A cost-effective and efficient framework to determine water quality monitoring network locations

Abstract

A crucial part in designing a robust water quality monitoring network is the selection of appropriate water quality sampling locations. Due to cost and time constraints, it is essential to identify and select these locations in an accurate and efficient manner. The main contribution of the present article is the development of a practical methodology for allocating critical sampling points in present and future conditions of the non-point sources under a case study of the Khoy watershed in northwest Iran, where financial resources and water quality data are limited. To achieve this purpose, the river mixing length method (RML) was applied to propose potential sampling points. A new non-point source potential pollution score (NPPS) was then proposed by the analytic network process (ANP) to classify the importance of each sampling point prior to selecting the most appropriate locations for a river system. In addition, an integrated cellular automata-Markov chain model (CA-Markov) was applied to simulate future change in non-point sources during the period 2026-2036. Finally, by considering anthropogenic activities through land-use mapping, the hierarchy value, the non-point source potential pollution score values and budget deficiency in the study area, the seven sampling points were identified for the present and the future. It is not expected, however, that the present location of the proposed sampling points will change in the future due to the forthcoming changes in non-point sources. The current study provides important insights into the design of a reliable water quality monitoring network with a high level of assurance under certain changes in non-point sources. Furthermore, the results of this study should be valuable for water quality monitoring agencies looking for a cost-effective approach for selecting sampling locations.

Designing the National Network for Automatic Monitoring of Water Quality Parameters in Greece

“…Most of the methodologies propose the selection of the water quality monitoring sites based on statistical approaches, such as principal component analysis, principal factor analysis, canonical correlation analysis, correlation analysis, cluster analysis, regression analysis, artificial neural networks, maintenance of variance extension, and matter element analysis (e.g., [30][31][32][33][92][93][94]). Other researchers employ other approaches and techniques, such as multicriteria decision-making analysis, genetic algorithm, entropy-based and fuzzy logic approaches, or combinations of the above (e.g., [27][28][29]34,95,96]).…”

“…Of the methodologies described above, GIS-MCDA is a cost-effective and easily applied approach that does not require qualified personnel and access to extended information, while it has been used successfully in the design of the networks for monitoring of water quality parameters in the past (e.g., [27,28,95,97]). In the present effort, the GIS-MCDA introduced takes into consideration…”

“…Of the methodologies described above, GIS-MCDA is a cost-effective and easily applied approach that does not require qualified personnel and access to extended information, while it has been used successfully in the design of the networks for monitoring of water quality parameters in the past (e.g., [27,28,95,97]). In the present effort, the GIS-MCDA introduced takes into consideration the most important nonpoint and point pressures due to human activities, and other criteria, such as increased vulnerability, while it has been focused on Greece's particularities (e.g., the general agricultural practices taken place, the existence of informal industrial activities, the vulnerability arisen due to karstic geological formations, and the increased interest for monitoring in areas included in networks of environmental interest).…”

“…A successful environmental water quality monitoring program relies on the following main elements [26]; identification of the monitoring objectives, national and EU institutional setting, preliminary surveys, monitoring design (selection of representative sampling sites and frequencies, and water quality variables), adjustment to budget and operational restrictions, design of field protocols, field monitoring operations, laboratory activities, quality-assurance procedures, data management and analysis, and product development and reporting. Based on the review concerning the design of water quality monitoring networks, there is a variety of methodological approaches, e.g., statistical, multicriteria decision-making, entropy-based and fuzzy logic approaches, or a combination of the above (e.g., [27][28][29][30][31][32][33][34]). These methodologies are usually difficult to be applied by not qualified personnel and require long and reliable time-series [35]; therefore, they are usually used in evaluation and optimization of an existing monitoring program, rather than in the initial design of water quality monitoring network.…”

Pollution mapping in the urban segment of a tropical river: is water quality index (WQI) enough for a nutrient-polluted river?

“…It has the ability to treat bulk of spatial and temporal data from different monitoring sites into comprehensive information. Different statistical techniques like principal component analysis (PCA), cluster analysis (CA) and factor analysis (FA) are very useful for riverine studies as these have the ability to evaluate spatial as well as temporal river quality variations and have the aptitude to identify the potential water contamination sources (Sabia et al 2018;Alilou et al 2018;Li et al 2017;Iqbal et al 2017;Villas-Boas et al 2017;Zhao et al 2016).…”