Andrew C. Winters

Achieving higher regional fidelity within computational constraints remains a key challenge in Earth system modeling; regionally refined models (RRMs) offer a development pathway. This paper documents the Energy Exascale Earth System Model version 3's RRM (v3.RRM) release, providing three AMIP-type configurations with ∼110 → 25 km atmosphere meshes over North America, the Amazon, and East Asia. All configurations share the same globally uniform 0.25 ^o land/river grid, and achieve ≻7 simulated years per wall-clock day, suited for large-ensemble, seasonal-to-decadal prediction, and artificial intelligence applications. Following the hybrid time step strategy of the previous version, all configurations reproduce the lower-resolution global mean result with minimal retuning. At the regional scale, all configurations improve surface air temperature and total precipitable water over complex terrain, particularly over major mountain ranges. The North American and Amazon RRMs reduce persistent low-cloud biases over the California and Peru/Chile coasts via stronger lower-tropospheric static stability facilitated by better-resolved orography. The North American RRM substantially improves snowpack and streamflow, and all configurations capture large-scale atmospheric variability and monsoon-driven precipitation cycles, with the East Asia and Amazon RRMs best matching observed North Atlantic Oscillation variability. Improvements vary by variable, region, and season. A 25-member single-year large ensemble experiment demonstrates v3.RRM's efficiency for studying resolution-sensitive regional processes, including the ENSO-snowpack teleconnection, at a fraction of the cost of multi-decadal simulations. More broadly, v3.RRM contributes to the regional downscaling community by enabling two-way coupling, unlike the one-way forcing of conventional dynamical downscaling.

Misrepresentation of warm-conveyor belts (WCBs) in weather forecasting models can impinge on the medium-range predictability of Rossby waves. Yet, the link between WCB errors and Rossby wave evolution remains poorly understood, despite the importance of Rossby wave breaking for weather extremes. We investigate this link using a WCB case from April 28 to May 4, 2020, which was associated with anticyclonic Rossby wave breaking over the North Atlantic and substantial ensemble variability in the forecasted Rossby wave evolution. Our analysis uses archived forecasts from the Global Ensemble Forecast System (GFS) and the European Centre for Medium-Range Weather Forecasts (ECMWF), as well as global simulations conducted using the Model for Prediction Across Scales (MPAS) with grid spacing from 60 km to 3.75 km. Simulations with more persistent WCBs capture Rossby wave amplification and subsequent anticyclonic wave breaking more accurately, while shorter-lived WCBs yield weaker amplification and cyclonic breaking. A systematic underestimation of the poleward extent of the 500 hPa meridional moisture flux within WCB ascent in GFS and ECMWF forecasts appears to be a key process limiting predictability. Accurate prediction of Rossby wave breaking depends not only on WCB representation but also on the upstream cyclone’s propagation and structure, mesoscale dynamics within the WCB, and upstream convection over North America. This case shows that forecasting Rossby wave breaking requires capturing the collective influence of multi-scale processes interacting with the broader Rossby wave packet. Forecast outcomes are strongly shaped by model choice, but initial conditions and horizontal resolution also play discernible roles.

Misrepresentation of warm‐conveyor belts (WCBs) in weather forecasting models can impinge on the medium‐range predictability of Rossby waves. Yet, the link between WCB errors and Rossby wave evolution remains poorly understood, despite the importance of Rossby wave breaking (RWB) for weather extremes. We investigate this link using a WCB case from 28 April to 4 May 2020, which was associated with anticyclonic RWB over the North Atlantic and substantial ensemble variability in the forecasted Rossby wave evolution. Our analysis uses archived forecasts from the Global Ensemble Forecast System (GFS) and the European Centre for Medium‐Range Weather Forecasts (ECMWF), as well as global simulations conducted using the Model for Prediction Across Scales with grid spacing from 60 to 3.75 km. Simulations with more persistent WCBs capture Rossby wave amplification and subsequent anticyclonic wave breaking more accurately, while shorter‐lived WCBs yield weaker amplification and cyclonic breaking. A systematic underestimation of the poleward extent of the 500 hPa meridional moisture flux within WCB ascent in GFS and ECMWF forecasts appears to be a key process limiting predictability. Accurate prediction of RWB depends not only on WCB representation but also on the upstream cyclone's propagation and structure, mesoscale dynamics within the WCB, and upstream convection over North America. This case shows that forecasting RWB requires capturing the collective influence of multi‐scale processes interacting with the broader Rossby wave packet. Forecast outcomes are strongly shaped by model choice, but initial conditions and horizontal resolution also play discernible roles.

