Using Remote Sensing and Multivariate Statistics in Analyzing the Relationship between Land Use Pattern and Water Quality in Tien Giang Province, Vietnam
This study was carried out to understand how land use patterns influence surface water quality in Tien Giang Province using remote sensing and statistical approaches. Surface water quality data were collected at 34 locations with the frequency of four times (March, June, September, and November) in 2019. Water quality parameters were used in the analysis, including pH, temperature, electrical conductivity (EC), total suspended solids (TSS), dissolved oxygen (DO), biological oxygen demand (BOD),
... chemical oxygen demand (COD), ammonium (N-NH4+), nitrite (N-NO2−), nitrate (N-NO3−), sulfate (SO42−), orthophosphate (P-PO43−), chloride (Cl−), total nitrogen (TN), total phosphorus (TP), and coliform. The relationship between land use patterns and water quality was analyzed using geographic information techniques (GIS), remote sensing (RS), statistical approaches (cluster analysis (CA), principal component analysis (PCA), and Krustal–Wallis), and weighted entropy. The results showed water quality was impaired by total suspended solids, nutrients (N-NH4+, N-NO2−, P-PO43−), organic matters (BOD, COD), and ions (Cl− and SO42−). Kruskal–Wallis analysis results showed that all water quality parameters in the water bodies in Tien Giang Province were seasonally fluctuated, except for BOD and TN. The highest levels of water pollutants were found mostly in the dry season (March and June). The majority of the land in the study area was used for rice cultivation (40.64%) and residential (27.51%). Water quality in the study area was classified into nine groups corresponding to five combined land use patterns comprising residential–aquaculture, residential–rice cultivation, residential–perennials, residential–rice–perennial, and residential–rice–perennial crops–aquacultural. The concentrations of the water pollutants (TSS, DO, BOD, COD, N-NH4+, N-NO2−, Cl−, and coliform) in the locations with aquaculture land use patterns (Clusters 1 and 2) were significantly larger than those of the remaining land use patterns. PCA analysis presented that most of the current water quality monitoring parameters had a great impact on water quality in the water bodies. The entropy weight showed that TSS, N-NO2−, and coliform are the most important water quality parameters due to residential–aquaculture and residential–rice cultivation; EC, DO, N-NH4+, N-NO2−, Cl−, and coliform were the significant variables for the land use type of residential–perennial crops; N-NO2−, P-PO43−, and coliform for the land use pattern of residential–rice cultivation–perennial crops) and N-NH4+, N-NO2−, Cl−, and coliform for the land use pattern of residential–rice cultivation–perennial crops–aquaculture. The current findings showed that that surface water quality has been influenced by the complex land use patterns in which residential and rice cultivation may have major roles in causing water impairment. The results of the water quality assessment and the variation in water properties of the land use patterns found in this study provide scientific evidence for future water quality management.