article · Sustainability
Assessing surface water quality over broad areas is often constrained by the limitations of standard manual testing methods. This research evaluates the use of hyperspectral reflectance indices and partial least square regression models to monitor key water quality indicators across Qaroun Lake in Egypt. Surface water samples from sixteen sites were examined across 2018 and 2019 for parameters including total dissolved solids, transparency, total suspended solids, chlorophyll-a, and total phosphorus. Newly developed three-band spectral reflectance indices achieved stronger correlations with these indicators than both traditional indices and newly developed two-band versions. When integrated with partial least squares regression, the three-band indices predicted water quality metrics with high statistical accuracy in both calibration and validation sets. These findings demonstrate that combining optimised spectral indices with multivariate statistical modelling offers a fast, non-destructive approach for tracking water pollution and assisting environmental management decisions.
Conventional water quality assessment requires labour-intensive laboratory testing, making large-scale, real-time pollution tracking difficult. Hyperspectral remote sensing techniques provide a faster, non-destructive way to evaluate aquatic ecosystems. By delivering accurate, timely data on water clarity, suspended materials, and nutrient indicators, this approach helps environmental managers and policy authorities detect pollution trends early and protect fragile lake environments.
This work demonstrates an applied, tested methodology for environmental monitoring organisations, lake authorities, and water resource managers seeking rapid surface water assessments. By integrating optimised spectral indices with statistical regression, the approach could be incorporated into commercial remote sensing software or drone and satellite imaging workflows. The technology remains at an applied research stage, requiring integration into operational software platforms before reaching full commercial deployment.
AI-generated from the published abstract. Always read the original work before citing.
Standard methods are limited for monitoring and managing water quality indicators (WQIs) in real-time and on a large scale. Consequently, there is an urgent need to use reliable, practical, swift, and cost-effective monitoring tools that can be easily deployed and assist decision makers in assessing key indicators relevant to surface water quality in a comprehensive manner. Surface water samples were collected and evaluated for water quality at 16 distinct sites across the Qaroun Lake in 2018 and 2019. Different WQIs, including total dissolved solids (TDS), transparency, total suspended solids (TSS), chlorophyll-a (Chl-a), and total phosphorus (TP), were tested for aquatic utilization. An integrated approach comprising WQIs, geospatial techniques, hyperspectral reflectance indices (SRIs) (commonly used SRIs, two-band and three-band SRIs (Spectral index calculated from water spectral reflectance of two or three wavelengths)), and partial least square regression (PLSR) models were used to assess the water quality of Qaroun Lake. According to the findings, the water quality attributes are polluted to varying degrees. The majority of commonly used SRIs presented moderately relationship with four WQIs (transparency, TSS, Chl-a, and TP) (R2 = 0.45 to 0.64), while the majority of newly two-band SRIs (NSRIs-2b) indicated moderate to strong relationships with WQIs (R2 = 0.51 to 0.74), and the majority of newly three band SRIs (NSRIs-3b) presented strong relationships with WQIs (R2 = 0.67 to 0.81). Broadly, the highest coefficients of determination were noticed with the NSRIs-3b followed by the NSRIs-2b and then the commonly used SRIs. For example, the NSRIs-3b (NDSI648,712,696) had stronger relationships with transparency, TSS, and Chl-a with R2 = 0.77, 0.66, and 0.81, respectively, than other SRIs. In addition, the NSRIs-3b (NDSI620,610,622) showed the highest R2 of 0.73 with TSS. The NSRIs-3b coupling with PLSR predicted the WQIs with satisfactory accuracy in the calibration (reach up R2 = 0.85) and validation (reach up R2 = 0.81) datasets. The overall findings of this research study showed that deriving an optimized NSRIs-3b from spectrum region and combining it with PLSR model could be a practical tool for managing water quality of the Qaroun Lake by accurately, timely, and non-destructively monitoring the WQIs.
This page summarises published work. The authoritative version sits with the publisher.
DOI: 10.3390/su131810408
Is something wrong with this record? Report it or request removal.
Discussion
Have you built on this work, tried to replicate it, or seen it applied in practice? Share what you know. Verified researchers and MARATTO™ domain experts can open a discussion, and any member can reply. Contributions are reviewed before they appear.
No discussion yet. Open the first thread.
New to MARATTO™? Create a free account.
