Scholarlink

<p>GCMs play a vital role in projecting multiple aspects of complex drought phenomena. However, relying on simulation from single GCM is not advised due to inherent uncertainties. An ensemble projection from multiple GCMs can help to minimize these uncertainties. Nonetheless, the inclusion of inefficient GCMs in the ensemble may limit the accuracy of the future projections. Therefore, selecting an appropriate set of GCMs for the entire region requires spatial ranking under a suitable statistical framework as a first step. In this study, we have utilized CMIM for GCMs ranking at each grid station. Based on the net strength, a subset of nine top-performing GCMs are then selected for the entire study region. To address multiple aspects of data uncertainties, we employed five different techniques from various statistical pools in the development of weighted MMEs. These pools include regression-based, geometric-based and BMA. During the evaluation phase of MMEs, based on the EDISO metric, the CLSE MME outperformed the other MMEs, exhibiting a minimum EDISO value of 1.098. The superiority of CLSE is further evidenced by Kling-Gupta Efficiency (KGE) and Diebold-Mariano test. Furthermore, based on CLSE weights, we proposed an SPI-based drought index with a variation on the classical Gamma-based approach, named MLMSDI. The proposed index demonstrated its effectiveness in assessing drought across various SSPs and seven-time scales. The study considered monthly precipitation data from 22 GCMs under CMIP6 for the historical period 1950–2014. For future drought projections, three SSPs—SSP1-2.6, SSP2-4.5, and SSP5-8.5—are used for the period 2015–2100, covering 28 locations across Punjab Province, Pakistan.</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

Optimizing GCM ensemble selection and weighted MME development for improved drought projection under global climate models simulations

  • 期刊论文

  • 发表日期：

    2026年3月26日

查看全文
  • Muhammad Shakeel,
  • Zulfiqar Ali,
  • Hussnain Abbas,
  • Muhammad Mohsin,
  • Mrim M. Alnfiai,
  • Naim Ahmad,
  • Rizwan Niaz

摘要

GCMs play a vital role in projecting multiple aspects of complex drought phenomena. However, relying on simulation from single GCM is not advised due to inherent uncertainties. An ensemble projection from multiple GCMs can help to minimize these uncertainties. Nonetheless, the inclusion of inefficient GCMs in the ensemble may limit the accuracy of the future projections. Therefore, selecting an appropriate set of GCMs for the entire region requires spatial ranking under a suitable statistical framework as a first step. In this study, we have utilized CMIM for GCMs ranking at each grid station. Based on the net strength, a subset of nine top-performing GCMs are then selected for the entire study region. To address multiple aspects of data uncertainties, we employed five different techniques from various statistical pools in the development of weighted MMEs. These pools include regression-based, geometric-based and BMA. During the evaluation phase of MMEs, based on the EDISO metric, the CLSE MME outperformed the other MMEs, exhibiting a minimum EDISO value of 1.098. The superiority of CLSE is further evidenced by Kling-Gupta Efficiency (KGE) and Diebold-Mariano test. Furthermore, based on CLSE weights, we proposed an SPI-based drought index with a variation on the classical Gamma-based approach, named MLMSDI. The proposed index demonstrated its effectiveness in assessing drought across various SSPs and seven-time scales. The study considered monthly precipitation data from 22 GCMs under CMIP6 for the historical period 1950–2014. For future drought projections, three SSPs—SSP1-2.6, SSP2-4.5, and SSP5-8.5—are used for the period 2015–2100, covering 28 locations across Punjab Province, Pakistan.