Climate change is likely to significantly reduce groundwater recharge in Ghana’s Densu Basin, raising concerns over future water security, according to a 2026 study led by researchers at the Kwame Nkrumah University of Science and Technology, Kumasi (KNUST).

The study Published in Groundwater for Sustainable Development found that rising temperatures and declining rainfall under multiple climate scenarios could lead to substantial reductions in groundwater recharge, particularly in the northern parts of the basin.

Using machine learning models, including artificial neural networks and random forest techniques, researchers showed that future climate conditions will weaken the natural processes that replenish underground water reserves.

The Densu Basin, which supplies water to parts of southern Ghana including Accra, is already under pressure from urbanisation and environmental degradation. The findings suggest that climate change could further strain the basin’s ability to sustain water supply.

The researchers identified short-term rainfall patterns, measured through indicators such as the Standardised Precipitation Index (SPI-3 and SPI-6), as key drivers of groundwater recharge variability. This highlights the basin’s sensitivity to near-term drought conditions.

To improve prediction accuracy, the study combined outputs from multiple machine learning models, finding that ensemble approaches provided more reliable projections than single-model methods.

The research also points to broader risks for Ghana, where groundwater remains a primary source of domestic water, particularly in rural communities. Reduced recharge could affect agriculture, household water supply and ecosystem stability.

The authors said the findings underscore the need for adaptive water management strategies, including improved monitoring, climate-resilient planning and sustainable groundwater use.

“Climate variability threatens groundwater sustainability, requiring reliable predictive tools,” the study noted, adding that data-driven approaches such as machine learning can support decision-making in water resource management.

The study is among the first in Ghana to apply machine learning and multiple climate models to project future groundwater recharge, addressing a gap in research on climate impacts on subsurface water systems.