Idrissou, Mouhamed: Modeling water availability for smallholder farming in inland valleys under climate and land use / land cover change in Dano, Burkina Faso. - Bonn, 2020. - Dissertation, Rheinische Friedrich-Wilhelms-Universität Bonn.
Online-Ausgabe in bonndoc:
author = {{Mouhamed Idrissou}},
title = {Modeling water availability for smallholder farming in inland valleys under climate and land use / land cover change in Dano, Burkina Faso},
school = {Rheinische Friedrich-Wilhelms-Universität Bonn},
year = 2020,
month = may,

note = {Effective water management in inland valley catchments is crucial for adaptation to the adverse impact of climate change and land use and land cover change (LULCC) on smallholder farming systems, poverty reduction, attaining food security, and ecosystem preservations in the West African region.
An intensive hydrological instrumentation of four sparse data catchments (Bankandi-Loffing, Mebar, Moutori, and Fafo in Dano, Burkina Faso) has been undertaken in order to better understand hydrological processes which control water availability, to calibrate and validate the physically-based and spatially distributed water balance simulation model WaSiM, to assess the impact of climate and land use and land cover change on water resources, and subsequently to derive strategies for improving the capacity of smallholder farmers to cope with water scarcity and climate variabilities.
The instrumentation of the catchment helped to achieve three years (2014-2016) of high temporal and spatial resolution data. The temporal resolutions of meteorological and stream flow data were 5 min to 10 min, six hours to a week for piezometric data, and 30 min to a week for soil moisture data.
Five rain recorders, seven stream gauges, 64 piezometers in shallow groundwater (< 5 m deep), and 64 soil moisture measurements at three different depths (5 cm, 30 cm, and 50 cm) were installed and operated in the four catchments (total area: 65 km2). Additionally, the groundwater tables of three relatively deep wells (6 m, 16 m, and 25 m deep) were monitored.
The analyses of hydrographs and the flow duration curves (FDC) using observed discharge show less discharge in the headwater sub-catchments compared to the downstream sub-catchments. This is due to the low contribution of base flow in the headwater sub-catchments. The decomposition of total runoff using observed hydrographs and stream electric conductivity suggests that interflow is the major contributor to total discharge.
The calibration and validation of the Bankandi-Loffing catchment achieved a good model performance using the coefficient of determination (R2), the Nash-Sutcliffe efficiency (NSE), the Kling-Gupta efficiency (KGE), and the percent bias (Pbias). The R2 ranges from 0.47 to 0.95, NSE from 0.40 to 0.95, and KGE from 0.57 to 0.84 between the observed and simulated discharge. The numerical performance for soil moisture modeling is 0.70 for both R2 and NSE, and 0.80 for KGE while for the groundwater table modeling the results are 0.30, 0.20, and 0.5 for R2, NSE, and KGE, respectively. The fact that the transfer of the parameter set from Bankandi-Loffing to Mebar catchment without recalibration resulted in a good model performance (R2: 0.93, NSE: 0.92, and KGE: 0.84 in 2014-2015; R2: 0.65, NSE: 0.64, and KGE: 0.59) suggests the strong robustness of WaSiM in the investigated area.
The resulting water balance shows that evapotranspiration is quantitatively the most important hydrological process, physical evaporation dominates the evapotranspiration, and 14% of rainfall runs out of the catchment as discharge. Interflow dominates runoff at the headwater sub-catchments whereas base flow is the major runoff component in the downstream area where the inland valley bottoms are located.
The conversion of savanna to cropland leads to an increase of surface runoff. This is potentially associated with an exacerbation of soil erosion and soil fertility loss. Therefore, supplementing the current erosion technique (stone-belt) with agroforestry and/or mulching will reduce the negative effects of land cover change.
Two scenarios were considered during the impact assessment. The first scenario evaluated exclusively the climate change impact by utilizing five regional climate models (RCMs) using land use and land cover (LULC) of the year 2013 for both the reference period (1971-2000) and the projection period (2021-2050). Each RCM is composed of the representative concentration pathways (RCPs) 4.5 and 8.5. The results indicate large uncertainty in the discharge projection for the future. Three RCMs predict an increase of total runoff for the projection period compared to the reference period. The mean total runoff increase is +61% (standard deviation Std= 31%) compared to the reference period. However, two RCMs project a decrease of total runoff. The mean total runoff decrease is -34% (Std= 10%) compared to the reference period.
The second scenario utilizes the five RCMs and LULC 2013 for the reference period and LULC 2030 for the projection period in order to assess the combined impact of climate change and LULCC. The results suggest that LULCC exacerbates the increase of total runoff in combination with the three RCMs with a mean increase in total runoff by +108% (Std= 38%) compared to the reference period (versus mean= +61% in the first scenario). However, for the two RCMs predicting a decrease of total runoff, LULCC reduces the decrease of total runoff. The mean decrease is -20% (Std= 10%) compared to the reference period (versus mean= -34% in the first scenario).
The results of this study can be used as input to water management models in order to derive strategies to cope with present and future water scarcities for smallholder farming in the investigated area.},

url = {}

The following license files are associated with this item: