Potential Precipitation Predictability Decreases Under Future Warming

Abstract

Precipitation predictability is likely to decrease with water cycle intensification under global warming, yet how much it will change spatiotemporally is unclear. We quantify the precipitation predictability changes under future warming using model simulations from Coupled Model Intercomparison Project Phase 5. The global-averaged potential precipitation predictability (PPP) is likely to decrease 0.3% per 1°C increase of global sea surface temperature (SST), with a decrease of 0.8% and 0.1% in tropical and extratropical regions, respectively, under future warming scenarios. The PPP changes are divergent among different models with considerable uncertainty. The estimated declining PPP is closely related to the increase of SST and the decrease of potential SST predictability based on a statistical regression analysis. These results unravel the changing pattern of precipitation predictability under future warming. Plain Language Summary Precipitation prediction for several weeks or months in advance provides opportunities for the preparation of extreme weather events such as droughts and floods. The predictable ability of precipitation determines how long it can be forecasted. We found that precipitation is becoming more difficult to predict when the global sea surface temperature (SST) continues to increase in the future, although different models show divergent results. The decreasing precipitation predictable ability is well explained by the increasing SST and the decreasing predictable ability of SST. Although extreme precipitation events may increase in a warmer world in the context of the enhanced SST variability and other changes such as sulfate aerosols and land use, precipitation is probably less predictable as a result of lower predictable ability of SST.

Precipitation predictability affects intra- and trans-generational plasticity and causes differential selection on root traits of Papaver rhoeas

“…Current climate change entails rising global temperatures, longer and more frequent dry periods, and changes in weather predictability (Xu et al, 2020;IPCC, 2021). Although global mean precipitation is expected to remain rather constant in the near future, precipitation patterns will change, leading to a reduction in precipitation predictability at different temporal scales, i.e., among days, weeks, seasons or years (Tonkin et al, 2017;Xu et al, 2020).…”

“…Current climate change entails rising global temperatures, longer and more frequent dry periods, and changes in weather predictability (Xu et al, 2020;IPCC, 2021). Although global mean precipitation is expected to remain rather constant in the near future, precipitation patterns will change, leading to a reduction in precipitation predictability at different temporal scales, i.e., among days, weeks, seasons or years (Tonkin et al, 2017;Xu et al, 2020). The degree of intrinsic precipitation predictability (i.e., the degree of temporal autocorrelation among precipitation events; Pennekamp et al, 2019) will determine the amount and timing of water availability for plants.…”

“…Although global mean precipitation is expected to remain rather constant in the near future, precipitation patterns will change, leading to a reduction in precipitation predictability at different temporal scales, i.e. , among days, weeks, seasons or years ( Tonkin et al., 2017 ; Xu et al., 2020 ). The degree of intrinsic precipitation predictability ( i.e.…”

Causal analysis of global warming: based on ARIMA and LSTM models

“…Sevellec et al [1] developed a new method for predicting global mean surface and sea surface temperatures based on transfer operators that allows probabilistic predictions by design, improving accuracy predictions. Xu et al [2] used model simulations from Phase 5 of the Coupled Model Comparison Project to quantify changes in precipitation predictability under future warming, revealing patterns of change in precipitation predictability under future warming. Grose et al [3] used a large amount of data with the latest global climate model simulations to complement six years of reliable observations of global warming, and their findings suggest that future warming will be slightly lower than the rates observed to date.…”

A functional vulnerability framework for biodiversity conservation

“…Such conditions-or quasi-pristine areas-are now virtually absent for most ecosystems on Earth 28,29 however, and past data about prehuman conditions are often biased and limited 27 . Moreover, the natural stochasticity in the Earth's climate (e.g., volcanism, extreme meteorological events), the lack of data relative to the combined effects of various disturbances on biodiversity, and the extreme complexity of biological responses themselves 30 , still prevent accurate predictions of species community shifts under global changes 31,32 . Such unpredictability hampers our ability to cope with future environmental and socioeconomic changes, weakening management and conservation efforts.…”