Temporal Stability Analysis for the Evaluation of Spatial and Temporal Patterns of Surface Water Quality

Abstract

Better characterizing the spatio-temporal pattern of water quality would increase the ability to effectively manage water resources. This study applied the concept of temporal stability analysis (TSA) to explore the temporal characteristics of spatial variability in surface water quality. Monitoring data from 41 monitoring stations in Qiantang River, China for 2017–2019 were used to assess four indicators: dissolved oxygen (DO), permanganate index (CODMn), total phosphorus (TP), and ammonia nitrogen (NH3–N). A Spearman’s rank correlation for each pair of monitoring times demonstrated that the spatial pattern of water quality was maintained for a specific period of time. The TP concentration was most temporally stable compared with the other three indicators across the study area. A temporal analysis of relative differences was applied to examine the temporal stability of the sampling sites. The mean concentrations, with acceptable errors, were estimated from the representative sites identified for the four water quality indicators. Different metrics were assessed to identify the temporally stable sites. The standard deviation of the relative difference (SDRD) and index of temporal stability (ITS) were found to be better for identifying the stable sites compared to the mean absolute bias error (MABE) and root mean square error (RMSE) in this study. A correlation analysis between the temporal stability indices and potential influencing factors showed that land use proportions (forest, built-up land, and agricultural land), and socio-economic indicators (gross domestic product [GDP] and population density) were closely associated with the temporal stability of water quality. The results showed evidence that the TSA method was feasible and effective in identifying the temporal stability of surface water quality and optimizing the water quality monitoring program. This study’s method and findings can help improve surface water quality monitoring strategies and water resource management.

What is the relationship between land use and surface water quality? A review and prospects from remote sensing perspective

“…The limitations of Spearman rank correlation coefficient is that when the data is converted to rank sum, it will lose information. If the data is normal distribution, Spearman rank correlation coefficient is not as strong as PCA (Zhang et al 2022).…”