Scholarlink

<p>The lack of precise temperature data impedes the hydro-climatological assessment for Afghanistan. This study developed high-resolution (0.08° × 0.08°) maximum and minimum temperature datasets at monthly scale for Afghanistan for the span of 60&#xa0;years (1961–2020). Initially the gaps in observed data were filled by the linear regression technique using the nearby stations and gridded datasets (CRU, ERA5, and TerraClimate). Subsequently, the gap-filled data were interpolated at 0.08° grid resolution using Ordinary Kriging (OK) spatial interpolation technique to develop a reference observed gridded product. Several gauge-based and reanalysis datasets (e.g. CRU, ERA5, PGFV2.1, NCEP/NCAR Reanalysis1, and TerraClimate) underwent accuracy assessment against the reference datasets developed in this study. The accuracy assessment was carried out using the commonly used statistical metrics such as KGE (Kling Gupta Efficiency), Coefficient of Determination (R<sup>2</sup>), and MAE (Mean Absolute Errors). The accuracy of linear scaling and quantile mapping methods of bias correction was also assessed, and the better method was used to develop the reference datasets. The developed products were analyzed for the spatio-temporal annual and seasonal trends using the Man Kendel and Sen’s Slope estimator methods. The results showed CRU as the most accurate dataset for Afghanistan. Furthermore, linear scaling outperformed quantile mapping as the average R<sup>2</sup>, and KGE increased for maximum temperature from 0.93 and 0.33 to 0.94 and 0.95, while MAE decreased from 3.84 to 1.15. While for minimum temperature, the average R<sup>2</sup> and KGE increased from 0.84 and 0.21 to 0.86 and 0.89, while MAE decreased from 3.48 to 1.17. Moreover, the findings of the annual trend showed that both the maximum and minimum temperatures had increasing trends. The seasonal trend analysis showed an increase in winter (DJF), spring (MAM), autumn (SON), and summer (JJA) for maximum temperature, with the exception of the east as well as southwest regions, which comprise Nangarhar, Kunar, Laghman, Nuristan, as well as the northern regions of Paktika. While for minimum temperature, seasonal trend analysis shows an increasing trend for all seasons, such as winter (DJF), spring (MAM), summer (JJA), and autumn (SON).</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

High-resolution monthly gridded temperature dataset development and trend analysis across Afghanistan: a spatio-temporal approach

  • 期刊论文

  • 发表日期：

    2025年5月5日

查看全文
  • Maghfoorullah Tasal,
  • Shakil Ahmad,
  • Muhammad Azmat,
  • Mohammad Uzair Rahil,
  • Khalil Ahmad

摘要

The lack of precise temperature data impedes the hydro-climatological assessment for Afghanistan. This study developed high-resolution (0.08° × 0.08°) maximum and minimum temperature datasets at monthly scale for Afghanistan for the span of 60 years (1961–2020). Initially the gaps in observed data were filled by the linear regression technique using the nearby stations and gridded datasets (CRU, ERA5, and TerraClimate). Subsequently, the gap-filled data were interpolated at 0.08° grid resolution using Ordinary Kriging (OK) spatial interpolation technique to develop a reference observed gridded product. Several gauge-based and reanalysis datasets (e.g. CRU, ERA5, PGFV2.1, NCEP/NCAR Reanalysis1, and TerraClimate) underwent accuracy assessment against the reference datasets developed in this study. The accuracy assessment was carried out using the commonly used statistical metrics such as KGE (Kling Gupta Efficiency), Coefficient of Determination (R2), and MAE (Mean Absolute Errors). The accuracy of linear scaling and quantile mapping methods of bias correction was also assessed, and the better method was used to develop the reference datasets. The developed products were analyzed for the spatio-temporal annual and seasonal trends using the Man Kendel and Sen’s Slope estimator methods. The results showed CRU as the most accurate dataset for Afghanistan. Furthermore, linear scaling outperformed quantile mapping as the average R2, and KGE increased for maximum temperature from 0.93 and 0.33 to 0.94 and 0.95, while MAE decreased from 3.84 to 1.15. While for minimum temperature, the average R2 and KGE increased from 0.84 and 0.21 to 0.86 and 0.89, while MAE decreased from 3.48 to 1.17. Moreover, the findings of the annual trend showed that both the maximum and minimum temperatures had increasing trends. The seasonal trend analysis showed an increase in winter (DJF), spring (MAM), autumn (SON), and summer (JJA) for maximum temperature, with the exception of the east as well as southwest regions, which comprise Nangarhar, Kunar, Laghman, Nuristan, as well as the northern regions of Paktika. While for minimum temperature, seasonal trend analysis shows an increasing trend for all seasons, such as winter (DJF), spring (MAM), summer (JJA), and autumn (SON).