Estimating groundwater recharge: A review

Since the mid-1980s there has been a relative explosion of recharge studies for (semi-)arid regions reported in the scientific literature. It is therefore relevant to assess what we now know and to offer additional guidance to water resource development practitioners. This paper identifies recurring recharge evaluation 'problems' and reviews some recent advances in estimation techniques. The determination of recharge fluxes in (semi-)arid regions remains fraught with uncertainty. Multiple tracer approaches appear to offer the best potential for reliable results in local studies which require 'at-point' information, though investigations in the Botswana Kalahari indicate that these are not straightforward. In the areas targeted, preferential flow contributes on average similar to 50% of the estimated total recharge, with values as high as 70%similar to 90%. Tracer results (e.g. Cl-, H-3) must therefore be interpreted with care in areas with multimodal flow in the vadose zone. Tritium is influenced by vapour transport when flux rates decrease below 10mm/a and accurate estimation of total chloride deposition is essential. The Botswana studies also indicate that long-term average areal recharge rate derived from the He-4 method has the same order of magnitude as values obtained by chloride mass balance, groundwater flow- and hydrochemical-/isotope modelling. The frequently studied issues of localized recharge and spatial variability need not be a problem if concern is with regional estimates. The key for practitioners is the project objective; this will dictate whether multiple 'at-point' or area-/groundwater-based estimation methods are appropriate. Many indirect (wadi) recharge studies reported in the literature are site specific; the relationship between 'at point' hydraulic properties and reach transmission losses demands further investigation.