Short-Term Spatio-Temporal Drought Forecasting Using Random Forests Model at New South Wales, Australia

Abstract

Droughts can cause significant damage to agriculture and water resources, leading to severe economic losses and loss of life. One of the most important aspect is to develop effective tools to forecast drought events that could be helpful in mitigation strategies. The understanding of droughts has become more challenging because of the effect of climate change, urbanization and water management; therefore, the present study aims to forecast droughts by determining an appropriate index and analyzing its changes, using climate variables. The work was conducted in three different phases, first being the determination of Standard Precipitation Evaporation Index (SPEI), using global climatic dataset of Climate Research Unit (CRU) from 1901–2018. The indices are calculated at different monthly intervals which could depict short-term or long-term changes, and the index value represents different drought classes, ranging from extremely dry to extremely wet. However, the present study was focused only on forecasting at short-term scales for New South Wales (NSW) region of Australia and was conducted at two different time scales, one month and three months. The second phase involved dividing the data into three sample sizes, training (1901–2010), testing (2011–2015) and validation (2016–2018). Finally, a machine learning approach, Random Forest (RF), was used to train and test the data, using various climatic variables, e.g., rainfall, potential evapotranspiration, cloud cover, vapor pressure and temperature (maximum, minimum and mean). The final phase was to analyze the performance of the model based on statistical metrics and drought classes. Regarding this, the performance of the testing period was conducted by using statistical metrics, Coefficient of Determination (R2) and Root-Mean-Square-Error (RMSE) method. The performance of the model showed a considerably higher value of R2 for both the time scales. However, statistical metrics analyzes the variation between the predicted and observed index values, and it does not consider the drought classes. Therefore, the variation in predicted and observed SPEI values were analyzed based on different drought classes, which were validated by using the Receiver Operating Characteristic (ROC)-based Area under the Curve (AUC) approach. The results reveal that the classification of drought classes during the validation period had an AUC of 0.82 for SPEI 1 case and 0.84 for SPEI 3 case. The study depicts that the Random Forest model can perform both regression and classification analysis for drought studies in NSW. The work also suggests that the performance of any model for drought forecasting should not be limited only through statistical metrics, but also by examining the variation in terms of drought characteristics.

Estimation of SPEI Meteorological Drought Using Machine Learning Algorithms

“…This outcome is consistent with the findings presented by [119], that the accuracy of the prediction process is greater for long-term drought episodes compared to the short-term periods. Further, our finding is in agreement with the study of SPEI prediction in New South Wales (Australia) which concluded that RF achieved RMSE and R 2 results of 0.53 and 0.76, respectively, for SPEI-3 [120]. In contrast, the SVM algorithm applied to Combined Terrestrial Evapotranspiration Index (CTEI) resulted in an R 2 value of 0.82 and the lowest errors in terms of the root mean squared error (RMSE) (0.33) and Mean Absolute Error (MAE) (0.20) in the Ganga River basin [42] which agree with our results, and were much better in comparison to the other ML models they applied.…”

Application of Machine Learning Techniques to Delineate Homogeneous Climate Zones in River Basins of Pakistan for Hydro-Climatic Change Impact Studies

“…The "no trend" in precipitation with time is assumed in the null hypothesis (H o ) of the test and vice versa for the alternative hypothesis (H a ). The Equations (9)- (12) show the test stats T of the MK test.…”

“…Simulation results from general circulation models (GCMs) are used to study the implications of hydro-climatic change and are considered the most comprehensive and valid instruments, with the ability to reproduce climate variables in chronological order as well as yield future projections using scenarios [8][9][10][11] recommended by the intergovernmental panel on climate change (IPCC). Numerous outputs from GCMs are available to gain insight into climate processes and the impact of various scenarios on these processes [12,13]. The question arises as to how to use these climate simulations to obtain meteorological inputs for impact models to characterize all the uncertainties involved in the models.…”

Agricultural drought risk assessment of Northern New South Wales, Australia using geospatial techniques

“…Four types of droughts are found in the literature: meteorological, agricultural, hydrological, and socio-economic (Sharafati et al 2019;Nabaei et al 2019;Deng et al 2018). Australia is frequently affected by agricultural drought events (Rahmati et al 2019;Dikshit et al 2020b).…”