Drought prediction based on an improved VMD-OS-QR-ELM model

To overcome the low accuracy, poor reliability, and delay in the current drought prediction models, we propose a new extreme learning machine (ELM) based on an improved variational mode decomposition (VMD). The model first redefines the output of the hidden layer of the ELM model with orthogonal triangular matrix decomposition (QR) to construct an orthogonal triangular ELM (QR-ELM), and then introduces an online sequence learning mechanism (OS) into the QR-ELM to construct an online sequence OR-ELM (OS-QR-ELM), which effectively improves the efficiency of the ELM model. The mutual information extension method was then used to extend both ends of the original signal to improve the VMD end effect. Finally, VMD and OS-QR-ELM were combined to construct a drought prediction method based on the VMD-OS-QR-ELM. The reliability and accuracy of the VMD-OS-QR-ELM model were improved by 86.19% and 93.20%, respectively, compared with those of the support vector regression model combined with empirical mode decomposition. Furthermore, the calculation efficiency of the OS-QR-ELM model was increased by 88.65% and 85.32% compared with that of the ELM and QR-ELM models, respectively.

Drought monitoring and prediction using SPI, SPEI, and random forest model in various climates of Iran

The aim of this study is to select the best model (combination of different lag times) for predicting the standardized precipitation index (SPI) and the standardized precipitation and evapotranspiration index (SPEI) in next time. Monthly precipitation and temperature data from 1960 to 2019 were used. In temperate climates, such as the north of Iran, the correlation coefficient of SPI and SPEI was 0.94, 0.95, and 0.81 at the time scales of 3, 12, and 48 months, respectively. Besides, this correlation coefficient was 0.47, 0.35, and 0.44 in arid and hot climates, such as the southwest of Iran because potential evapotranspiration (PET) depends on temperature more than rainfall. Drought was predicted using the random forest (RF) model and applying 1–12 months lag times for next time. By increasing of time scale, the prediction accuracy of SPI and SPEI will improve. The ability of SPEI is more than SPI for drought prediction, because the overall accuracy (OA) of prediction will increase, and the errors (i.e., overestimate (OE) and underestimate (UE)) will reduce. It is recommended for future studies (1) using wavelet analysis for improving accuracy of predictions and (2) using the Penman–Monteith method if ground-based data are available.

Applying machine learning for drought prediction in a perfect model framework using data from a large ensemble of climate simulations

There is a strong scientific and social interest in understanding the factors leading to extreme events in order to improve the management of risks associated with hazards like droughts. In this study, artificial neural networks are applied to predict the occurrence of a drought in two contrasting European domains, Munich and Lisbon, with a lead time of 1 month. The approach takes into account a list of 28 atmospheric and soil variables as input parameters from a single-model initial-condition large ensemble (CRCM5-LE). The data were produced in the context of the ClimEx project by Ouranos, with the Canadian Regional Climate Model (CRCM5) driven by 50 members of the Canadian Earth System Model (CanESM2). Drought occurrence is defined using the standardized precipitation index. The best-performing machine learning algorithms manage to obtain a correct classification of drought or no drought for a lead time of 1 month for around 55 %–57 % of the events of each class for both domains. Explainable AI methods like SHapley Additive exPlanations (SHAP) are applied to understand the trained algorithms better. Variables like the North Atlantic Oscillation index and air pressure 1 month before the event prove essential for the prediction. The study shows that seasonality strongly influences the performance of drought prediction, especially for the Lisbon domain.

Drought assessment and prediction for Gidra river, Slovakia

Climate change is affecting every aspect of the world including water resources and water scarcity. Drought is one of many big problems associated with climate change that could occur all over the world. Moreover, hydrological drought is one form of drought that relates to decreased river discharges, below-normal groundwater level, declining the area of wetlands and low water level in lakes or reservoirs. In this study, an assessment of hydrological drought in Gidra river is conducted to characterize dry and normal hydrological years according to Slovak Hydrometeorological Institute (SHMI) Methodology. Furthermore, making benefit of machine learning and artificial intelligence in this field is applicable now, as data of many types are being recorded every day. Deploying machine learning algorithms for the purpose of drought prediction is one way to regulate many operations of water management to prevent irrigation problems. By catching patterns through historical data and deploying machines to learn from those patterns, it is possible to use the values of daily average discharges for January, February, March, and April to correctly predict the hydrological situation in Gidra river whether it is dry or normal, knowing that normal situation refers to wet or normal hydrologically assessed years as the optimal goal in this study is drought assessment and prediction of Gidra river.

Defining Homogeneous Drought Zones Based on Soil Moisture across Japan and Teleconnections with Large-Scale Climate Signals

Droughts are widespread disasters worldwide and are concurrently influenced by multiple large-scale climate signals. This is particularly true over Japan, where drought has strong heterogeneity due to multiple factors such as monsoon, topography, and ocean circulations. Regional heterogeneity poses challenges for drought prediction and management. To overcome this difficulty, this study provides a comprehensive analysis of teleconnection between climate signals and homogeneous drought zones over Japan. First, droughts are characterized by simulated soil moisture from land surface model during 1958-2012. The Mclust toolkit, distinct empirical orthogonal function, and wavelet coherence analysis are used, respectively, to investigate the homogeneous drought zone, principal component of each homogeneous zone, and teleconnection between climate signals and drought. Results indicate that nine homogeneous drought zones with different characteristics are defined and quantified. Among these nine zones, zone-1 is dominated by extreme drought events. Zone-2 and zone-6 are typical representatives of spring droughts, while zone-7 is wet for most of the period. The Hokkaido region is divided into wetter zone-4 and drier zone-9. Zone-3, zone-5 and zone-8 are distinguished by the topography. The analyses also reveal almost nine zones have a high level of homogeneity, with more than 60% explained variance. Also, these nine zones are dominated by different large-scale climate signals: the Arctic Oscillation has the strongest impact on zone-1, zone-7, and zone-8; the influence of the North Atlantic Oscillation on zone-3, zone-4, and zone-6 is significant; zone-2 and zone-9 are both dominated by the Pacific Decadal Oscillation; El Niño-Southern Oscillation dominates zone-5. The results will be valuable for drought management and drought prevention.