Estimating the Standardized Precipitation Evapotranspiration Index Using Data-Driven Techniques: A Regional Study of Bangladesh

Abstract

Drought prediction is the most effective way to mitigate drought impacts. The current study examined the ability of three renowned machine learning models, namely additive regression (AR), random subspace (RSS), and M5P tree, and their hybridized versions (AR-RSS, AR-M5P, RSS-M5P, and AR-RSS-M5P) in predicting the standardized precipitation evapotranspiration index (SPEI) in multiple time scales. The SPEIs were calculated using monthly rainfall and temperature data over 39 years (1980–2018). The best subset regression model and sensitivity analysis were used to determine the most appropriate input variables from a series of input combinations involving up to eight SPEI lags. The models were built at Rajshahi station and validated at four other sites (Mymensingh, Rangpur, Bogra, and Khulna) in drought-prone northern Bangladesh. The findings indicated that the proposed models can accurately forecast droughts at the Rajshahi station. The M5P model predicted the SPEIs better than the other models, with the lowest mean absolute error (27.89–62.92%), relative absolute error (0.39–0.67), mean absolute error (0.208–0.49), root mean square error (0.39–0.67) and highest correlation coefficient (0.75–0.98). Moreover, the M5P model could accurately forecast droughts with different time scales at validation locations. The prediction accuracy was better for droughts with longer periods.

Optimizing ETo Prediction in the Mahanadi Basin: A Comprehensive Evaluation of Machine Learning Models with Emphasis on ANFIS Performance

“…The weighted scores based on multiple evaluation metrics provide a comprehensive assessment of the model's predictive capabilities. M5P models also did a remarkable performance in estimating ETo same work was done by Elbeltagi et al, (Elbeltagi et al, 2022) with the M5P tree and their hybridized versions (AR-RSS, AR-M5P, RSSM5P, and AR-RSS-M5P). The evaluation of the GEP model showed slightly lower performance in terms of NSE and MAE compared to the ANFIS and M5P models.…”

“…In recent years, data-driven models, including stochastic and arti cial intelligence methods, have demonstrated their e cacy in modeling and predicting hydrometeorological variables, presenting e cient approaches for ETo modeling and forecasting. These models have shown promising performance and offer a valuable tool set for studying and predicting ETo variations in different spatiotemporal scales (Dehghanisanij et al, 2022) (Elbeltagi et al, 2022) (Azad et al, 2022). Accurate prediction of evapotranspiration holds immense signi cance, particularly in regions like Iran that face limited water resources.…”

Spatiotemporal Characteristics of Meteorological Drought and Wetness Events across the Coastal Savannah Agroecological Zone of Ghana

“…Although we selected the 3 and 12 month timescales, several research studies, such as [34,136,137], have reported the feasibility of considering other timescales. Drought and wetness analyses can be performed on a variety of timescales, depending on the need.…”

Modeling Various Drought Time Scales via a Merged Artificial Neural Network with a Firefly Algorithm

“…Due to the complexity and non-linearity of the drought process, simulations using non-linear time series data are necessary. Consequently, machine learning (ML) systems for drought forecasting have attracted considerable interest [14,15]. In addition, several types of research have shown that AI algorithms outperform conventional approaches [16][17][18][19], such as artificial neural network (ANN) [20], support vector machines (SVMs) [21], random forests [22], and the adaptive neuro-fuzzy inference system (ANFIS) [23], which are examples of these ML systems.…”