Abstract
The lack of data and suitable methods to quantify regional hydrological processes often hinders sustainable water management and adaptation to climate change in semiarid regions, particularly in the Sahel, which is known for its climatic variability. Here we show that 36Cl from nuclear tests is a promising method to estimate water transit times and groundwater recharge rates on the catchment scale, and to distinguish water and chloride cycles. 36Cl was measured in 131 surface and groundwater samples in the Chari-Logone sub-catchment of the emblematic Lake Chad Basin, located in central Sahel. It was found that only 12 ± 8% of the catchment is connected to the main rivers. Groundwater supporting rivers in the upper humid part of the catchment has a mean transit time of 9.5 ± 1 years and a recharge rate of 240 ± 170 mm yr−1. In the lower Sahelian part of the catchment, stream-focused recharge yields recharge rates up to 78 ± 7 mm yr−1 in riparian groundwater against 16 ± 27 mm yr−1 elsewhere. Our estimates suggest that aquifers in the Sahel host a significant amount of renewable water, which could therefore be used as a strategic freshwater resource.
Highlights
Freshwater resources in semi-arid regions are facing a number of stress factors, such as rapid population growth with the associated economic and agricultural developments, and climate change[1]
Dissolved gas tracers provide temporal constraints on the water transit time of surface and subsurface flows within a catchment[13]
Bomb-produced tritium was widely applied as a young age tracer in the 1960s6, but the combined effect of removal by rain and radioactive decay (t1/2 3H = 12.3 years) leads to an ambiguous age determination for recharge periods between 1975 and 201015,16. 36Cl appears to be a relevant alternative tracer[17]: chloride is a ubiquitous ion, the production of 36Cl associated with nuclear
Summary
Measurements of 36Cl were carried out by Accelerator Mass Spectrometry[48] at the French AMS National facility, ASTER at CEREGE49. Because of low chloride contents in surface and some groundwater samples, the addition of a 35Cl enriched spike was required to reach a total chloride amount of 2 mg and to precisely determine the Cl concentration using the ID-AMS technique[50]. Water samples were processed in batches of 10–15 samples with a spiked blank and an unspiked blank at the end of each batch to estimate chemical contaminations on chloride concentrations and on 36Cl/Cl ratios. The measured 36Cl/Cl of spiked samples are at least one order of magnitude above spiked blanks (36Cl/Cl = 4.0 10−15 at at−1, n = 12) and the calculated 36Cl/Cl sample ratios were corrected from the blanks. Total uncertainties, including internal errors and external reproducibility, on the determination of 36Cl/Cl ratios and chloride concentrations by ID-AMS were respectively estimated at 7% and 5%50. The calculations made on the 36Cl data are succinctly described in the main paper, and more extensively in the Supplementary Materials
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
