期刊
METEOROLOGY AND ATMOSPHERIC PHYSICS
卷 129, 期 2, 页码 173-186出版社
SPRINGER WIEN
DOI: 10.1007/s00703-016-0466-9
关键词
-
资金
- U.K. Department for International Development (DFID)-Met Office Hadley Centre Climate Science Research Partnership (CSRP) program [L0830]
- DfID/NERC HyCRISTAL project [NE/M019985/1]
- Livestock-Climate Change Collaborative Support Research Program (LCC-CSRP) (based at Colorado University, USA) [G-9650-12]
- Addis Ababa University
- Natural Environment Research Council [NE/M019985/1, NE/M02038X/1] Funding Source: researchfish
- NERC [NE/M02038X/1, NE/M019985/1] Funding Source: UKRI
Rainfall variability in Ethiopia has significant effects on rainfed agriculture and hydropower, so understanding its association with slowly varying global sea surface temperatures (SSTs) is potentially important for prediction purposes. We provide an overview of the seasonality and spatial variability of these teleconnections across Ethiopia. A quasi-objective method is employed to define coherent seasons and regions of SST-rainfall teleconnections for Ethiopia. We identify three seasons (March-May, MAM; July-September, JAS; and October-November, ON), which are similar to those defined by climatological rainfall totals. We also identify three new regions (Central and western Ethiopia, CW-Ethiopia; Southern Ethiopia, S-Ethiopia; and Northeast Ethiopia, NE-Ethiopia) that are complementary to those previously defined here based on distinct SST-rainfall teleconnections that are useful when predicting interannual anomalies. JAS rainfall over CW-Ethiopia is negatively associated with SSTs over the equatorial east Pacific and Indian Ocean. New regional detail is added to that previously found for the whole of East Africa, in particular that ON rainfall over S-Ethiopia is positively associated with equatorial east Pacific SSTs and with the Indian Ocean Dipole (IOD). Also, SST-to-rainfall correlations for other season-regions, and specifically for MAM in all regions, are found to be negligible. The representation of these teleconnections in the HadGEM2 and HadGEM3-GA3.0 coupled climate models shows mixed skill. Both models poorly represent the statistically significant teleconnections, except that HadGEM2 and the low resolution (N96) version of HadGEM3-GA3.0 better represent the association between the IOD and S-Ethiopian ON rainfall. Additionally, both models are able to represent the lack of SST-rainfall correlation in other seasons and other parts of Ethiopia.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据