Modelling the discharge capacity of Ronggojalu Spring under land cover change scenarios

Ronggojalu Lake, located in the Bromo-Tengger region of East Java, Indonesia, is a vital water source for the Probolinggo–Lumajang Regional Drinking Water Supply System. This study introduces a novel approach by integrating land cover classification accuracy with time-dependent spatial regression modeling to estimate the long-term discharge capacity of Ronggojalu Spring. The objective is to evaluate whether the spring can reliably meet future water demands—projected at 300 L/s over the next 25 years with 99% reliability—amid increasing regional water needs, land development, and climate variability. Using hydrological and geological data from 2022, including precipitation and discharge measurements, we developed a hydrological model to evaluate spring recharge capacity. A scenario-based analysis was conducted to assess the impact of land cover changes and climate variation in the recharge area. The model achieved a Root Mean Square Error of 2.65 (2.5% of average discharge), indicating high accuracy. While the Nash–Sutcliffe Efficiency (NSE) was calculated at 1.00, suggesting perfect fit, a Bias of − 1.43 pointed to a slight underestimation. The findings indicate that land cover change, particularly deforestation and urban expansion, significantly affects recharge rates and can reduce long-term discharge capacity by over 20% under certain scenarios. However, current discharge estimates at 766.9 L/s exceed the projected demand, providing a buffer for future needs. These insights underscore the importance of sustainable land use and water resource management policies to safeguard long-term water supply reliability in the region.