The DIGAP project confronts a problem observed in the field by Malagasy partners and humanitarian actors:

the difficulty of anticipating humanitarian crisis situations linked to droughts. DIGAP argues that a consideration of groundwater status is crucial to anticipating water scarcity-related humanitarian crises in semi-arid regions that are almost devoid of surface water. In fact for some years environmental markers have provided scientific evidence on the hydric situation (e.g. rainfall deficit, drop in piezometric levels). We seek to determine how we can better use water status and more specifically groundwater status, in anticipating humanitarian crises.

First (work-package 1, in which the postdoc will be involved), we will develop an innovative methodology based on remote sensing and modeling to characterize groundwater dynamics at the regional scale in an area that lacks basic documentation. Second (work-package 2), we will quantify the associations among various environmental factors and food security, in addition to human health indicators through the use of a multilevel epidemiological modeling approach. We expect that our results will unravel the effects of environmental and anthropic factors in the development of drought-related humanitarian crises and thereby sustainably strengthen Madagascar’s early warning system.

Work program

In the semi-arid southwest Madagascar region, permanent streams are essentially absent. These conditions allow us to assume that the precipitation minus evapotranspiration (P-ET) term should be a valid proxy for estimating groundwater recharge. We will therefore be able to estimate recharge in a spatialized manner (1*1km) for the entire southern region by subtracting ET from precipitation (resulting from another part of the project). Propagation of the precipitation and ET uncertainties will provide recharge uncertainty.

This recharge estimate can be evaluated at two spatial scales: i) at the regional scale through GRACE (Gravity Recovery and Climate Experiment) satellite monitoring, which provides a resolution of 300*300km (thus, the entire southwest region will be considered a single entity). We will use specific maps from the GRACE processing chain that provide mass variations related to hydrological dynamics and which are expressed in millimeters (Prevost et al. 2019); ii) at the local scale by comparing P-ET with recharge calculated using GROSoM piezometric monitoring (Carrière et al. 2021). Half of the piezometric dataset (dataset 1) will be used here and the remainder (dataset 2) will be used later. We will construct ten new piezometers in the vicinity of existing boreholes to perform long-term pumping tests to estimate the storage coefficient, a key parameter for expressing in millimeters the recharge calculated by piezometric monitoring. Pumping test interpretations will be performed using the OUAIP software developed by BRGM (French Geological Service). We will use the ESPERE-v2 software to calculate recharge according to the water table fluctuation (WTF) approach.

To finish, we will reproduce the general groundwater dynamics at the municipality and district scale through a simplified and robust reservoir-modeling approach. We will use recharge (P-ET) as an input signal into the model. Calibration and validation of these models will be done with the second part of the GROSoM dataset (dataset 2), which was not used in the previous step. The groundwater dynamics will allow us to consider multi-year effects such as the accumulation of dry years.

Main activities

• Characterization of aquifer hydrodynamic properties (select drilling company, drilling monitoring, pumping tests in Madagascar and interpretation).
• Involvement in the preparation of the GROSOM observatory’s hydrogeological newsletter (3 to 4 times a year) in close collaboration with local partners.
• Analysis and modeling of piezometric time series (interpretation and gap filling).
• Estimating recharge from piezometric time series (e.g. WTF method).
• Estimate recharge from GRACE data.
• Comparison of different recharge estimates and quantification of associated uncertainty.
• Regional « reservoir-type » hydrogeological modeling based on global products and calibration/validation with
  GROSOM data.

Starting date

Beginning 2024 (the exact date can be discussed).

Application

Send to: Simon Carrière:

• CV
• cover letter
• name of three referees0

Application analysis

First round: November 20^th

Transversal knowledge needed

The candidate should have a PhD in hydrological sciences or hydrogeology.

He/she will have experience in modeling or hydrology on a regional scale.

An experience in a developing country would be a plus.
The candidate will need to be able to present scientific work, in both oral and written forms.

Skills

Hydrogeology, hydrology, modeling (bucket type), drilling follow-up, programming (R and/or Python).

Social skills

Openness to interdisciplinary work, curiosity, autonomy, ability to work in a poor country with difficult field
conditions (we plan around 2 months in Madagascar)

References

Carrière, S.D., Heath, T., Rakotomandrindra, P.F., Ollivier, C., Rajaomahefasoa, R.E., Rakoto, H.A., Lapègue, J.,
Rakotoarison, Y.E., Mangin, M., Kempf, J., Razakamanana, T., Chalikakis, K., 2021. Long‐term Groundwater
Resource Observatory for Southwestern Madagascar (GROSoM). Hydrological Processes e14108.
https://doi.org/10.1002/hyp.14108

Prevost, P., Chanard, K., Fleitout, L., Calais, E., Walwer, D., van Dam, T., Ghil, M., 2019. Data-adaptive spatiotemporal filtering of GRACE data. Geophysical Journal International 219, 2034–2055.
https://doi.org/10.1093/gji/ggz409