Scholarlink

<p>Drought is among the most severe natural disasters, with its frequency and intensity increasing due to global climate change. Accurate drought characterization is essential for ensuring environmental sustainability. As precipitation is a key variable in drought assessment, many recent studies utilize precipitation data from global climate models (GCMs). However, inconsistencies among GCM outputs limit the effectiveness of individual models. To address this, multi-model ensemble (MME) approaches combine outputs from multiple GCMs, though traditional methods often rely on linear metrics like Pearson correlation, which fail to capture nonlinear dependencies. This study introduces the standardized dual Ddivergence-correlation (SDDC) index, which incorporates a novel Dual Divergence-Correlation Weighting (DDCW) scheme. The DDCW method integrates distance correlation and divergence-based weighting to effectively capture both nonlinear associations and distributional differences between observed and modeled data. Using precipitation data from 22 CMIP6 GCMs, the DDCW method is compared with Simple Model Averaging (SMA) and a recent Weighted Ensemble (WE) approach. The comparison is performed using quality assessment measures, including the correlation coefficient and mean absolute error (MAE), to evaluate the accuracy and reliability of the projected precipitation outcomes. Results demonstrate that DDCW consistently outperforms over traditional methods,&#xa0;achieving a higher mean correlation (0.5175) compared to SMA (0.4676) and WE (0.5140), and a lower mean MAE (18.458) compared to SMA (19.164) and WE (18.906). For future drought characterization, steady state probabilities (StSP) were computed under three shared socio-economic pathway (SSP) scenarios, revealing that "No Drought" conditions are most probable, while extreme events remain less frequent. This outcome likely reflects the regional hydroclimatic behavior of Pakistan, where projected precipitation increases under CMIP6 models moderate drought severity even in high-emission scenarios.</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

Overcoming global climate model inconsistencies in drought projection with the development of dual divergence correlation weighting scheme over Pakistan

  • 期刊论文

  • 发表日期：

    2026年2月9日

查看全文
  • Mahrukh Yousaf,
  • Amara Farooq,
  • Sadia Qamar,
  • Naim Ahmad,
  • Muhammad Shakeel,
  • Aamina Batool,
  • Zulfiqar Ali,
  • Veysi Kartal

摘要

Drought is among the most severe natural disasters, with its frequency and intensity increasing due to global climate change. Accurate drought characterization is essential for ensuring environmental sustainability. As precipitation is a key variable in drought assessment, many recent studies utilize precipitation data from global climate models (GCMs). However, inconsistencies among GCM outputs limit the effectiveness of individual models. To address this, multi-model ensemble (MME) approaches combine outputs from multiple GCMs, though traditional methods often rely on linear metrics like Pearson correlation, which fail to capture nonlinear dependencies. This study introduces the standardized dual Ddivergence-correlation (SDDC) index, which incorporates a novel Dual Divergence-Correlation Weighting (DDCW) scheme. The DDCW method integrates distance correlation and divergence-based weighting to effectively capture both nonlinear associations and distributional differences between observed and modeled data. Using precipitation data from 22 CMIP6 GCMs, the DDCW method is compared with Simple Model Averaging (SMA) and a recent Weighted Ensemble (WE) approach. The comparison is performed using quality assessment measures, including the correlation coefficient and mean absolute error (MAE), to evaluate the accuracy and reliability of the projected precipitation outcomes. Results demonstrate that DDCW consistently outperforms over traditional methods, achieving a higher mean correlation (0.5175) compared to SMA (0.4676) and WE (0.5140), and a lower mean MAE (18.458) compared to SMA (19.164) and WE (18.906). For future drought characterization, steady state probabilities (StSP) were computed under three shared socio-economic pathway (SSP) scenarios, revealing that "No Drought" conditions are most probable, while extreme events remain less frequent. This outcome likely reflects the regional hydroclimatic behavior of Pakistan, where projected precipitation increases under CMIP6 models moderate drought severity even in high-emission scenarios.