Rainfall is a very important weather and climate parameter that affects social and economic activities in Djibouti. This leads to droughts, floods and humanitarian disasters over the country. The rainfall climatology of the country has some changes which need to be taken into account while planning for economic activities such as civil and structural engineering. The whole country receives significant rains in the months of (October – December) and (March – April) as known except for the southern parts of the country. The southern part of the country gets its rains in the months of (June to August). Djibouti is located in the Eastern Africa it is believed that OND rainfall over the Eastern Africa is strongly caused by IOD and ENSO events. The roles of Indian-Pacific Ocean SSTs in climate variability over Eastern Africa have been discussed in a number of studies, and since most of previous studies focused on the entire region of eastern Africa. But less concerns remained on the analysis of specific countries like Djibouti. To explain it is OND season over Djibouti strongly caused by IOD and ENSO? This study will try to cover this gap by study inter-annual variability of October-December rainfall season over Djibouti.
The study inter-annual variability of OND rainfall over Djibouti focuses on analyzing by using different methods such as EOF, composite analysis and correlation analysis, the study tried to assess the spatial and temporal distribution of OND season rainfall over Djibouti and its associate circulating weather system covering a period of 1971 – 2010. The first EOF mode analysis is referred to be a dominant mode and was taken into consideration in this study. The identification of dry and wet years was based on the interannual standardized rainfall anomaly, we have identified the five wettest (1972, 1982, 1994, 1997, 2000) and three driest years (1974, 1975, 1978). The Mann–Kendall test statistics for abrupt change in rainfall was using there was observed abrupt change throughout the study period occurred 1979. The composite analysis of the wet and dry years with respect to the different variables including, wind, velocity potential/divergence (convergence) reveals that the wet (dry) years were associated with convergence (divergence) at 850hpa and
divergence (convergence) at the upper level (200hpa), implying rising (sinking) motion, especially over the study area and western Indian Ocean. During wet years the sea level pressure is weakens over the West Indian Ocean and the eastern African coast due to the presence of low pressure in that region. Because of low pressure at the coast (Djibouti coastal area), the area experience a lot of activities (convection) favored for rain formation. The results of correlation analysis show that there exists a significant correlation between OND rainfall, Nino 3.4 and DMI over the country. The correlation coefficient of OND rainfall and DMI is 0.66 while that of Nino 3.4 and OND rainfall is 0.51. It was found that IOD exerts significant influence on Djibouti OND rainfall such that rainfall is enhanced (suppressed) during warm cold events in the events Indian Ocean dipole. Although IOD is the dominant mode over the country but Djibouti’s rainfall pattern seemed to be stronger when the large positive IOD indices matched with El Niño events. This these aims to investigate how Indo-Pacific Oceans modulate Djibouti rainfall and provide us with a physical basis to consider IOD as a predictor for variations of the OND season. It was found significant relationship between the anomalous SST in the Indian Ocean and enhanced rainfall intensity over Djibouti especially the coastal, south-eastern and center parts of the country where the positive correlation. Therefore the previous climate events must be well understood so as to make reliable and accurate forecasts that will enable to improve planning and management of climate sensitive activities.
Notes: If you wish to download the full text, please contact the library(at)wmo.int (Please replace (at) by @).
Format: Digital (Available online for logged-in users)