Identification of Water Quality Pollution Sources in a Transitional Urban Area Using Principal Component Analysis (PCA) of Physico-Chemical Parameters

Rapid urban growth in semi-arid India is increasingly affecting surface-water quality, although the specific sources responsible for pollution across urban–rural transition zones remain insufficiently resolved. This study examines the principal pollution drivers along an urbanization gradient in Hyderabad, Telangana, by combining principal component analysis (PCA) with land use/land cover (LULC) buffer analysis at distances of 100–1000 m. Pre-monsoon grab samples were collected from five contrasting locations: Vikarabad as a rural reference, M.G.B.S. representing the urban core, Nagole representing a residential area, Peerzadiguda representing a peri-urban setting, and Pratapsingaram representing a mixed urban–agricultural environment. Fifteen physico-chemical parameters were analyzed following APHA (Citation2017) and IS 10500:2012 protocols, including pH, alkalinity, acidity, hardness, Cl-, NO3-, NO2-, NH4+, Fe, TSS, TDS, electrical conductivity, turbidity, and initial and final dissolved oxygen; BOD5 was included as a supplementary variable. PCA was conducted using the z-standardized correlation matrix in XLSTAT 2024.4.2, with loadings of ≥|0.7| considered meaningful. Two components were retained, jointly explaining 79.96% of the total variance. PC1 accounted for 44.86% and showed strong positive contributions from ammonium (+0.99), alkalinity (+0.98), acidity (+0.84), iron (+0.79), and BOD5 (+0.73), alongside strong negative loadings for DO_final (−0.91) and DO_initial (−0.83). This component represents a combined organic and ammoniacal pollution signature associated with domestic sewage inputs. PC2 explained 35.10% of variance and was characterized by TDS (+0.95), nitrate (−0.91), conductivity (−0.82), pH (−0.70), and nitrite (−0.70), indicating ionic and pH variability linked to surrounding land-cover conditions. Integrating PCA with LULC data clearly differentiated the natural-cover rural reference site from the urban, peri-urban, and agricultural locations. Notably, PC1 intensity was more strongly associated with sewerage deficiencies than with built-up density, identifying untreated domestic sewage as the principal pollution driver in the investigated urban and peri-urban waters. A targeted monitoring approach prioritizing BOD5 and NH4+ is therefore recommended for cost-effective surveillance of Hyderabad’s surface waters under IS 10500:2012.