| 英文摘要: | Persistent drought in the Sahel in the 1970s and 1980s was caused by subtle changes in global sea surface temperatures. Now model results show that the direct effect of increasing greenhouse-gas concentrations led to the subsequent recovery. The Sahel is the semi-arid southern shore of the Saharan 'sea of sand'. It holds a special place in climate science, because of the long-standing debate on the causes of persistent drought in the 1970s and 1980s. Since the driest mid-1980s, which include 1984 — the year of the Ethiopian famine made famous by Live Aid — there has been a recovery in rainfall1, provoking further questions on what controls precipitation in the region. Writing in Nature Climate Change, Buwen Dong and Rowan Sutton2 suggest that higher atmospheric concentrations of greenhouse gases, and the consequent atmospheric temperature increase known as the direct effect, were primarily responsible for the recovery. Given the high climatic vulnerability of the region, this study2 is sure to captivate a broader audience, including development practitioners. This presents an opportunity to synthesize current understanding of the dynamics of future climate change in the context of past drought persistence. The twentieth-century evolution of Sahel rainfall — including the decades of anomalously abundant rains that preceded the long-term drought — has been attributed to sea surface temperature (SST) variations. This was conclusively demonstrated only in 20033, freeing Sahelian farmers and pastoralists from blame4. Prior attempts to attribute variations in Sahel rainfall to emissions from industrialization presented arguments for indirect effects, whereby emissions, both greenhouse gases (GHGs) and aerosols, affect rainfall by inducing persistent SST anomalies5. For example, the cooling effect of sulphate aerosols on the surface temperature of the North Atlantic, the moisture source for the West African monsoon, has long been recognized as a key component of late 20th century Sahelian drought6. The generalized warming of the oceans attributed to GHGs, that emerged around 1970, is understood to have exacerbated drought persistence in the 1970s and 1980s3, 7, 8. Comparison to El Niño events9 is illustrative: warming of the oceans locally favours the rising motions in deep convection that cause the air to cool, and water vapour in it to condense and fall in precipitation. At the same time, it raises the bar for the same processes to occur globally, because convection communicates the surface warming to the upper levels of the troposphere. As the upper-tropospheric warming spreads, it stabilizes columns of air remote from the surface warming itself. When this “upped ante for convection” cannot be met, tropical land dries7, as in the case of the Sahel10. In contrast to previous periods of persistently abundant or deficient rainfall, the recent recovery in the Sahel is characterized by increased year-to-year variability1. This variability can be understood by revisiting the two elements implicated in drought — variations in North Atlantic SSTs, and warming of the tropical oceans. Whether due to a decrease in atmospheric aerosol loading prompted by legislation in the US and Europe, or to a recovery in the strength of the Atlantic Meridional Overturning Circulation11, the North Atlantic is now also warming. North Atlantic SSTs that are warmer than the global tropical mean imply that the “upped ante”9 associated with warming tropical oceans can be met by the increased moisture supplied in monsoon flow8. In years when that happens, the Sahel can receive abundant rains once again. But that is not guaranteed year in, year out, due to the large internal variability in the system: there is concern that this year's rainy season may be delayed or deficient precisely because the present configuration is that of a relatively cool North Atlantic in the face of a rapidly warming El Niño event in the tropical Pacific (http://go.nature.com/UgPmP8). The work by Dong and Sutton investigates the three factors — SSTs, GHGs and anthropogenic aerosols — that may contribute to precipitation changes. Based on simulations run with a single model, and testing these in combination and alone, they come to the conclusion that the direct effect of GHGs on the climate of the Sahel is sufficient to explain the rainfall recovery since the driest early to mid-1980s. The direct and indirect effects of GHGs cannot be easily separated. In models, the sum of their separate effects is different from that of their simultaneous combination12. Herein lies the uncertainty in regional projections of precipitation change5. Dong and Sutton2 analyse output from one model (HadGEM2), but do these results hold for other models? Comparison of the model used in the study with others included in the Coupled Model Intercomparison Project Phase 5 (CMIP5) confirms that all models tested make the Sahel wetter when subjected to only the direct effect of GHGs, when SSTs are prevented from warming in response12. However, one reason why the model used by Dong and Sutton emphasizes the direct effect of GHGs may be its underperformance in reproducing the effect of historical SST on Sahel rainfall. Testing the version of this model made available in CMIP5 finds it does not reproduce the conclusion of SST influence reported in the literature (see refs 3,13, for example) that provides the scientific basis for seasonal prediction. The correlation between modelled precipitation in a single simulation run with the atmospheric component of HadGEM2 over observed SST, and observed precipitation (from the Global Precipitation Climatology Project, a blend of satellite retrievals and station observations) is not significant for 1979–2008 (Fig. 1). In comparison, the ensemble mean (5 simulations) from another CMIP5 model, GFDL-CM3, which is the current generation of the model used in ref. 13 and 7, shows a robust SST influence.
|
![Sahel precipitation anomalies (10-20[deg] N, 20[deg] W-40[deg] E) for 1979-2008.](http://www.nature.com/nclimate/journal/v5/n8/images_article/nclimate2739-f1.jpg)