Yazar "Ali, Zulfiqar" seçeneğine göre listele

Listeleniyor 1 - 3 / 3
  • [ X ]
    Öğe
    A novel semi data dimension reduction type weighting scheme of the multi-model ensemble for accurate assessment of twenty-first century drought
    (Springer, 2024) Mukhtar, Alina; Ali, Zulfiqar; Nazeer, Amna; Dhahbi, Sami; Kartal, Veysi; Deebani, Wejdan
    Accurately and reliably predicting droughts under multiple models of Global Climate Models (GCMs) is a challenging task. To address this challenge, the Multimodel Ensemble (MME) method has become a valuable tool for merging multiple models and producing more accurate forecasts. This paper aims to enhance drought monitoring modules for the twenty-first century using multiple GCMs. To achieve this goal, the research introduces a new weighing paradigm called the Multimodel Homo-min Pertinence-max Hybrid Weighted Average (MHmPmHWAR) for the accurate aggregation of multiple GCMs. Secondly, the research proposes a new drought index called the Condensed Multimodal Multi-Scalar Standardized Drought Index (CMMSDI). To assess the effectiveness of MHmPmHWAR, the research compared its findings with the Simple Model Average (SMA). In the application, eighteen different GCM models of the Coupled Model Intercomparison Project Phase 6 (CMIP6) were considered at thirty-two grid points of the Tibet Plateau region. Mann-Kendall (MK) test statistics and Steady States Probabilities (SSPs) of Markov chain were used to assess the long-term trend in drought and its classes. The analysis of trends indicated that the number of grid points demonstrating an upward trend was significantly greater than those displaying a downward trend in terms of spatial coverage, at a significance level of 0.05. When examining scenario SSP1-2.6, the probability of moderate wet and normal drought was greater in nearly all temporal scales than other categories. The outcomes of SSP2-4.5 demonstrated that the likelihoods of moderate drought and normal drought were higher than other classifications. Additionally, the results of SSP5-8.5 were comparable to those of SSP2-4.5, underscoring the importance of taking effective actions to alleviate drought impacts in the future. The results demonstrate the effectiveness of the MHmPmHWAR and CMMSDI approaches in predicting droughts under multiple GCMs, which can contribute to effective drought monitoring and management.
  • [ X ]
    Öğe
    Assessing the generalization of forecasting ability of machine learning and probabilistic models for complex climate characteristics
    (Springer, 2024) Batool, Aamina; Ali, Zulfiqar; Mohsin, Muhammad; Masmoudi, Atef; Kartal, Veysi; Satti, Samina
    Climate changes and global warming increase risk of recurrent extreme and complex climatic features. It necessitates accurate modeling and forecasting of climate phenomena for sustainable development goals. However, machine learning algorithms and advanced statistical models are extensively employed to analyze complex data and make predictions related to climate phenomena. It is important to have comprehensive knowledge to use these models and consider their potential implications. This study aims to evaluate and compare some popular machine learning and probabilistic methods by analyzing various time series indices associated with precipitation and temperature. For application, time series data of Standardized Precipitation Temperature Index (SPTI), Standardized Temperature Index (STI), Standardized Compound Drought and Heat Index (SCDHI), and Biased Diminished Weighted Regional Drought Index (BDWRDI) are used from various meteorological regions of Pakistan. The performance of each algorithm is compared using Residual Mean Square Error (RMSE) and Mean Average Error (MAE). The outcomes associated with this research indicate a higher preference of neural networks over machine learning methods in the training sets. However, the efficiency varies from model to model, indicator to indicator, time scale to time scale, and location to location during the testing phase. The most appropriate models are found by considering a list of candidates forecasting models and investigating the performance of each model.
  • [ X ]
    Öğe
    Development of Divergence and Interdependence-based Hybrid Weighting Scheme (DIHWS) for accurate assessment of regional drought
    (Springer Wien, 2024) Mukhtar, Alina; Ali, Zulfiqar; Kartal, Veysi; Karakoyun, Erkan; Yousaf, Mahrukh; Sammen, Saad Sh.
    Accurate ensembles of precipitation data play an important role in precise and efficient drought monitoring systems at the regional level. This article proposes a weighted aggregation scheme - the Divergence and Interdependence-based Hybrid Weighting Scheme (DIHWS) - to ensemble precipitation data for accurate regional drought analysis. The derivation of weights is based on the interdependence among meteorological observatories and the divergence from the mean characteristics of regional data. Here, the interdependence among meteorological observatories is assessed using the Bayesian Network theory. At the same time, the divergence from the mean characteristics of regional data is based on the set of equations used for regional aggregation in (Ali et al., Water Resour Manage 36:4099-4114, 2022). Consequently, the paper introduces a new regional drought index - the Bayesian Network-based Adaptive Regional Drought Index (BNARDI). BNARDI is a standardized regional index and used estimated at multiple time scales. The application of DIHWS and BNARDI is based on five regions of varying observatories. We observed smaller MAE values associated with DIHWS than its Simple Model Average (SMA) and one other of its relevant compitator in all the regions. Therefore, we conclude that the proposed weighting scheme and drought index are more reliable for regional drought monitoring and forecasting. Additionally, the research includes various forecasting models to assess their appropriateness for forecasting the new regional index. The results of this research demonstrate that no single method is suitable for forecasting complex drought data, as generated by BNARDI. Therefore, we suggest using varying methods or a hybrid of various candidate forecasting models for forecasting BNARDI.