Long-Term Patterns of Groundwater E. coli Contamination in Tripoli and Souq al‑Jomaa, Libya (1990–1995): Seasonal Variability, Rainfall Influence, and Public Health Implications

Abstract

Domestic water supply in Libya and many other arid and semi-arid regions is primarily from groundwater. This study assessed historical trends of Escherichia coli (E. coli) contamination in private wells in Tripoli and Souq al-Jomaa, Libya, and their relationship with seasonal rainfall variability and reported waterborne disease cases for the period 1990–1995. Methods: A retrospective observational study was conducted using archival records from the Ras Hassan Water Laboratory, which included microbiological analyses of 2,428 wells (1,230 in Tripoli and 1,198 in Souq al‑Jomaa). Annual contamination prevalence and monthly contamination percentages were derived from routine laboratory testing for E. coli using membrane filtration techniques (0.45 μm cellulose ester, 100 mL, m‑FC agar, 44.5 °C for 24 hours). The rainfall data were obtained from the National Meteorological Center, and the waterborne disease data were collected from hospital archives. Seasonal comparisons were performed using repeated-measures ANOVA, the correlation between rainfall and contamination was calculated using Pearson's correlation, and temporal trends were assessed using the Mann-Kendall test. Results: In Tripoli, the six-year average E. coli prevalence was 70.4% compared to 56.5% in Souq al-Jomaa (Mann-Whitney U test, p<0.01). Monthly contamination was highest in winter (Dec–Feb: 79% in Tripoli, 71% in Souq al‑Jomaa) and lowest in summer (66% and 44%, respectively). Annual rainfall was variable, ranging from 420 mm in 1990 (highest) to 240–280 mm in 1994–1995 (lowest), with intermediate values in 1992–1993 (~330 mm). Monthly rainfall was strongly correlated with contamination in Souq al‐Jomaa (r = 0.78, p < 0.01) and moderately in Tripoli (r = 0.52, p < 0.05). Peak contamination occurred during the wettest months (Dec–Jan, 60–90mm). A total of 1,748 cases of waterborne diseases were reported, with the highest incidence in 1994 (431 cases). Females represented 62% of cases. No significant temporal trend was detected for contamination or disease, but the six-year limit statistical power. Conclusion: Rainfall appears to be an important contributor to seasonal E. coli contamination in shallow wells, especially in shallower aquifers (Souq al‑Jomaa). The observed temporal patterns suggest a need for pre‑winter well protection and disinfection.