Model Comparisons Between Canonical Vine Copulas and Meta-Gaussian for Agricultural Drought Forecasting over China

Agricultural drought is caused by reduced soil moisture and precipitation and affects the growth of crops and vegetation, and in turn agricultural production and food security. For developing measures for drought mitigation, reliable agricultural drought forecasting is essential. In this study, we developed an agricultural drought forecasting model based on canonical vine copulas under three-dimensions (3C-vine model), in which the antecedent meteorological drought and agricultural drought persistence were utilized as predictors. Besides, the meta-Gaussian (MG) model was selected as a reference model to evaluate the forecast skill. The agricultural drought in August of 2018 was selected as a case study, and the spatial patterns of 1–3-month lead forecasts of agricultural drought utilizing the 3C-vine model resembled the corresponding observations, indicating the predictive ability of the model. The performance metrics (NSE, R2, and RMSE) showed that the 3C-vine model outperformed the MG model for August under diverse lead times. Also, the 3C-vine model exhibited excellent forecast skills in capturing the extreme agricultural drought over different selected typical regions. This study may help with drought early warning, drought mitigation, and water resources scheduling.

Temporal analysis of rainfall and drought characteristics over Jalore District of S-W Rajasthan

Drought forecasting is being considered an important tool to help understand the rainfall pattern and climate change trend. Drought is a prolonged period of months or years in which an area, whether surface water or groundwater, becomes insufficient in its water supplies. Drought is considered as most difficult but least known environmental phenomenon, impacting more persons than any other. There are several indices used to classify droughts. For this study, precipitation-based drought indices are considered (i.e., SPI, RAI and Percentage Departure of Rainfall). The objective of the research is to examine and determine the possible rainfall trends over the Jalore district of South-West Rajasthan in Luni river basin. In this research, trend analysis using the rainfall data from the years 1901 to 2021 was carried out on monthly, seasonal and annual basis. To define the current trend path, the Mann-Kendall test and Sen's slope estimator test were used. In order to detect the trend and its change in magnitude over a particular period of time, Sen's slope estimator was used. During the southwest monsoon, declining rainfall leads to short-term meteorological droughts, which have severe effect on the agriculture sector and Jalore district's water supplies, while rising rainfall during other seasons tends to mitigate the severity of drought. The result of research reveals that there is rise of pre-monsoon and post-monsoon rainfall, but it also depicts a fall in the annual rainfall which reflects in reduced Winter and S-W monsoon rainfall.

Improving Drought Modeling Using Hybrid Random Vector Functional Link Methods

Drought modeling is essential in water resources planning and management in mitigating its effects, especially in arid regions. Climate change highly influences the frequency and intensity of droughts. In this study, new hybrid methods, the random vector functional link (RVFL) integrated with particle swarm optimization (PSO), the genetic algorithm (GA), the grey wolf optimization (GWO), the social spider optimization (SSO), the salp swarm algorithm (SSA) and the hunger games search algorithm (HGS) were used to forecast droughts based on the standard precipitation index (SPI). Monthly precipitation data from three stations in Bangladesh were used in the applications. The accuracy of the methods was compared by forecasting four SPI indices, SPI3, SPI6, SPI9, and SPI12, using the root mean square errors (RMSE), the mean absolute error (MAE), the Nash–Sutcliffe efficiency (NSE), and the determination coefficient (R2). The HGS algorithm provided a better performance than the alternative algorithms, and it considerably improved the accuracy of the RVFL method in drought forecasting; the improvement in RMSE for the SPI3, SP6, SPI9, and SPI12 was by 6.14%, 11.89%, 14.14%, 24.5% in station 1, by 6.02%, 17.42%, 13.49%, 24.86% in station 2 and by 7.55%, 26.45%, 15.27%, 13.21% in station 3, respectively. The outcomes of the study recommend the use of a HGS-based RVFL in drought modeling.

SPI-Based Hybrid Hidden Markov–GA, ARIMA–GA, and ARIMA–GA–ANN Models for Meteorological Drought Forecasting

