Liquid and ice water content in clouds and their variability with temperature in Africa based on ERA-Interim, JRA-55, MERRA and ISCCP

Abstract

This study presents seasonal vertical structure from 1000 to 100 hPa of liquid water content (LWC) and ice water content (IWC) and investigates their variability with temperature in Africa from climatological perspective. Both LWC and IWC may increase or decrease with temperature by season and by region. We compared vertical distribution of LWC and IWC from ERA-Interim, JRA-55 and MERRA reanalysis to the distribution of cloud types from ISCCP and we found that cloud type with highest liquid content is Stratocumulus (Sc) in JJA for both East of Africa (13.3 %) and West of Africa (19.3 %) while Cumulus (Cu) in MAM for both South of Africa (8.6 %) and North of Africa (13.5 %). Cirrus (Ci) is the most frequent cloud with ice with 24.6 % presence in East Africa, 20.4 % in West Africa, 7.2 % in North Africa during MAM and 21.3 % in South Africa during SON. The LWC exists below 400 hPa for both ERA-I and JRA-55, whereas below 200 hPa based on MERRA in all regions of Africa. The IWC exists above 600 hPa for both East and West Africa, above 800 hPa in South Africa and above 900 hPa in North Africa for both ERA-I and JRA-55, whereas above 500 hPa in both East and West of Africa, above 800 hPa in South of Africa and above 850 hPa in North of Africa based on MERRA. The ERA-I, JRA-55 and MERRA are in agreement in low and mid troposphere while in upper troposphere only ERA-I and JRA-55 are in agreement. The MERRA disagrees with ERA-I and JRA-55 in the upper troposphere. The difference of LWC and IWC seasonal vertical distribution in Africa is due to different climatic features between tropic regions and subtropic regions. All three reanalyses show the LWC and IWC between −22.7 and 1.6 °C over Africa. The coexistence temperature between LWC and IWC may be useful to estimate the temperature at which there are the supercooled liquid water and mixed clouds. The ERA-I and JRA-55 may be good proxies in precipitation and cloud monitoring in regions with a lack of in situ observations in Africa. Our results reveal that ISCCP and the two reanalyses (ERA-I and JRA-55) are more reliable for characterizing cloud types and the vertical profile of LWC and IWC climatology, respectively.