Advancing the characterization of dry spells during the rainy season

Author: Teuku Ferijal

  • Thesis download: available for open access on 18 Jan 2022.

Ferijal, Teuku, 2020 Advancing the characterization of dry spells during the rainy season, Flinders University, College of Science and Engineering

Terms of Use: This electronic version is (or will be) made publicly available by Flinders University in accordance with its open access policy for student theses. Copyright in this thesis remains with the author. You may use this material for uses permitted under the Copyright Act 1968. If you are the owner of any included third party copyright material and/or you believe that any material has been made available without permission of the copyright owner please contact with the details.


Global climate change modifies the global hydrological cycle, including precipitation. The changes that occur in precipitation are mostly a decrease in frequency and an increase in intensity. This results in an increased frequency and duration of dry spells, which aggravates the drought or dryness severity. Drought is commonly estimated using an accumulated precipitation index. However, this approach fails to detect the drought progression at a finer time scale due to the presence of dry spells. With increasing temperatures, dry spells will become an important element of drought development because it escalates more water to evaporate. During the rainy season, an increasing frequency and duration of dry spells would cause severe impact on agriculture, especially regions that depend on rainy season precipitation. Because of high temporal variation in onset and cessation dates of the rainy season, existing drought indices are not appropriate tools for drought estimation during the rainy season. Meanwhile, although dry spell has been used to assess dryness conditions, no study has linked duration and frequency of dry spells to dryness severity. Therefore, this study aims to investigate the changes in dryness level during the rainy season using frequency and duration information of dry spells. The Indonesian Maritime Continent (IMC) was selected as a study area because it is the largest rainy area on the earth and has strong heterogeneous precipitation types.

The first part of this thesis deals with the determination of rainy season onset and cessation. It investigates the usage of the driest period as a starting point for determination of onset and cessation. The driest period is simply the longest period without precipitation or the 14 consecutive days having the least amount of cumulative precipitation in the year. The definition produces a flexible starting time approach, which is different from the commonly used regional fixed starting date that is based on the regional driest month in the year. The results reveal that the proposed method prevents the occurrence of premature cessation dates and keeps the cessation dates within the corresponding year. The gap between the first day of the driest period and the onset date has a significant linear relationship with the onset date at most stations over the region, particularly with the onset of anomalous accumulation.

The second part of the thesis offers a new approach of using dry spell frequency and duration to characterise the dryness severity during the rainy season. It analyses trends in meteorological drought severity during the rainy season. The dry spell index was proposed and compared with Standardised Precipitation Index (SPI). The Mann-Kendall trend test was used to investigate the occurrence of a trend within the indices. The results suggested that increasing drought is mostly found in the southern hemisphere region, which is most likely caused by the increase in dry spell parameters such as number of dry days, contribution of long dry spells, and duration of extreme dry spells. For the 1982 to 2014 period, the DSI identified 40% more stations with increasing drought trend than that SPI identified.

The last part aims to evaluate the spatial and temporal variability in dryness severity during the rainy season. DSI was integrated with an evaporation rate to evaluate the dryness severity. The analysis uses TRMM 3B42 V7 data for daily precipitation and GLEAMS remote sensing derived data for the daily evaporation rates. Resulting DSI trends were evaluated using the Mann-Kendall test with 90% confidence level. The results demonstrate significant increasing dryness in the southern part of IMC and a significant decrease in dryness near the equator. A similar pattern was also observed in total precipitation, which indicates that the precipitation frequency reduces, and intensity increases. The identified increasing dryness severity, especially in Java, has important consequences for agricultural activities and water resources management.

The overall results of this thesis suggest there is a general tendency, supported by the precipitation station and remote sensing-based data that the southern hemisphere regions are experiencing an increase in dryness in the rainy season. The frequency and intensity of precipitation have changed toward more extreme conditions resulting in more and longer dry spells. Future research should focus on defining the driest period in real time for improving rainy season onset prediction. It is also important to standardise the dryness severity resulting from DSI, in order to obtain an index that is globally applicable.

Keywords: rainy season, drought, dry spell, dry spell index, Indonesian Maritime Continent

Subject: Meteorology and Oceanography thesis

Thesis type: Doctor of Philosophy
Completed: 2020
School: College of Science and Engineering
Supervisor: Okke Batelaan