Nuclear war has the potential to cause global catastrophe, as soot from widespread fires blocks sunlight, triggers abrupt cooling, and collapses agriculture. This study examines the long‐term response of weather and climate patterns to nuclear conflict, simulated by injecting ∼150 million tons of smoke into the stratosphere within the Community Earth System Model. An algorithm detecting extreme moisture transport events is applied to evaluate global hydrological impacts. The simulations reveal widespread declines in event frequency, particularly during each hemisphere's summer, but also narrow ocean‐to‐continent corridors of sharp frequency increases during the cold season, including across the North Pacific to the U.S. West Coast, the eastern North Atlantic to North Africa and extending eastward toward South and Southeast Asia, the South Pacific to Chile, and the South Indian Ocean to Australia. In these regions, extreme moisture transport events make disproportionately large contributions to total precipitation. Changes in event distribution are largely driven by circulation changes following the smoke injection: subtropical jets contract equatorward, continental surface anticyclones intensify, and anomalously strengthened offshore cyclones establish anomalous coastal pressure gradients, favoring enhanced moisture transport toward coastal locations. Over southwest North America, these events feature a classic atmospheric river (AR)‐like pattern during all seasons following the smoke injection, which contrasts with the control simulation where AR‐like characteristics appear only during the cold season. This study offers novel insight into how nuclear conflict could profoundly disrupt the global hydrological cycle and extreme precipitation, with critical implications for regional water resources in a severely perturbed climate system.

The state and evolution of the North Pacific jet (NPJ) stream exert considerable influence on weather conditions across North America. The NPJ is modulated by several multiscale processes during the cold season (November–March), including the effects of organized tropical convection on synoptic to seasonal time scales. This study specifically explores the influence of El Niño–Southern Oscillation (ENSO), the Madden–Julian oscillation (MJO), and atmospheric blocking on the short-to-medium-range (0–9-day) predictability of the NPJ. NPJ predictability is evaluated using an NPJ phase diagram that is constructed from the jet’s two leading patterns of variability: a zonal extension or retraction of the climatological exit region of the NPJ and a poleward or equatorward shift of the climatological exit region of the NPJ. Results demonstrate that NPJ forecasts during El Niño or MJO phases 7–8 feature enhanced forecast skill relative to the Global Ensemble Forecast System (GEFS) model climatology, particularly when those forecasts coincide with an extended or poleward-shifted NPJ at the time of forecast initialization. Reduced forecast skill occurs during any active MJO phase if the NPJ is retracted at the time of initialization and during forecasts initialized at the onset of northeast Pacific blocking events. Analyses further suggest that NPJ predictive skill may be derived from the constructive or destructive interference of the separate teleconnections induced by ENSO and the MJO and depends heavily on the incipient state of the NPJ. Significance Statement: The evolution of the North Pacific jet stream plays a considerable role in dictating downstream weather conditions across North America. Consequently, this study examines whether there are certain situations in which the North Pacific jet stream is more predictable at 0–9-day forecast lead times, with a particular focus on the influence of subseasonal-to-seasonal modes of atmospheric variability, such as El Niño, the Madden–Julian oscillation, and midlatitude blocking events. Analyses show that North Pacific jet stream predictability depends not only on the aforementioned modes of variability but also on the state of the jet stream at the time a forecast is made. Forecasts are generally improved during El Niño and particularly when the North Pacific jet stream is extended or poleward shifted. Forecasts are less accurate during periods characterized by a retracted North Pacific jet stream and at the start of northeast Pacific blocking events.

