High prediction skill of decadal tropical cyclone variability in the North Atlantic and East Pacific in the Met Office decadal prediction system DePreSys four

2tkn-2026-02-09_20_54_20-monerie-2025-high-decadal-prediction-skill.pdf

High prediction skill of decadal tropical cyclone variability in the North Atlantic and East Pacific in the Met Office decadal prediction system DePreSys four.

The UK Met Office decadal prediction system DePreSys four shows skill in predicting the number of tropical cyclones and tropical cyclone track density over the eastern Pacific and tropical Atlantic Ocean on the decadal timescale, up to anomaly correlation coefficient equals zero point nine three and anomaly correlation coefficient equals zero point eight three, respectively. The high skill in predicting the number of tropical cyclones is related to the simulation of the externally forced response, with internal climate variability also allowing the improvement in prediction skill. The skill is due to the model's ability to predict the temporal evolution of surface temperature and vertical wind shear over the eastern Pacific and tropical Atlantic Ocean. We apply a signal-to-noise calibration framework and show that DePreSys four predicts an increase in the number of tropical cyclones over the eastern Pacific and the tropical Atlantic Ocean in the next decade, twenty twenty-three to twenty thirty, potentially leading to high economic losses.

Tropical cyclones are strong atmospheric weather systems that travel long distances over land and the oceans. Tropical cyclones are associated with high winds, storm surges, large waves and heavy rainfall, causing casualties and economic losses. Tropical cyclones mostly affect many regions, including the tropical Atlantic and Caribbean Sea, the tropical Pacific, and the Indian Ocean.

Predicting the future evolution in the number and intensity of tropical cyclones is of high importance for populations and decision-makers, especially in vulnerable areas. Many studies have focused on understanding and improving tropical cyclone prediction at sub-seasonal to annual timescales, i.e. sub-seasonal and seasonal predictions. Some dynamical prediction systems have demonstrated an ability to predict the year-to-year evolution of the number of tropical cyclones over the North Atlantic Ocean and the western North Pacific Ocean. One source of such prediction is the El Niño Southern Oscillation, with El Niño events associated with an increased prediction skill for tropical cyclones relative to neutral years.

Evidence shows that anthropogenic activities can modulate the tropical cyclone activity on multi-decadal to longer timescales. For example, one study has found that changes in anthropogenic aerosol emissions have affected tropical cyclone activity over the North Atlantic and western North Pacific over the past forty years through their effects on sea surface temperature and atmospheric circulation. The externally forced response may lead to a future increase in the intensity of tropical cyclones, leading to a higher risk of tropical cyclonerelated damage over the tropics. Several models project a future decrease in the number of tropical cyclones globally but with an increase in the proportion of the strong tropical cyclones. However, the future change in tropical cyclone activity at regional scales remains uncertain.

While decadal prediction provides useful pre-planning information for stakeholders and policy-makers, the performance of decadal prediction systems remains largely unexploited. The decadal timescale bridges the gap between seasonal prediction and climate projection. Studies have shown that dynamical prediction systems have skill for proxied tropical cyclones on a decadal timescale in the North Atlantic Ocean. They link skill for tropical cyclone activity to skill in predicting the Atlantic Multidecadal Variability. Skill has also been reported in predicting tropical cyclone activity over the western North Pacific. However, these aforementioned studies are based on a simplified approach, using proxies of tropical cyclone, e.g., using the daily minimum of mean sea level pressure, and have relied on statistical or hybrid statistical-dynamic prediction frameworks. These studies have severe limitations because non-stationarities in the tropical cyclone-environment proxy relationship can strongly impact the statistical approach. The ability of dynamical prediction systems to explicitly predict tropical cyclones on decadal timescales is unknown. Here, we assess the prediction skill of tropical cyclones globally, using an explicit tracking algorithm to identify tropical cyclones, in the newly developed decadal prediction system of the UK Met Office, DePreSys four.

We address the following questions:

· Can DePreSys four predict the evolution of the tropical cyclone activity up to ten years ahead?

· Can we define sources of predictive skill for tropical cyclone activity?

· Can we increase the predictive skill of tropical cyclones by addressing the signal-to-noise paradox in climate models?

Results

Representation of the climatology of tropical cyclone activity

We first assess the ability of DePreSys four to simulate the tropical cyclone track density over the nineteen sixty-one to twenty twenty-one period, focusing on interannual variability to use the largest number of start dates possible. Simulated tropical cyclones are validated against tropical cyclones tracked in ERA five using the same tropical cyclone tracking method. Climatologically, the tropical cyclone track density is highest over the eastern and western North Pacific Ocean in ERA five, as shown in other studies. DePreSys four simulates the geographical distribution of track density well, with the highest values seen over the Pacific Ocean and resembling the track density pattern in ERA five. However, DePreSys four underestimates the track density over the North Atlantic Ocean and overestimates it over the Pacific and Indian Oceans. These biases in the representation of the track density are consistent with other prediction systems from the Met Office. Consistent with the track density, DePreSys four underestimates tropical cyclone genesis over the North Atlantic Ocean and overestimates it over the Pacific and Indian Oceans.

We estimate the ability of DePreSys four to simulate the temporal variability in the tropical cyclone track density. The DePreSys four ensemble mean strongly underestimates the interannual variability of tropical cyclone track density. We resample the ensemble members to assess the interannual variability of the track density, using a single realization for each start date. We confirm that the underestimation of the interannual variability of the track density is not solely in the ensemble mean but also inherent to individual members of DePreSys four.