Identification of Extreme Rainfall Events for the Period 1970–2019 in Sri Lanka Using Percentile-Based Analysis and Its Projections for 2100 for the Emission Scenarios of RCP 4.5 and 8.5

2021 
The hydrological hazards associated with extreme weather events have increased globally over the past few decades leading to floods, landslides, drought, heat waves etc. Major reason for such hazards is an increase in the amplitude and frequency of weather extremes. Sri Lanka, ranked second among countries most affected by extreme weather events in 2017 is expected to see a 1.2% annual GDP loss by 2050 due to climate change. Identification of trends in extreme climate events and projecting them into the future will be important to find adaptation strategies in order to mitigate the disasters associated with extreme weather. Therefore, this study is focused to identify the change in extreme rainfall events during last fifty years by comparison of number of heavy and very heavy rainfall events for the period 1970–1979, 1980–1989, 990–1999, 2000–2009 and 2010–2019. Daily rainfall data for 16 rainfall stations in the period 1970–2019 is used and percentile based extreme rainfall events is analyzed. The results show an increasing trend of extreme rainfall events in Sri Lanka at some places during the past two decades, but not at a significant level. Analysis was further extended to identify future extreme events, using the NEX-NASA CMIP5 daily rainfall data for the moderate emission scenario (RCP 4.5) and high emission scenario (RCP 8.5) for the three (3) climatic zones in Sri Lanka and for the four monsoon seasons. Number of extreme events above the 90th and 95th percentiles for the period 2005–2100 was considered for the analysis. Base period used was 1975–2005. Results show a significant increasing trend for heavy and very heavy rainfall events in the three climatic zones for both emission scenarios. However, the increasing trend is not much significant for some monsoon seasons in Sri Lanka.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    13
    References
    0
    Citations
    NaN
    KQI
    []