳定的问题，且需要在空间上连续采样覆盖整个监测区域以保证计算精度。本研究提出了一个特征相互作用波纹网络（FIRN）来检索WQPs的浓度，包括总磷（TP）、总氮（TN）、叶绿素a （Chl-a）、化学需氧量（COD）、生化需氧量（BOD）和总悬浮物（TSS）。FIRN降低了预测精度对水采样空间连续性和训练样本数量的依赖。所有wqp均以无人机（UAV）高光谱图像为基础，在统一框架下进行量化，并通过信息共享和传递将采样和未采样区域进行关联。利用FIRN可视化了黄阳河各WQPs浓度在2024年的区域空间分布，其中水质WQPs平均浓度从2024年3月至2024年6月下降，从2024年6月至2024年9月有所改善。对温度、日长、降水、PM2.5等环境因子对水质的影响进行了多向分析，阐明了环境因子与WQPs空间分布的相关性。WQPs浓度的空间分布可视化显示了不同时间段潜在污染源的位置。将该方法应用于黄阳河水质随时间变化的监测，为制定城市河流水环境治理方案奠定了理论和技术基础。实验结果表明，FIRN的最佳平均绝对百分误差（MAPE）和决定系数（R2）分别为7.85%和0.96。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Journal of Hydrology-Regional Studies Earth and Planetary Sciences-Earth and Planetary Sciences (miscellaneous)

CiteScore

6.70

自引率

8.50%

发文量

284

审稿时长

60 days

期刊介绍： Journal of Hydrology: Regional Studies publishes original research papers enhancing the science of hydrology and aiming at region-specific problems, past and future conditions, analysis, review and solutions. The journal particularly welcomes research papers that deliver new insights into region-specific hydrological processes and responses to changing conditions, as well as contributions that incorporate interdisciplinarity and translational science.