An official website of the United States government
Abstract Machine learning (ML) models are successful with weather forecasting and have shown progress in climate simulations, yet leveraging them for useful climate predictions needs exploration. Here we show this feasibility using Neural General Circulation Model (NeuralGCM), a hybrid ML-physics atmospheric model developed by Google, for seasonal predictions of large-scale atmospheric variability and Northern Hemisphere tropical cyclone (TC) activity. Inspired by physical model studies, we simplify boundary conditions, assuming sea surface temperature (SST) and sea ice follow their climatological cycle but persist anomalies present at the initialization time. With such forcings, NeuralGCM can generate 100 simulation days in ~8 minutes with a single Graphics Processing Unit (GPU), while simulating realistic atmospheric circulation and TC climatology patterns. This configuration yields useful seasonal predictions (July–November) for the tropical atmosphere and various TC activity metrics. Notably, the predicted and observed TC frequency in the North Atlantic and East Pacific basins are significantly correlated during 1990–2023 (r=~0.7), suggesting prediction skill comparable to existing physical GCMs. Despite challenges associated with model resolution and simplified boundary forcings, the model-predicted interannual variations demonstrate significant correlations with the observation, including the sub-basin TC tracks (p<0.1) and basin-wide accumulated cyclone energy (p<0.01) of the North Atlantic and North Pacific basins. These findings highlight the promise of leveraging ML models with physical insights to model TC risks and deliver seamless weather-climate predictions.
more »
« less
Subseasonal Tropical Cyclone Prediction and Modulations by MJO and ENSO in CESM2
Abstract Subseasonal tropical cyclone (TC) reforecasts from the Community Earth System Model version 2 (CAM6) subseasonal prediction system are examined in this study. We evaluate the modeled TC climatology and the probabilistic forecast skill of basin‐wide TC genesis at weekly temporal resolution. Prediction skill is calculated using the Brier skill score relative to a constant annual mean climatology and to a monthly varying seasonal climatology during TC season. The model captures the observed basin‐wide climatological TC seasonality and spatial distributions at weeks 1–6, but TC genesis is largely underestimated from Week 2 onward. For some basins and lead times, the predicted TC genesis is primarily controlled by the number of TC “seeds” and the mean‐state climate condition. The model has good prediction skill relative to the constant climatology across all the basins and lead times, but is only skillful in the eastern Pacific, North Indian Ocean, and Southern Hemisphere at Week 1 when compared to the seasonal climatology, indicating limited skill in predicting deviations from the seasonal cycle. We find strong modulations of the predicted TC genesis at up to 3 weeks of forecast lead time by the Madden‐Julian Oscillation. The interannual variability of predicted TC genesis and accumulated cyclone energy are skillfully predicted in the North Atlantic and the Northwestern Pacific, with a strong modulation by the El Nino‐Southern Oscillation.
more »
« less
- Award ID(s):
- 1652289
- PAR ID:
- 10385043
- Publisher / Repository:
- DOI PREFIX: 10.1029
- Date Published:
- Journal Name:
- Journal of Geophysical Research: Atmospheres
- Volume:
- 127
- Issue:
- 22
- ISSN:
- 2169-897X
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Abstract It is widely agreed that subseasonal-to-seasonal (S2S) predictability arises from the atmospheric initial state during early lead times and from the land and ocean during long lead times. We test this hypothesis for the large-scale mid-latitude atmosphere by training numerous XGBoost models to predict weather regimes (WRs) over North America at 1-to-8-week lead times. Each model uses a different predictor from one Earth system component (atmosphere, ocean, or land) sourced from reanalysis. According to the models, the atmosphere provides more predictability during the first two forecast weeks, and the three components performed similarly afterward. However, the skill and sources of predictability are highly dependent on the season and target WR. Our results show greater WR predictability in fall and winter, particularly for the Pacific Trough and Pacific Ridge regimes, driven primarily by the ocean (e.g., El Niño-Southern Oscillation and sea ice). For the Pacific Ridge in winter, the stratosphere also contributes significantly to predictability across most S2S lead times. Additionally, the initial large-scale tropospheric structure (encompassing the tropics and extra-tropics, e.g., Madden-Julian Oscillation) and soil conditions play a relevant role—most notably for the Greenland High regime in winter. This study highlights previously identified sources of predictability for the large-scale atmosphere and gives insight into new sources for future study. Given how closely linked WRs are to surface precipitation and temperature anomalies, storm tracks, and extreme events, the study results contribute to improving S2S prediction of surface weather.more » « less
