This study investigates the effect of pH on the water quality of King Talal Dam (KTD) in Jordan, utilizing data mining analysis. With treated wastewater use becoming increasingly reliant and the region facing water scarcity, pH dynamics are crucial for sustainable water resource management. A five-year record (2018-2022) of physical, chemical, and biological parameters was analyzed at four strategic sampling stations along the dam system. The study employed supervised machine learning models like XGBoost, Random Forest, and SVR for predicting pH fluctuation and identifying the parameters with an influence (Chen & Guestrin, 2016). Results indicated that bicarbonate (HCO3-), ammonium nitrogen (NH4+-N), nitrite (NO2--N), total suspended solids (TSS), and calcium (Ca2+) significantly influence pH. XGBoost delivered the highest predictive capability among models (R2 = 0.92). The outcomes highlight the complex interlinkages between pH and other water quality parameters, noting the potential of data mining towards enhancing proactive environmental monitoring. The study recommends integrating such predictive models into national water management policy to improve the precision for monitoring and decision-making in arid ecosystems.
Sura Taha Al-Harahsheh (Tue,) studied this question.