Scholarlink

<p>Rapid and unplanned urbanization in Raipur, Chhattisgarh, has exacerbated urban flooding, particularly during intense rainfall events exceeding 50&#xa0;mm. This study compares five disaggregation methods to break down daily rainfall data into hourly data using the Indian Monsoon Data Assimilation and Analysis (IMDAA) dataset: The Support Vector Machine (SVM), the Method of Fraction (MOF), the Artificial Neural Network (ANN), and the Indian Meteorological Department’s 1/3rd rule (IMD 1/3rd). The Random Forest method demonstrated the highest accuracy in replicating observed hourly rainfall patterns. Using CMIP6 climate data, we generated hourly rainfall forecasts for Raipur from 2021 to 2035 using this method. However, projected monsoon and summer rainfall durations are to increase under the SSP5-8.5 and SSP2-4.5 scenarios, with monsoon rainfall exhibiting periodic extremes. Conversely, we expect a decrease in summer rainfall and an early peak but gradual decrease in winter rainfall. Shorter-duration rainfall events may experience occasional spikes, while longer-duration events, especially those with a 100-year return period, are projected to increase in intensity, heightening flood risks. Notable spikes are projected in 2023, 2027, 2031, and 2033, underscoring the urgent need for enhanced flood management strategies in Raipur. </p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Temporal Disaggregation and Short-Term Forecasting of Hourly Rainfall Data: A Case Study of Raipur City Chhattisgarh, India

  • 期刊论文

  • 发表日期：

    2025年5月27日

查看全文
  • Gulshan Patel,
  • Shashikant Verma,
  • Mani Kant Verma

摘要

Rapid and unplanned urbanization in Raipur, Chhattisgarh, has exacerbated urban flooding, particularly during intense rainfall events exceeding 50 mm. This study compares five disaggregation methods to break down daily rainfall data into hourly data using the Indian Monsoon Data Assimilation and Analysis (IMDAA) dataset: The Support Vector Machine (SVM), the Method of Fraction (MOF), the Artificial Neural Network (ANN), and the Indian Meteorological Department’s 1/3rd rule (IMD 1/3rd). The Random Forest method demonstrated the highest accuracy in replicating observed hourly rainfall patterns. Using CMIP6 climate data, we generated hourly rainfall forecasts for Raipur from 2021 to 2035 using this method. However, projected monsoon and summer rainfall durations are to increase under the SSP5-8.5 and SSP2-4.5 scenarios, with monsoon rainfall exhibiting periodic extremes. Conversely, we expect a decrease in summer rainfall and an early peak but gradual decrease in winter rainfall. Shorter-duration rainfall events may experience occasional spikes, while longer-duration events, especially those with a 100-year return period, are projected to increase in intensity, heightening flood risks. Notable spikes are projected in 2023, 2027, 2031, and 2033, underscoring the urgent need for enhanced flood management strategies in Raipur.