Recharge Happens in Extremes

01The evidence

• Rainfall intensity distribution and extreme events were shown to control recharge dynamics in a shallow karst aquifer, with the intensity structure mattering more than the total.Journal of Hydrology, 2025
• Shallow aquifer dynamics respond to climate variability in ways that depend on event structure, which complicates recharge estimates built on seasonal totals.Groundwater dynamics

02The tension

Recharge is estimated with water-balance models that distribute rainfall over the season and apply a runoff coefficient. These models work on average and fail in the extremes, which are exactly the years that matter for long-term storage. The residual is absorbed into a calibration factor that then mis-forecasts.

03The theory

For water to reach an aquifer, it must travel through the unsaturated zone. Under gentle rain, soil moisture rises and most water is returned to the atmosphere by evapotranspiration. Under intense rain, infiltration exceeds the soil's capacity to redistribute, and water moves through preferential paths to depth before it can be used.

The consequence is a threshold: below a critical intensity, recharge is small and roughly proportional to excess rainfall; above it, recharge rises sharply because the flow path changes. That makes the annual total a poor predictor and the storm distribution a good one.

The theory predicts that recharge should correlate with the number of events above a threshold rather than with annual rainfall, and that a single exceptional season can dominate a decade of storage. It also predicts that recharge estimates will diverge most in regions where the intensity distribution is changing fastest.

This matters for groundwater management because it inverts the usual advice: conserving water in wet years is less important than capturing the extreme events, which requires different infrastructure.

04What this is not

05Predictions

• P1Recharge should correlate with the count of events above a threshold intensity better than with annual rainfall.
• P2A single exceptional season should account for a disproportionate share of multi-year storage change.
• P3Regions with changing intensity distributions should show the largest divergence between modelled and observed recharge.

06What would refute it

Recharge tracking annual totals after intensity is controlled would return the problem to the water balance.

07What it would change

It would change how aquifers are managed and how recharge infrastructure is sited.