In many regions, extreme temperatures increase faster than background global warming due to complex atmospheric circulation changes. This has been overlooked in Antarctica. Here, we examine extreme temperatures recorded at 16 weather stations since 1950, and identify their changes and drivers. Over the Antarctic Peninsula and the South Pole, 10-year return level temperatures increased by up to 1.19°C per decade since 1980, outpacing mean temperature trends. Most extreme events, including statistically implausible or black swan events, were driven by intense atmospheric rivers linked to Rossby wave trains forced by tropical Pacific convection. In contrast, 10-year return level temperature decreased over most of East Antarctica, despite increasing atmospheric river activity, as the mean cooling muted the impacts of warm intrusion until recently. Our results highlight how the entire continent could become vulnerable in a warmer climate or enhanced tropical convective activity, as illustrated by the March 2022 heatwave.

Nuclear war has the potential to cause global catastrophe, as soot from widespread fires blocks sunlight, triggers abrupt cooling, and collapses agriculture. This study examines the long-term response of weather and climate patterns to nuclear conflict, simulated by injecting ~150 million tons of smoke into the stratosphere within the Community Earth System Model. An algorithm detecting extreme moisture transport events is then applied to evaluate global hydrological impacts. The simulations reveal widespread declines in event frequency, particularly during each hemisphere’s summer, but also narrow, near zonal, ocean-to-continent corridors of sharp increases during the cold season, including across the North Pacific to the U.S. West Coast, the eastern North Atlantic to North Africa and South Asia, the South Pacific to Chile, and the South Indian Ocean to Australia. In these regions, extreme moisture transport events make disproportionately large contributions to total precipitation. Changes in event distribution are largely driven by circulation changes following the smoke injection: subtropical jets contract equatorward, continental surface anticyclones intensify, and anomalously strengthened offshore cyclones establish steep coastal pressure gradients, favoring enhanced moisture transport in atmospheric river (AR)-like environments. Over southwest North America, these events feature a classic AR-like pattern during all seasons following the smoke injection, contrasting with the control simulation where AR-like characteristics appear only during cold seasons. This study offers novel insight into how nuclear conflict could profoundly disrupt the global hydrological cycle and extreme precipitation, with critical implications for regional water resources in a severely perturbed climate system.

Snowfall is the primary contributor to Antarctic surface mass balance. Identifying regional‐scale mechanisms that drive heavy snowfall provides context for changes in Antarctic surface mass balance in a warmer climate. We compare drivers of top snowfall days across five Antarctic regions using machine learning and traditional synoptic diagnostics. A convolutional neural network identifies top snow days with an accuracy of 92%–94% per region when trained on just atmospheric moisture and low‐level meridional wind, highlighting the importance of atmospheric river‐like structures to top Antarctic snowfall days. The network's skill depends mainly on low‐level wind in East Antarctica and atmospheric moisture in West Antarctica, suggesting that dynamic processes are comparatively more important in driving East Antarctic top snowfall days. We leverage the quasi‐geostrophic omega equation to identify mechanisms for ascent and snowfall production, and we find that East Antarctic top snowfall days feature stronger synoptic‐scale forcing for ascent compared to West Antarctica.

Despite their relative rarity, atmospheric rivers are key contributors to the surface mass balance of Antarctica. However, the future role of atmospheric rivers in modulating Antarctic sea-level contributions is a major area of uncertainty. Here, we leverage high-resolution climate simulations to show that Antarctic atmospheric rivers are highly sensitive to future increases in atmospheric moisture, leading to a doubling of atmospheric river frequencies and 2.5 × increase in precipitation from 2066–2100 under present-day thresholds for atmospheric river detection. However, future precipitation impacts are critically dependent on the detection threshold: accounting for moisture increases in the threshold produces smaller, regional changes in atmospheric river frequency, primarily resulting from an eastward shift in the polar jet maximum wind speeds. Our results underscore the importance of using large ensembles to quantify Antarctic atmospheric river responses to variability in projected moisture, which may not be captured when using only a few ensemble members.