the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Combined impacts of climate and land-use change on future water resources in Africa
Celray James Chawanda
Albert Nkwasa
Wim Thiery
Ann van Griensven
Abstract. Africa depends on its water resources for hydroelectricity, inland fisheries, and water supply for domestic, industrial, and agricultural operations. Anthropogenic climate change (CC) has changed the state of these water resources. Land use and land cover has also undergone significant changes due to the need to provide resources to a growing population. Yet, the impact of the Land Use and Land Cover Change (LULCC) in addition to CC on the water resources of Africa is underexplored. Here we investigate how precipitation, evapotranspiration (ET), and river-flow respond to both CC and LULCC scenarios across the entire African continent. We set up a SWAT+ model for Africa and calibrated it using the Hydrological Mass Balance calibration (HMBC) methodology detailed in Chawanda et al., (2020a). The model was subsequently driven by an ensemble of bias-adjusted global climate models to simulate the hydrological cycle under a range of CC and LULCC scenarios. The results indicate that the Zambezi and the Congo River basins are likely to experience reduced river flows under CC by up to 7 % decrease, while the Limpopo will likely have higher river flows. The Niger River basin is likely to experience the largest decrease in river flows in all of Africa due to CC. The Congo River basin has the largest difference in river flows between scenarios with (over 18 % increase) and without LULCC (over 20 % decrease). The projected changes have implications on agriculture and energy sectors and hence the livelihood of people on the continent. Our results highlight the need to adopt policies to halt global greenhouse gas emissions and to combat the current trend of deforestation to avoid the high combined impact of CC and LULCC on water resources in Africa.
- Preprint
(2700 KB) - Metadata XML
- BibTeX
- EndNote
Celray James Chawanda et al.
Status: final response (author comments only)
-
RC1: 'Comment on hess-2023-93', Anonymous Referee #1, 14 Jun 2023
Review comments
Journal: Hydrology and Earth Systems Sciences
Manuscript id: https://doi.org/10.5194/hess-2023-93 Â
Title: Combined impacts of climate and land-use change on future water resources in Africa
Â
The study projects how runoff and river flow across Africa may change under different future CC and LULCC scenarios by first calibrating the Hydrological Mass Balance Calibration methodology at continental-scale using SWAT+ hydrological model. Subsequently, authors used an ensemble of bias-adjusted global climate models ISIMIP to simulate future runoff, evapotranspiration (ET), and river flow projections under CC and LULCC scenarios for RCP 2.6, 6.0 and 8.5 as compared to the historical period.
Â
Generally, the manuscript is well-written with novel research results at the continental level. And I can recommend its acceptance by HESS following some editorial and suggestions outlined under. I hope these comments will not prevent its publication in HESS.
Â
Comments and suggestions
Â
- The authors used two experiments of CC and LULCC scenarios to evaluate their results. Yes, this is good to use but my concern is that what about the consideration of other factors that are interlinked to the oceans' atmospheric pressures as they may affect the precipitation distributions.
Â
- The authors simulate the historical (1975 - 2005) and the far future (2070 - 2100) periods. I think it would have been better if they could produce the near (2021-2040) and midterm (2041-2070) results for better preparation of the societies for the climate variabilities that may evolve as the climate changes in the near future.
Â
- Regarding model calibration, it is good that the authors have used NSE values. But some NSE values are still below the threshold values in some stations. Would have been added some other model performance measures would be good to crosscheck the results and see the outputs.
Â
- Figure 2: I couldn’t see figures 1a and 1b, Could you please provide and level it in the figure. And the same is in Figure 3 as well.
Â
- In the results section authors have repeatedly used the latitude grids to spatially elaborate their results. That is good but it could have been better if at least one map of Africa could show the latitude and longitude values so that we can easily understand where it would be they are referring to.
Â
- In section 3.1.2, it could be good if the authors explain how they evaluate the model performance in quantifying the ET performance across the river basins.
Â
Citation: https://doi.org/10.5194/hess-2023-93-RC1 -
RC2: 'Comment on hess-2023-93', Anonymous Referee #2, 18 Jul 2023
The manuscript assesses the impact of climate change and land cover change over Africa. It is a good case study over Africa. The manuscript is overall well written but some additonal details must be presented to make it acceptable to HESS.
1) The model calibration part has been completely referred to a previous paper. It is necessary to explain the relavant calibration principles and details here in brief. Please try to make the paper self explanatory to the extent possible. It can be provided as appendix if the authors feel so. The term soft data has been defined only in Chawanda et al. (2020a) and not here.
2) I could not read the previous paper in full (Chawanda et al., 2020a). However, I think that this paper is somewhat similar to the previous paper referred here in terms of assessment of climate change impact. A comparison of results over the common landmass is needed to be presented in the current manuscript.
3) What is the benefit of doing this climate change and LULC study? The results are averaged over large basins for long time periods. Hence, how beneficial this study will be for local scale adaptation or management?
4) Results should be presented for near term periods (e.g. 2030-2050, 2050-2070). Only presenting long term results may not be useful and verifiable in a possible time frame.
5) It has been mentioned that over some zones where streamflow data is not available, calibration was done only using ET. Highlight those zones and explain how this ET calibration was done.
6) The calibration is based on preserving the long term averages of the water cycle components. Please explain conceptually how this can be used to model yearly dynamics of the water cycle.
7) I understand that ground water component for calibration was calculated as a residual of longterm averaged water budget. Hence, the uncertainties in the other datasets would have propagated to the GW data. Please explain how far this will affect the result.
8) Is it safe to assume that the catchment proeprties and the model parameters are stationary in time for such a long time period? How to account for the non-stationarity in catchment properties and model parameters?
9) Please use continous colour bar instead of ranges for presenting ET, rainfall and their differences (figures 5-9). The class ranges are quite large and a continous colour bar will provide more information that these maps I feel. Especially, with the current ET difference maps, I get an impression that there are large differences between the SWAT+ ET and WaPOR ET. A continous colour bar may help to understand this better.
10) Please limit the y-axis value range in figure 10. I think the flow value is maximising at 40000 cubic metre per second. This will help us visualise the temporal variation in the flow better.
11) Hydrological modelling exercises are generally not useful for simulating extreme events such as floods and droughts. Please include your views on how to model extremes under future climate change and land cover change scenarios.
Citation: https://doi.org/10.5194/hess-2023-93-RC2
Celray James Chawanda et al.
Celray James Chawanda et al.
Viewed
HTML | XML | Total | BibTeX | EndNote | |
---|---|---|---|---|---|
426 | 176 | 21 | 623 | 14 | 17 |
- HTML: 426
- PDF: 176
- XML: 21
- Total: 623
- BibTeX: 14
- EndNote: 17
Viewed (geographical distribution)
Country | # | Views | % |
---|
Total: | 0 |
HTML: | 0 |
PDF: | 0 |
XML: | 0 |
- 1