Spatiotemporal Assessment of Bacteriological Water Quality in Rivers: A Censoring-Aware Integrated Approach (Harşit Stream, Türkiye)

Authors : Buse Eraslan Akkan, Cengiz Mutlu, Bülent Verep

Pages : 913-922

Doi:10.35229/jaes.1793470

View : 62 | Download : 150

Publication Date : 2025-11-30

Article Type : Research Paper

Abstract :This paper presents an integrated, reproducible framework for assessing bacteriological quality in rivers. The aim is to quantify spatial variation and temporal trends with non-parametric statistics using censored Kaplan-Meier (KM) summaries and a threshold event approach. For this purpose, monthly surface water samples were collected under sterile conditions at seven stations on the Harşit Stream (Giresun, Türkiye), and transported to the laboratory under a cold chain protocol. Total coliform (TC), Fecal coliform (FC), Escherichia coli (EC), and fecal Streptococci (FS) counts (MPN/100mL) were used as indicators of bacteriological quality. Findings were examined using descriptive statistics and Kaplan–Meier summaries, considering right censoring. Furthermore, station differences and relationships were tested. Moreover, multivariate structure and quality trends were evaluated using statistical analyses. Furthermore, bacteriological quality assessment was performed according to national and international guidance. This study verified the effect of KM summaries on censoring due to clustering at the upper limit. Wide-ranging positive and strong relationships were detected between parameters (p<0.001). A common gradient was detected in all bacterial indicators. The trend over time showed a decrease only at Station 2. Furthermore, the stations are clearly classified into high load (stations 6–7), low load (stations 1–3), and intermediate load (stations 4–5; with 2 higher distinctions) clusters. In conclusion, KM-based censoring of sensitive summaries and threshold event analysis comprehensively manage censoring in bacteriological stream data. In this way, decision makers will realize the risk indicators that show how long it will take to exceed the threshold free from substitution-based bias. This will facilitate rapid and equitable interventions in sustainable basin management. This application is one of the first systematic KM applications for river bacteriology in Türkiye and has the potential to be applied to different watersheds.
Keywords : Su kalitesi, uzamsal-zamansal analiz, sansürlü veri, sağkalım analizi, çok değişkenli analiz, Harşit Çayı

ORIGINAL ARTICLE URL