Knowledge on the distribution of intense convective precipitation is useful for mitigation of natural disaster caused by precipitation. An increase of natural disasters due to intense convective precipitation for the past three decades has been reported . The vertical distribution of intense convective precipitation can be investigated through the presence of convective clouds such as cumulonimbus (Cb) clouds. Such vertical structure or distribution is inferred from radar reflectivity factor (dBZ) given by weather or precipitation radar (PR) such as Tropical Rainfall Measuring Mission (TRMM- PR) . The use of TRMM-PR data to study vertical distribution of intense convective clouds has been reported for tropical areas . It was found that average vertical profile of cumulonimbus tower (CbT) and Intense Convective Clouds (ICC) over ocean are similar for all locations, whereas they are different over land areas. However, Indonesia was categorized as oceanic intense convection in this study.
Indonesia is not a maritime or continental area because it consists of large landmass and ocean, so that it is commonly called “maritime-continent” area. Therefore, characteristics of intense convective cloud in Indonesia may be different from maritime and continental areas. Study on the vertical profile of intense convective in Indonesia is limited to Kototabang, west Sumatra [4-6]. Recently, a study was conducted to study seasonal and diurnal variations of vertical distribution of precipitation over Indonesia using the TRMM-PR data . However, this study did not investigate the vertical profile of intense convective precipitation. Furthere advances in understanding must be made in order to improve our overall understanding of vertical distribution of intense convective clouds over Indonesia.
2. Materials and Methods
In this study the primary data are TRMM PR 2A25 attenuation correction Z (version 7), for a 10 year period (1998-2007). Data have vertical resolution of 250 m and total 80 levels in the vertical. Detail information about TRMM PR can be found in Kummerow et al. . As in a previous study , we only used the TRMM profiles with scan angle of less than 7 on either sides of nadir.
Intense convective clouds is classified into two types of convective cells following the classification method of Bhat and Kumar , namely, Cumulonimbus Tower (CbT) and Intense Convective Cloud (ICC). The cloud is assumed as CbT when the Z profiles have 20 dBZ at the altitude of 12 km and this value lasts up to 3 km (9 km depth). Moreover, the ICC type has Z value of 30 and 41 dBZ at 8 and 3 km altitude, respectively. Occurence frequency of each cloud type is contured on 1 1 boxes and 15 locations are selected as region of interest. The regional differences in the vertical distribution of intense convective are depicted in the form of normalized frequency by altitude diagram (CFAD) .
Fig. 1 shows the number of Z profiles for CbT, ICC8 and ICC3 over Indonesia with the grid of contour plot being 1 x 1 . It is found that intense convective cloud is frequently observed over large landmass such as Jawa, Sumatra, Kalimantan and Irian Jaya.
From Fig. 1, we selected 15 locations for case study with more detailed analysis. Table 1 summaries the position of all locations. It is found that Jawa island (Jav) has the largest number of intense convective data, followed by Sumatra2 (SM2), Strait of Malacca (SMal), Kalimantan (Kal), Coastal Irian Jaya (CIJ) and Irian Jaya (Irj). On the other hand, small number of intense convective clouds data is observed in Indian Ocean, Sulawesi and Strait of Karimata. This finding reinforces previous study about the role of large landmass in initiating the intense convective clouds in Indonesia . Number of CbT cloud is always larger than that of ICC8 and ICC3. However, number of ICC8 and ICC3 fluctuates; ICC8 number is larger than ICC3 at one location but it can be smaller at other locations.
Fig. 2 shows the CFAD (countoured frequency by altitude diagram) of CbT and ICCs. The lines in CFAD indicate percentiles of 10% to 90% and thick lines indicate 50% percentile which is mean value of data. Below the melting layer ( 4 km), CbT radar reflectivity shows downward increasing (DI) pattern over coastal areas of Sumatra (CSm1, CSm2, CSm3) and Indian Ocean (Ioc), whereas downward decreasing pattern are osberved in other areas. The DI pattern indicated a significant raindrop growth at below the melting layer ( 4 km). On the other hand, ICC8 and ICC3 radar reflectivity over all locations show downward decreasing (DD) toward the surface.
Fig. 3 shows mean vertical distribution of CbT, ICC8, and ICC3 over different locations. Intense convective of CbT (Fig. 3a) shows larger Z in Sumatra2 (SM2), followed by Kalimantan (Kal), Sumatra3 (SM3) and Jawa (JAV). For ICC8 (Fig. 3b) strong intense convective is observed in Sumatra (SM1, SM2, SM3), Jawa (Jav) and Kalimantan (Kal). Moreover, for ICC3 (Fig. 3c), strong intense convective is observed in Sumatra2 (SM2) and followed by Kalimantan (Kal), Sumatra (SM1, SM3) and Jawa (Jav). Intense convective clouds shows weaker or smaller Z in Indian Ocean (IOC) and Pacific Ocean (POC).
The mean vertical profile in Indonesia (Fig. 3) shows significant regional variation. This variation may be influenced by many factors such as topography, global and local atmospheric circulation .
Intense convective precipitation in Indonesia both CbT, ICC8 and ICC3 are dominant over large island such as Sumatra, Kalimantan, Jawa and Irian Jaya. Besides larger data number, intense convective clouds over large island is also much stronger than over open ocean and coastal areas. For CbT, below the melting layer ( 4 km), radar reflectivity over coastal areas of Sumatra and Indian Ocean shows downward increasing pattern while downward decreasing were observed over other locations. For ICC3 and ICC8, below the melting layer ( 4 km), radar reflectivity over all locations shows downward decreasing pattern.
This study was supported by the 2017 and 2018 International Joint Collaboration and Scientific Publication grant from the Ministry of Research, Technology and Higher Education of the Republic of Indonesia (Contract No. 02/UN.16.1.17/PP.KLN/LPPM/2017 and 050/SP2H/LT/DRPM/2018). The authors thank to National Aeronautics and Space Administration (NASA) for providing the TRMM-PR data.