In this post, I will be presenting some research findings on water projections in the SRB under climate change, predicted using hydrological and climate models. As there is no single “correct” hydrological or climate model, it is important to compare the results of different model predictions, as it sheds light on uncertainties within those projections (Jiang et al., 2007; Cornelissen et al., 2013).
In terms of river discharge, Mbaye et al. (2015) used the hydrological model “MPI-HM” and forced it with a Regional Climate Model (RCM) “REMO”, to predict changes in the SRB by 2017-2100 compared to 1971-2000 under the emission pathways RCP 4.5 and 8.5. It is projected that there will be a general decrease in mean monthly discharge as well as the 10th (low flow) and 90th percentile discharge (high flow).
 |
Projected mean monthly changes in discharge and 10th and 90th percentile discharge in the SRB (Mbaye et al., 2015).
|
In Bodian et al. (
2018), the GR4J hydrological model was calibrated to predict mean annual discharge change by 2050 compared to 1971-2000. Similarly, RCP 4.5 and 8.5 were chosen, but 6 General Circulation Models (GCMs) were used instead of RCMs. All except one GCM predicted that there will be either no change or a decrease in annual discharge. |
| Projected change in mean annual discharge in the SRB under (a) RCP 4.5 and (b) RCP 8.5 (Bodian et al., 2018). |
Finally, Rameshwaran et al. (
2021) employed the “HMF-WA” model, alongside CMIP5 GCM ensembles and RCP 2.6 and 8.5 to predict change in river flow in Senegal in the 2050s and 2080s compared to 1970-1999. Under RCP 2.6, mean monthly flows and median peak flows will remain relatively constant, however, large reductions in flows are predicted under RCP 8.5, especially in July to October.
Hydrological models can simulate other hydrological characteristics as well. For example, Mbaye et al. (2015) also predicted changes in soil moisture, which is projected to decline in the northern to central part of the basin, with a severe decrease in northern SRB especially under RCP 8.5.
Uncertainties in climate change projections
While there is a general consensus that river flow is going to decline in the SRB under climate change, particularly under high emission scenarios (RCP 8.5), the magnitude of change varies and there are also some discrepancies in predictions. This highlights the “cascade of uncertainty” when using climate and hydrological models, referring to the uncertainties in models caused by uncertain emission trajectories, errors due to low spatial resolution of GCMs, downscaling of GCMs to RCMs, and in hydrological models (Vaghefi et al., 2019). Therefore, while models are helpful in helping us understand future water trends, it is also necessary to interpret the results cautiously and be aware of the uncertainties and limitations of the predictions.
Hi Selina! I loved this post on water projections - I learnt so much! In regards to the uncertainty in climate change projections, do you believe we are close to more accurate models in the near future? or is the technology still a while away?
ReplyDeleteHello Cristina, I'm glad you liked my post! I think models can improve in accuracy, since researchers are still constantly trying to minimize uncertainties in models, for example CMIP is actively advancing to produce new GCMs. Computational power is also increasing rapidly, allowing more complex models to run. Technologies such as remote sensing can also help collect observational data globally at much finer spatial resolution, which is helpful for reducing uncertainties in downscaling of GCMs.
Delete