Comment on “Feedbacks on convection from an African wetland” by Christopher Taylor in Geophysical Research Letters (VOL. 37, L05406, doi:10.1029/2009GL041652, 2010)
Faisal Hossain, Tennessee Technological University
Chris Taylor presents a fascinating paper in Geophysical Research letters titled “Feedbacks on convection from an African wetland” (VOL. 37, L05406, doi:10.1029/2009GL041652, 2010). Using satellite thermal datasets spanning 24 years, the dynamics of convective cloud cover is correlated to wetland size in the inner delta of the Niger river during the wet season. Taylor’s analysis shows that the daytime initiation of convective storms actually doubles during periods of inundation of the wetland during August to September. This appears to lend a lot of credibility to the “wetland breeze effect” (similar to the lake effect) where the inundation leads to the formation of larger scale and long-lived convective systems that travel several hundreds of kilometers west. Consequently this effect leads to more rainfall and a wet Sahel that sustains the wetland in the long-run. The whole study epitomizes a classic case of positive feedback and a water cycle in action in the West African region. The societal implications of this work go far beyond the feedback mechanisms. Taylor points out that his newly observed wetland effect raises the question of how humans will modify rainfall patterns by artificial impoundments upstream. If more dams are built in the upstream region of the Niger river, the flow to the wetland will decrease resulting in less inundation and less initiation of convection. Consequently, this may mean less rainfall to keep the wet Sahel. In the long run, this will have serious implications of the water resources management and the sustainability of the artificial reservoir itself.
“There have been a number of previous studies looking at the impact of surface hydrological conditions on local rainfall. This study provides observational evidence that a wet surface feature can influence convection over a much larger region. Such a scale interaction is likely affected by the conditions prevalent in West Africa, where long‐lived travelling storms produce the vast majority of rainfall. Similar up‐scaling may affect the operation of soil moisture ‐ rainfall feedbacks in the absence of lateral redistribution of the water via the river network.”
Recently, Roger and I have been trying to understand how large dams, their impoundments and the associated land use change may have altered heavy rainfall (convective) across the globe and the implications that may have for flood frequency alteration and dam safety. From a data-base exercise, we have seen trends prominent in arid and semi-arid regions (see Hossain et al., 2010 – Dam Safety Effects due to Human Alteration of Extreme Precipitation, Water Resources Research, 46, W03301, (DOI:10.1029/2009WR007704) where the 99th percentile of rainfall has been observed to have increased anywhere from 4-20% per year in the post-dam period. In another related work (On the Empirical Relationship between the Presence of Large Dams and the Alteration in Extreme Precipitation, Natural Hazards Review, DOI: 10.1061/(ASCE)NH.1527-6996.0000013), a case study for Albacete in Southern Spain shows the frequency of heavy rainfall having experienced a marked increase after the construction of the last large reservoir in 1983. The need for atmospheric modeling to understand the impact on rainfall by large dams was mentioned to understand the mechanism behind the observed changes to rainfall patterns. This is in line with Taylor’s suggested implications due to artificial impoundments.
To further quote-
“…they suggest that the construction of a major hydroelectric dam proposed in Guinea would produce an additional decrease of 33%, resulting in a 48% reduction in flooded area during September. Such a change would significantly reduce the window in the seasonal cycle when the wetland can influence rainfall. This raises the possibility that major upstream hydroelectric schemes on the Niger could reduce rainfall at the regional scale, a suggestion which would benefit from further examination with an atmospheric model.”
Taylor’s study is perhaps the first study to look closely at the ‘climatology’ of convective initiations as a function of wetland size and evaporation and track them with the use of satellite thermal imagery. We believe his work will be a pathfinder to achieving a better understanding of what large dams and their associated land use may have done over the years to the heavy flood-inducing storm events.