-
Abstract Seasonal predictions of tropical cyclone (TC) landfalls are challenging because seasonal landfall count not only depends on the number and spatial distribution of TC genesis, but also whether those TCs are steered toward land or not. Past studies have separately examined genesis and landfall as a function of large-scale ocean and atmospheric environmental conditions. Here, we introduce a practical statistical framework for estimating the seasonal count of TC landfalls as the product of a Poisson model for seasonal TC genesis and a logistic model for landfall probability. We compute spatial variations in TC landfall and genesis by decomposing TC activity in the western North Pacific (WNP) basin into 10° × 10° bins, then identify coherent regions where El Niño–Southern Oscillation (ENSO) and the western extent of the Pacific subtropical high (WPSH) have significant influences on seasonal landfall count. Our framework shows that ENSO and the WPSH are weakly related to basinwide landfalls but strongly related to regional genesis and landfall probability. ENSO modulates the zonal distribution of TC genesis, consistent with past work, whereas the WPSH modulates the meridional distribution of landfall probability due to variations in steering flow associated with the Pacific subtropical high. These spatial patterns result in four coherent subregions of the WNP basin that define seasonal landfall variations: landfall count increases in the southwestern WNP during a positive WPSH and La Niña, the south-central WNP during a positive WPSH and El Niño, the eastern WNP during a negative WPSH and El Niño, and the northern WNP during a negative WPSH and La Niña.more » « less
-
Abstract El Niño–Southern Oscillation (ENSO) influences seasonal Atlantic tropical cyclone (TC) activity by impacting environmental conditions important for TC genesis. However, the influence of future climate change on the teleconnection between ENSO and Atlantic TCs is uncertain, as climate change is expected to impact both ENSO and the mean climate state. We used the Weather Research and Forecasting Model on a tropical channel domain to simulate 5-member ensembles of Atlantic TC seasons in historical and future climates under different ENSO conditions. Experiments were forced with idealized sea surface temperature configurations based on the Community Earth System Model (CESM) Large Ensemble representing: a monthly varying climatology, eastern Pacific El Niño, central Pacific El Niño, and La Niña. The historical simulations produced fewer Atlantic TCs during eastern Pacific El Niño compared to central Pacific El Niño, consistent with observations and other modeling studies. For each ENSO state, the future simulations produced a similar teleconnection with Atlantic TCs as in the historical simulations. Specifically, La Niña continues to enhance Atlantic TC activity, and El Niño continues to suppress Atlantic TCs, with greater suppression during eastern Pacific El Niño compared to central Pacific El Niño. In addition, we found a decrease in the Atlantic TC frequency in the future relative to historical regardless of ENSO state, which was associated with a future increase in northern tropical Atlantic vertical wind shear and a future decrease in the zonal tropical Pacific sea surface temperature (SST) gradient, corresponding to a more El Niño–like mean climate state. Our results indicate that ENSO will remain useful for seasonal Atlantic TC prediction in the future.more » « less
-
Abstract Atmospheric rivers (ARs) and Santa Ana winds (SAWs) are impactful weather events for California communities. Emergency planning efforts and resource management would benefit from extending lead times of skillful prediction for these and other types of extreme weather patterns. Here we describe a methodology for subseasonal prediction of impactful winter weather in California, including ARs, SAWs and heat extremes. The hybrid approach combines dynamical model and historical information to forecast probabilities of impactful weather outcomes at weeks 1–4 lead. This methodology uses dynamical model information considered most reliable, that is, planetary/synoptic‐scale atmospheric circulation, filters for dynamical model error/uncertainty at longer lead times and increases the sample of likely outcomes by utilizing the full historical record instead of a more limited suite of dynamical forecast model ensemble members. We demonstrate skill above climatology at subseasonal timescales, highlighting potential for use in water, health, land, and fire management decision support.more » « less