Drought is a severe environmental disaster that results in significant social and economic damage. As such, efficient mitigation plans must rely on precise modeling and forecasting of the phenomenon. This study was designed to enhance drought forecasting through developing and evaluating the applicability of three hybrid models—the hidden Markov model–genetic algorithm (HMM–GA), the auto-regressive integrated moving average–genetic algorithm (ARIMA–GA), and a novel auto-regressive integrated moving average–genetic algorithm–ANN (ARIMA–GA–ANN)—to forecast the standard precipitation index (SPI) in the Bisha Valley, Saudi Arabia. The accuracy of the models was investigated and compared with that of classical HMM and ARIMA based on a performance evaluation and visual inspection. Furthermore, the multi-class Receiver Operating Characteristic-based Area under the Curve (ROC–AUC) was applied to evaluate the ability of the hybrid model to forecast drought events. We used data from 1968 to 2008 to train the models and data from 2009 to 2019 for validation. The performance evaluation results confirmed that the hybrid models provided superior results in forecasting the SPI one month in advance. Furthermore, the results demonstrated that the GA-induced improvement in the HMM forecasts was matched by an approximate 16.40% and 23.46% decrease in the RMSE in the training and testing results, respectively, compared to the classical HMM model. Consequently, the RMSE values of the ARIMA–GA model were reduced by an average of 10.06% and 9.36% for the training and testing processes, respectively. Finally, the ARIMA–GA–ANN, which combined the strengths of the linear stochastic model ARIMA and a non-linear ANN, achieved a greater reduction values in RMSE by an average of 32.82% and 27.47% in comparison with ARIMA in the training and testing phases, respectively. The ROC–AUC results confirmed the capability of the developed models to distinguish between events and non-events with reasonable accuracy, implying the appropriateness of these models as a tool for drought mitigation and warning systems.

Forecasting of Drought: A Case Study of Water-Stressed Region of Pakistan

Demand for water resources has increased dramatically due to the global increase in consumption of water, which has resulted in water depletion. Additionally, global climate change has further resulted as an impediment to human survival. Moreover, Pakistan is among the countries that have already crossed the water scarcity line, experiencing drought in the water-stressed Thar desert. Drought mitigation actions can be effectively achieved by forecasting techniques. This research describes the application of a linear stochastic model, i.e., Autoregressive Integrated Moving Average (ARIMA), to predict the drought pattern. The Standardized Precipitation Evapotranspiration Index (SPEI) is calculated to develop ARIMA models to forecast drought in a hyper-arid environment. In this study, drought forecast is demonstrated by results achieved from ARIMA models for various time periods. Result shows that the values of p, d, and q (non-seasonal model parameter) and P, D, and Q (seasonal model parameter) for the same SPEI period in the proposed models are analogous where “p” is the order of autoregressive lags, q is the order of moving average lags and d is the order of integration. Additionally, these parameters show the strong likeness for Moving Average (M.A) and Autoregressive (A.R) parameter values. From the various developed models for the Thar region, it has been concluded that the model (0,1,0)(1,0,2) is the best ARIMA model at 24 SPEI and could be considered as a generalized model. In the (0,1,0) model, the A.R term is 0, the difference/order of integration is 1 and the moving average is 0, and in the model (1,0,2) whose A.R has the 1st lag, the difference/order of integration is 0 and the moving average has 2 lags. Larger values for R2 greater than 0.9 and smaller values of Mean Error (ME), Mean Absolute Error (MAE), Mean Percentile Error (MPE), Mean Absolute Percentile Error (MAPE), and Mean Absolute Square Error (MASE) provide the acceptance of the generalized model. Consequently, this research suggests that drought forecasting can be effectively fulfilled by using ARIMA models, which can be assist policy planners of water resources to place safeguards keeping in view the future severity of the drought.

Scientometric Analysis-Based Review for Drought Modelling, Indices, Types, and Forecasting Especially in Asia

An extended drought period with low precipitation can result in low water availability and issues for humans, animals, and plants. Drought forecasting is critical for water resource development and management as it helps to reduce negative consequences. In this study, scientometric analysis and manual comprehensive analysis on drought modelling and forecasting are used. A scientometric analysis is used to determine the current research trend using bibliometric data and to identify relevant publication field sources with the most publications, the most frequently used keywords, the most cited articles and authors, and the countries that have made the greatest contributions to the field of water resources. This paper also tries to provide an overview of water issues, such as drought classification, drought indices, historical droughts, and their impact on Asian countries such as China, Pakistan, India, and Iran. There have been many models established for this purpose and choosing the appropriate model for study is a long procedure for researchers. An appropriate, comprehensive, pedagogical study of model ideas and historical implementations would benefit researchers by helping them to avoid overlooking viable model options, thus reducing their time spent on the topic. As a result, the goal of this paper is to review drought-forecasting approaches and recommend the best models for the Asian region. The models are divided into four categories based on their mechanisms: Regression analysis, stochastic modelling, machine learning, and dynamic modelling. The basic concepts of each approach in terms of the model’s historical use, benefits, and limitations are explained. Finally, prospects for future drought research in Asia are discussed as well as potential modelling techniques.