Forecasting of SPI and Meteorological Drought Based on the Artificial Neural Network and M5P Model Tree

Abstract

Climate change has caused droughts to increase in frequency and severity worldwide, which has attracted scientists to create drought prediction models to mitigate the impacts of droughts. One of the most important challenges in addressing droughts is developing accurate models to predict their discrete characteristics, i.e., occurrence, duration, and severity. The current research examined the performance of several different machine learning models, including Artificial Neural Network (ANN) and M5P Tree in forecasting the most widely used drought measure, the Standardized Precipitation Index (SPI), at both discrete time scales (SPI 3, SPI 6). The drought model was developed utilizing rainfall data from two stations in India (i.e., Angangaon and Dahalewadi) for 2000–2019, wherein the first 14 years are employed for model training, while the remaining six years are employed for model validation. The subset regression analysis was performed on 12 different input combinations to choose the best input combination for SPI 3 and SPI 6. The sensitivity analysis was carried out on the given best input combination to find the most effective parameter for forecasting. The performance of all the developed models for ANN (4, 5), ANN (5, 6), ANN (6, 7), and M5P models was assessed through the different statistical indicators, namely, MAE, RMSE, RAE, RRSE, and r. The results revealed that SPI (t-1) is the most sensitive parameters with highest values of β = 0.916, 1.017, respectively, for SPI-3 and SPI-6 prediction at both stations on the best input combinations i.e., combination 7 (SPI-1/SPI-3/SPI-4/SPI-5/SPI-8/SPI-9/SPI-11) and combination 4 (SPI-1/SPI-2/SPI-6/SPI-7) based on the higher values of R2 and Adjusted R2 while the lowest values of MSE values. It is clear from the performance of models that the M5P model has higher r values and lesser RMSE values as compared to ANN (4, 5), ANN (5, 6), and ANN (6, 7) models. Therefore, the M5P model was superior to other developed models at both stations.

Estimating the monthly pan evaporation with limited climatic data in dryland based on the extended long short-term memory model enhanced with meta-heuristic algorithms

“…6 . Taylor diagram shows the standard deviation, RMSE,and Pearson correlation coefficient on a two-dimensional chart, which provides an intuitive way to compare the model performance and reflects the simulation capability of the proposed models 10 , 11 , 18 , 35 . On the whole, Fig.…”

“…In general, the direct measurements method (e.g., Class A pan, Lysimeter group) is largely restricted due to the limitation of experimental conditions in dryland 14 – 16 , and the physically-based methods (e.g., Dalton model, FAO-56 Penman–Monteith method, etc.) have the drawbacks that the estimated results are very sensitive to the errors of parameters 17 , 18 , and the key meteorological factors(e.g., relative humidity, latent heat of evaporation, radiation) are sometimes difficult to be measured in the arid sand land 19 , 20 . Therefore, it is necessary to construct the data-driven models to estimate the Ep with less meteorological information.…”

“…Accurate estimation of evaporation by using is a significant issue in ecological management [6][7][8][9][10][11] , especially in arid sand land, where the stability and sustainability of the artificially re-vegetated belts depend on the effective utilization of the limited available water resources 12,13 .In general, the direct measurements method (e.g., Class A pan, Lysimeter group) is largely restricted due to the limitation of experimental conditions in dryland [14][15][16] , and the physically-based methods (e.g., Dalton model, FAO-56 Penman-Monteith method, etc.) have the drawbacks that the estimated results are very sensitive to the errors of parameters 17,18 , and the key meteorological factors(e.g., relative humidity, latent heat of evaporation,…”

“…In general, the direct measurements method (e.g., Class A pan, Lysimeter group) is largely restricted due to the limitation of experimental conditions in dryland [14][15][16] , and the physically-based methods (e.g., Dalton model, FAO-56 Penman-Monteith method, etc.) have the drawbacks that the estimated results are very sensitive to the errors of parameters 17,18 , and the key meteorological factors(e.g., relative humidity, latent heat of evaporation,…”

Decision Tree-Based Modeling of the Aeration Effectiveness of Circular Plunging Jets

“…M5P and ANN were used in another study to predict SPI and meteorological draft. The models were compared, and M5P was found to be a better-predicting model than ANN with a higher R value and fewer errors . The predictive models for the dynamic modulus of asphalt concretes were also created by using the M5P model tree approach.…”