13 Assessment of Hydrological Impacts …
301
(Pulido-Velazquez et al. 2007); however, their application at the scale of the Diarha
catchment is difficult because of the scarcity of hydrogeological data. Overall, despite
its inability to integrate the atmosphere-water-vegetation cycle, the semi-distributed
conceptual approach of the GR4J model correlates simulated and observed Diarha
hydrographs well.
After calibrating the GR4J model and taking into account the interannual variability of the Diarha flows, the outputs of the RCM (GFDL-ESM2G, INM-CM4 and
IPSL-CM5A-LR) of the CMIP5 (Coupled Model Intercomparison Project 5) experiment under two Representative Concentration Pathways (RCP4.5 and RCP8.5) were
used to simulate the trend of flows by 2050, compared to the 1961–2012 reference
period. The models predict that by 2050 there will be a decrease in annual mean
flows of the Diarha compared to the 1961–2012 reference period.
Under the RCP4.5 scenario, the models predict a decrease of 2.1% in precipitation and an average temperature increase of 1.4 °C, resulting in a decrease of 5%
in Diarha flows. The decrease of precipitation and, therefore, of streamflow is more
pronounced under the RCP8.5 scenario, which estimates a rainfall decrease of 12%
and an increase in temperatures of 1.8 °C, causing a net decrease of 25% in rainflow.
The models project, beyond the decline in annual flows, a decrease in the characteristic flow rates (DMAX, DCC_10j, DCC_20j, DC1, DC2 and DC3) and an increase
in low flows in the two emissions scenarios (RCP4.5 and RCP8.5). This increase
in characteristic low flows (DCE) may be due to the limitation of the hydrological
model (GR4J) to model an extremely dry climate.
These results are consistent with those of Bodian et al. (2018) in the Gambia River
at Mako station and Senegal River at Bakel station. In that study, the authors used six
GCMs (CanESM2, CNRM, CSIRO, HadGEM2-CC, HadGEM2-ES and MIROC5)
from CMIP5, and the GR4J model to evaluate the hydrological impact of climate
change. They predicted a decline in Senegal’s streamflow of 8% (RCP4.5) to 16%
(RCP8.5), and a decline in the Gambia’ from 22% (RCP4.5) to 26% (RCP8. 5) by
2050. In addition, Mbaye et al. (2015), using a CORDEX simulation of the regional
REMO climate model as input to the MPI-HM hydrological model, found a decline
in flows, runoff, ETR, and soil moisture in the upper basin of the Senegal River
by 2071–2100, under the scenarios RCP4.5 and RCP8.5. The authors predicted a
decline of more than 50% in the water resources of the Senegal River, especially in
the northern part.
Furthermore, Setegn et al. (2011) used the outputs of nine GCMs under the SRESA2 emission scenario and the SWAT model to assess the impact of climate change
on the water resources of Lake Tana, in Ethiopia. They found a significant decrease
in average annual flows by 2080–2100, consistent with the work of Biao, (2017)
in the sub-basins Betou and Bonou of the Oueme River in Benin. In that study,
the HyMoLAP hydrological model was submitted to the outputs of the HIRHAM5
and RCA4 climate models under the RCP4.5 and RCP8.5 scenarios to assess the
hydrological impact of climate change. It noted, under the RCP4.5 scenario, a drop
in river flows with a magnitude ranging respectively from 25% to -39% and from
20% to 37% by 2020 (2011–2040), 2050 (2041–2070) and 2080 (2071–2100), and
301
(Pulido-Velazquez et al. 2007); however, their application at the scale of the Diarha
catchment is difficult because of the scarcity of hydrogeological data. Overall, despite
its inability to integrate the atmosphere-water-vegetation cycle, the semi-distributed
conceptual approach of the GR4J model correlates simulated and observed Diarha
hydrographs well.
After calibrating the GR4J model and taking into account the interannual variability of the Diarha flows, the outputs of the RCM (GFDL-ESM2G, INM-CM4 and
IPSL-CM5A-LR) of the CMIP5 (Coupled Model Intercomparison Project 5) experiment under two Representative Concentration Pathways (RCP4.5 and RCP8.5) were
used to simulate the trend of flows by 2050, compared to the 1961–2012 reference
period. The models predict that by 2050 there will be a decrease in annual mean
flows of the Diarha compared to the 1961–2012 reference period.
Under the RCP4.5 scenario, the models predict a decrease of 2.1% in precipitation and an average temperature increase of 1.4 °C, resulting in a decrease of 5%
in Diarha flows. The decrease of precipitation and, therefore, of streamflow is more
pronounced under the RCP8.5 scenario, which estimates a rainfall decrease of 12%
and an increase in temperatures of 1.8 °C, causing a net decrease of 25% in rainflow.
The models project, beyond the decline in annual flows, a decrease in the characteristic flow rates (DMAX, DCC_10j, DCC_20j, DC1, DC2 and DC3) and an increase
in low flows in the two emissions scenarios (RCP4.5 and RCP8.5). This increase
in characteristic low flows (DCE) may be due to the limitation of the hydrological
model (GR4J) to model an extremely dry climate.
These results are consistent with those of Bodian et al. (2018) in the Gambia River
at Mako station and Senegal River at Bakel station. In that study, the authors used six
GCMs (CanESM2, CNRM, CSIRO, HadGEM2-CC, HadGEM2-ES and MIROC5)
from CMIP5, and the GR4J model to evaluate the hydrological impact of climate
change. They predicted a decline in Senegal’s streamflow of 8% (RCP4.5) to 16%
(RCP8.5), and a decline in the Gambia’ from 22% (RCP4.5) to 26% (RCP8. 5) by
2050. In addition, Mbaye et al. (2015), using a CORDEX simulation of the regional
REMO climate model as input to the MPI-HM hydrological model, found a decline
in flows, runoff, ETR, and soil moisture in the upper basin of the Senegal River
by 2071–2100, under the scenarios RCP4.5 and RCP8.5. The authors predicted a
decline of more than 50% in the water resources of the Senegal River, especially in
the northern part.
Furthermore, Setegn et al. (2011) used the outputs of nine GCMs under the SRESA2 emission scenario and the SWAT model to assess the impact of climate change
on the water resources of Lake Tana, in Ethiopia. They found a significant decrease
in average annual flows by 2080–2100, consistent with the work of Biao, (2017)
in the sub-basins Betou and Bonou of the Oueme River in Benin. In that study,
the HyMoLAP hydrological model was submitted to the outputs of the HIRHAM5
and RCA4 climate models under the RCP4.5 and RCP8.5 scenarios to assess the
hydrological impact of climate change. It noted, under the RCP4.5 scenario, a drop
in river flows with a magnitude ranging respectively from 25% to -39% and from
20% to 37% by 2020 (2011–2040), 2050 (2041–2070) and 2080 (2071–2100), and