Scholarlink

<p>The Standardised Precipitation Index (SPI) was considered in this work for evaluating non-stationarity in India's meteorological drought using the Generalised Additive Model in Location, Scale, and Shape (GAMLSS) structure for the period 1989–2023, with time and climate covariates. The non-stationary model's performance is evaluated in a comparative research study on time scales of 0.5, 1, 3, 6, 12, 24, and 48&#xa0;months. The Akaike Information Criteria (AIC) is used to select the best models. Using the Kernel Density Estimate (KDE), the characteristics of drought, specifically Drought Duration (DD) and Drought Severity (DS), are examined. Furthermore, significant variations are observed when drought characteristics are taken into account. The significant explanatory covariates for each of the 4641 grid locations across all time scales are highlighted. The findings showed that, for various drought scales, non-stationarity is prevalent in India's meteorological drought. The Diurnal Temperature Range (DTR) is the most significant of the chosen climatic covariates. Spatial variations in precipitation and temperature indicate that non-stationary location scale (NS–LS) parameters outperform those with only location non-stationarity (NS–L) across India. A comparative analysis at Rajkot and Ramanadapuram reveals that Non-Stationary (NS) models outperform stationary ones, with DTR as the dominant covariate across all time scales. Compared with KDE plots, the Stationary (S) model plot varies with drought characteristics in the absence of external influences, whereas the NS model plot shows a different behaviour. During the intense 2016 drought, NS models accurately depicted the spatial distribution and intensity of drought across India, underscoring the importance of climatic covariates. Deficient monsoonal rains caused decreased agricultural output and further strained water supplies. Sustainable mitigation and adaptation methods that account for non-stationary trends in climatic data are necessary to effectively address the evolving nature of drought. In a changing environment, this innovative drought analysis method yields consistent results across the research area. The findings carry considerable consequences for climate risk assessment, drought assessment, and policy formulation. Spatial mapping of dominant covariates across India enables region-specific drought early warning systems, where DTR-sensitive regions can be prioritised for temperature-based mitigation measures. The methodology can also be applied to other climate indices and regions to examine non-stationary drought dynamics. For policymakers and planners, this approach provides a quantitative basis for improving irrigation scheduling, water allocation, and agricultural resilience planning in response to changing drought risks.</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

Assessment of non-stationarity in meteorological drought across India under changing climate

  • 期刊论文

  • 发表日期：

    2026年7月21日

查看全文
  • S. P. Swarna Latshmi,
  • Degavath Vinod,
  • Amai Mahesha

摘要

The Standardised Precipitation Index (SPI) was considered in this work for evaluating non-stationarity in India's meteorological drought using the Generalised Additive Model in Location, Scale, and Shape (GAMLSS) structure for the period 1989–2023, with time and climate covariates. The non-stationary model's performance is evaluated in a comparative research study on time scales of 0.5, 1, 3, 6, 12, 24, and 48 months. The Akaike Information Criteria (AIC) is used to select the best models. Using the Kernel Density Estimate (KDE), the characteristics of drought, specifically Drought Duration (DD) and Drought Severity (DS), are examined. Furthermore, significant variations are observed when drought characteristics are taken into account. The significant explanatory covariates for each of the 4641 grid locations across all time scales are highlighted. The findings showed that, for various drought scales, non-stationarity is prevalent in India's meteorological drought. The Diurnal Temperature Range (DTR) is the most significant of the chosen climatic covariates. Spatial variations in precipitation and temperature indicate that non-stationary location scale (NS–LS) parameters outperform those with only location non-stationarity (NS–L) across India. A comparative analysis at Rajkot and Ramanadapuram reveals that Non-Stationary (NS) models outperform stationary ones, with DTR as the dominant covariate across all time scales. Compared with KDE plots, the Stationary (S) model plot varies with drought characteristics in the absence of external influences, whereas the NS model plot shows a different behaviour. During the intense 2016 drought, NS models accurately depicted the spatial distribution and intensity of drought across India, underscoring the importance of climatic covariates. Deficient monsoonal rains caused decreased agricultural output and further strained water supplies. Sustainable mitigation and adaptation methods that account for non-stationary trends in climatic data are necessary to effectively address the evolving nature of drought. In a changing environment, this innovative drought analysis method yields consistent results across the research area. The findings carry considerable consequences for climate risk assessment, drought assessment, and policy formulation. Spatial mapping of dominant covariates across India enables region-specific drought early warning systems, where DTR-sensitive regions can be prioritised for temperature-based mitigation measures. The methodology can also be applied to other climate indices and regions to examine non-stationary drought dynamics. For policymakers and planners, this approach provides a quantitative basis for improving irrigation scheduling, water allocation, and agricultural resilience planning in response to changing drought risks.