Scholarlink

<p>Drought is one of the most significant repercussions of climate change, globally impacting food security and ecological productivity. Bangladesh has encountered several droughts over the&#xa0;past few decades, leading to substantial economic and environmental consequences. The northwestern region of the country is the most vulnerable to drought because of its geographical location and semi-arid climate. Given the increasing&#xa0;frequency and severity of droughts, rapid&#xa0;and reliable drought information is essential&#xa0;for maintaining agroecological production and ensuring food security. Using the Standardized Precipitation Index (SPI) and three time-series models– Auto Regressive Moving Average (ARMA), PROPHET, and a hybrid ARMA-Generalized Autoregressive Conditional Heteroskedasticity (ARMA-GARCH)–we assessed drought trends across five meteorological stations (Bogura, Dinajpur, Ishwardi, Rajshahi, and Rangpur) in the north-western region of Bangladesh for the period 1980–2019. Results show that SPI trends were significant for the Dinajpur and Ishwardi stations but insignificant for the other three stations (Bogura, Rajshahi, and Rangpur). Among the three models, the hybrid model (ARMA-GARCH) outperformed the individual models (ARMA and PROPHET), suggesting that the ARMA-GARCH model could be effectively utilized to predict droughts as it showed higher accuracy than the individual models. This study provides empirical evidence of (i) an intensification of drier climates in the north-western region of the country over the past 40&#xa0;years, which has practical implications for introducing climate-adaptive practices in agriculture and other livelihood sectors, and (ii) the better performance of a hybrid model compared to individual models in predicting drought, which is of great significance for government decision-making.</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

Hybrid model outperformed individual models in predicting droughts in a semi-arid region of Bangladesh

  • 期刊论文

  • 发表日期：

    2026年4月16日

查看全文
  • Shahed Mahmud,
  • Shihab A. Shahriar,
  • Md. Lokman Hossain,
  • Rashik Islam,
  • Ashim C. Das,
  • Mohammed Abdus Salam

摘要

Drought is one of the most significant repercussions of climate change, globally impacting food security and ecological productivity. Bangladesh has encountered several droughts over the past few decades, leading to substantial economic and environmental consequences. The northwestern region of the country is the most vulnerable to drought because of its geographical location and semi-arid climate. Given the increasing frequency and severity of droughts, rapid and reliable drought information is essential for maintaining agroecological production and ensuring food security. Using the Standardized Precipitation Index (SPI) and three time-series models– Auto Regressive Moving Average (ARMA), PROPHET, and a hybrid ARMA-Generalized Autoregressive Conditional Heteroskedasticity (ARMA-GARCH)–we assessed drought trends across five meteorological stations (Bogura, Dinajpur, Ishwardi, Rajshahi, and Rangpur) in the north-western region of Bangladesh for the period 1980–2019. Results show that SPI trends were significant for the Dinajpur and Ishwardi stations but insignificant for the other three stations (Bogura, Rajshahi, and Rangpur). Among the three models, the hybrid model (ARMA-GARCH) outperformed the individual models (ARMA and PROPHET), suggesting that the ARMA-GARCH model could be effectively utilized to predict droughts as it showed higher accuracy than the individual models. This study provides empirical evidence of (i) an intensification of drier climates in the north-western region of the country over the past 40 years, which has practical implications for introducing climate-adaptive practices in agriculture and other livelihood sectors, and (ii) the better performance of a hybrid model compared to individual models in predicting drought, which is of great significance for government decision-making.