Scholarlink

<p>Global warming and climate change have increased the risk of recurrent extreme climatic events. However, accurate joint estimates of drought and heat wave risks are essential for mitigating global warming effects. These factors are crucial for developing sustainable policies. The current research emphasizes the significance of these factors to address the impact of global warming. The study proposes a joint index known as the Adaptive Joint Standardized Drought and Heatwave Index (AJSDHI). The AJSDHI methodology includes ensembled models with a Weighted Aggregation (WA) scheme, K-components Gaussian Mixture Distribution (K-CGMD) for distribution fitting, and a bivariate probabilistic distribution approach. WA weighting mitigates model variability and extreme values, and the bivariate method examines the joint relationship between variables such as precipitation and temperature. Moreover, the suggested index enables the inference of the probabilistic behavior of these joint events using steady-state probabilities. In application, we analyze time series data collected from thirty-two sites around the Tibetan plateau. The dataset spans from 1961 to 2014 of monthly precipitation and temperature observations. Data from eighteen GCMs obtained from the Coupled Model Intercomparison Project phase 6 (CMIP6), served as the primary source for climate projections. In this study accuracy of KCGMD is determined by comparing the Bayesian Information Criterion (BIC) values with suitable univariate distributions. In addition, three forecasting techniques are evaluated for their suitability: Autoregressive Integrated Moving Average (ARIMA), Extreme Learning Machine (ELM), and Multilayer Perceptron (MLP). Findings suggest that K-CGMD is the most suitable fitting approach for ensemble data. In the majority of the points, the joint long-term chances of moderate wet and cold events are much higher than those of moderate dry and hot events. Overall, the frequency counts show that both ELM and MLP generally outperformed ARIMA, with ELM exhibiting a significant performance, specifically in terms of RMSE.</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

Development of co-integrated standardized procedure for the joint monitoring, forecasting and probabilistic characterization of climate extremes under global climate models

  • 期刊论文

  • 发表日期：

    2026年2月27日

查看全文
  • Aamina Batool,
  • Mahrukh Yousaf,
  • Muhammad Shakeel,
  • Amina Magdich,
  • Reem Alreshidi,
  • Zulfiqar Ali,
  • Veysi Kartal

摘要

Global warming and climate change have increased the risk of recurrent extreme climatic events. However, accurate joint estimates of drought and heat wave risks are essential for mitigating global warming effects. These factors are crucial for developing sustainable policies. The current research emphasizes the significance of these factors to address the impact of global warming. The study proposes a joint index known as the Adaptive Joint Standardized Drought and Heatwave Index (AJSDHI). The AJSDHI methodology includes ensembled models with a Weighted Aggregation (WA) scheme, K-components Gaussian Mixture Distribution (K-CGMD) for distribution fitting, and a bivariate probabilistic distribution approach. WA weighting mitigates model variability and extreme values, and the bivariate method examines the joint relationship between variables such as precipitation and temperature. Moreover, the suggested index enables the inference of the probabilistic behavior of these joint events using steady-state probabilities. In application, we analyze time series data collected from thirty-two sites around the Tibetan plateau. The dataset spans from 1961 to 2014 of monthly precipitation and temperature observations. Data from eighteen GCMs obtained from the Coupled Model Intercomparison Project phase 6 (CMIP6), served as the primary source for climate projections. In this study accuracy of KCGMD is determined by comparing the Bayesian Information Criterion (BIC) values with suitable univariate distributions. In addition, three forecasting techniques are evaluated for their suitability: Autoregressive Integrated Moving Average (ARIMA), Extreme Learning Machine (ELM), and Multilayer Perceptron (MLP). Findings suggest that K-CGMD is the most suitable fitting approach for ensemble data. In the majority of the points, the joint long-term chances of moderate wet and cold events are much higher than those of moderate dry and hot events. Overall, the frequency counts show that both ELM and MLP generally outperformed ARIMA, with ELM exhibiting a significant performance, specifically in terms of RMSE.