Status and prospects for drought forecasting: opportunities in artificial intelligence and hybrid physical–statistical forecasting

Abstract

Despite major improvements in weather and climate modelling and substantial increases in remotely sensed observations, drought prediction remains a major challenge. After a review of the existing methods, we discuss major research gaps and opportunities to improve drought prediction. We argue that current approaches are top-down, assuming that the process(es) and/or driver(s) are known—i.e. starting with a model and then imposing it on the observed events (reality). With the help of an experiment, we show that there are opportunities to develop bottom-up drought prediction models—i.e. starting from the reality (here, observed events) and searching for model(s) and driver(s) that work. Recent advances in artificial intelligence and machine learning provide significant opportunities for developing bottom-up drought forecasting models. Regardless of the type of drought forecasting model (e.g. machine learning, dynamical simulations, analogue based), we need to shift our attention to robustness of theories and outputs rather than event-based verification. A shift in our focus towards quantifying the stability of uncertainty in drought prediction models, rather than the goodness of fit or reproducing the past, could be the first step towards this goal. Finally, we highlight the advantages of hybrid dynamical and statistical models for improving current drought prediction models. This article is part of the Royal Society Science+ meeting issue ‘Drought risk in the Anthropocene’.

Advances and gaps in the science and practice of impact‐based forecasting of droughts

“…Substantial progress has been made in hydro‐climatic forecasting of droughts, with advancements in detection, causation, prediction, and climate change attribution (AghaKouchak et al, 2022; Fung et al, 2020; Sutanto, Wetterhall, et al, 2020a). However, the gap between predicting drought as a hydro‐meteorological event and understanding its real‐world impact on the society and the economy hinders effective response measures (WMO, 2015).…”

Skillful Decadal Flood Prediction

“…Most operational flood forecasts are produced by forcing a conceptual or physics-based hydrological model with dynamical daily or sub-daily climate inputs (Emerton et al, 2016). Statistical-dynamical, or hybrid, methods present an attractive alternative approach to produce streamflow forecasts, as they avoid the need to run an offline land model and benefit from advances in statistical modeling (AghaKouchak et al, 2022;Slater et al, 2022). Streamflow quantiles are predicted using statistical or machine-learning models driven by dynamical weather or climate predictions.…”

Drought risk in the Anthropocene

“…This themed issue documents important scientific advances that are improving understandings of droughts and the evidence needed for drought management. There have been rapid advances in drought forecasting, which are documented in the paper by AghaKouchak et al (Status and prospects for drought forecasting: opportunities in artificial intelligence and hybrid physicalstatistical forecasting) [12]. These advances have been enabled by a combination of new data sources, in particular from satellite Earth observation, and increasingly powerful methods to analyse these vast datasets, notably with machine learning (also employed by Marvel et al [5] for analysis of climate data) and artificial intelligence.…”

“…These advances have been enabled by a combination of new data sources, in particular from satellite Earth observation, and increasingly powerful methods to analyse these vast datasets, notably with machine learning (also employed by Marvel et al [5] for analysis of climate data) and artificial intelligence. AghaKouchak et al [12] demonstrate how physical insights can help to constrain and improve the performance of data-based methods.…”

“…AghaKouchak et al . [ 12 ] demonstrate how physical insights can help to constrain and improve the performance of data-based methods. Increasing computational resources are enabling climatic, hydrological and water resource systems modelling on scales that were hitherto considered to be infeasible—for example the paper by Murgatroyd et al .…”