Abstract This study analyzes storm-total precipitation forecasts from the National Centers for Environmental Prediction Global Forecast System (GFS) and European Centre for Medium-Range Weather Forecasts (ECMWF) for the contiguous United States (CONUS) landfalling tropical cyclones (TCs) from 2018 to 2024 using an object-based methodology. Forecast errors are quantified in terms of location, precipitation intensity, and size errors for 1- and 5-in. threshold precipitation clusters for 40 cases from 36 landfalling TCs. While there are few signs of forecast error improvement as forecast lead time decreases, the GFS consistently underforecasts precipitation intensities within 1-in. clusters when compared to the ECMWF. In addition, cluster forecast errors are stratified by different TC-related characteristics. Overall, there is no systematic relationship between any cluster forecast error and TC track forecast errors; however, high track error cases have centroid location errors 75–100 km greater than low track error cases for 5-in. threshold clusters in both models. The most robust difference in cluster forecast errors is by landfall intensity, where 5-in. clusters disproportionally occur in hurricanes. Moreover, when compared to hurricanes, tropical depressions and storms exhibit higher 5-in. cluster location errors in the GFS and underforecast all amounts of precipitation in 5-in. clusters in both models. In addition, a GFS forecast for Hurricane Henri (2021) is examined to investigate potential factors leading to a poor location forecast. The GFS forecast appears to more slowly advance a nearby cutoff low, leading to a weaker interaction between the low and Henri and resulting in precipitation being forecast for the wrong location. Significance Statement Tropical cyclone (TC) precipitation forecast verification has traditionally been done using methods that can struggle to provide information on error sources that could be used to improve forecasts. Our study verifies TC precipitation forecasts from 2018 to 2024 using object-based methods that independently assess spatial and intensity errors. We found that the GFS underforecasts the amount of precipitation within 1-in. objects compared to the ECMWF. Additionally, the GFS and ECMWF tend to underforecast all amounts of precipitation for TCs that make landfall as a tropical depression or storm compared to TCs that make landfall as a hurricane. We also looked closer at a forecast from Hurricane Henri to understand what led to a poor precipitation location forecast.

Rainfall-related hazards pose some of the most severe threats from tropical cyclones (TCs) and can be challenging to forecast. Compared to traditional dynamical models, statistical and parametric models offer a computationally inexpensive way to predict probabilistic precipitation forecasts for landfalling TCs. In this study, we systematically evaluate the probabilistic precipitation forecast performance of the ensemble Parametric Hurricane Rainfall Model (ePHRaM) for landfalling TCs in the 2017-23 hurricane seasons and compare its skill against both global [Global Ensemble Forecast System (GEFS) National Oceanic and Atmospheric Administration (NOAA) and Integrated Forecasting System (IFS) European Centre for Medium-Range Weather Forecasts (ECMWF)] and regional model ensembles [Hurricane Weather Research and Forecasting (HWRF) Model (NOAA)]. For these cases, ePHRaM forecasts have some degree of reliability for light precipitation thresholds [1 and 3 in. (120 h)-1]; however, the model loses its ability to distinguish an event versus nonevent for higher precipitation rates [5 and 7 in. (120 h)-1]. Conversely, dynamical models have overall better reliability, especially for precipitation thresholds less than 7 in. (120 h)-1. These results are further supported by the Brier skill score (BSS) and continuous ranked probability skill score (CRPSS), which show that the ePHRaM overall performs worse than the dynamical models. Individual case studies suggest that ePHRaM's forecasts are especially less skillful for cases with large precipitation asymmetries due to interaction with nearby synoptic features.

This study investigates the impact of dropwindsonde data from NOAA Gulfstream IV (G-IV) synoptic surveillance missions on Atlantic basin tropical cyclone (TC) track forecasts from the European Centre for Medium-Range Weather Forecasts (ECMWF) ensemble prediction system (EPS) and the National Centers for Environmental Prediction Global Ensemble Forecast System (GEFS) from 2018 to 2022. Track error and forecast skill are computed for the 69 forecast initializations that assimilated G-IV data and are compared to the 712 forecast initializations from the same period, but without these data, to quantify forecast impacts. Overall, ECMWF EPS and GEFS track forecasts containing G-IV data are up to 24% more skillful than forecasts without the assimilation of G-IV data. Moreover, these results suggest that the greatest positive impact on track forecast skill occurs during the first forecast initialized with G-IV data for each TC. Additionally, two case studies, Hurricanes Marco and Zeta (2020), which are characterized by notable track forecast improvements relative to the forecast prior to assimilating G-IV data, are analyzed to quantify how the G-IV data may have altered the steering flow and hence the track forecast. The track error reduction for Hurricane Marco appears to be the result of a change in the 0-h position that placed the storm in a more westerly steering flow, resulting in a more eastern track. Meanwhile, Hurricane Zeta had a 5-kt (1 kt ' 0.51 m s21) reduction in the along-track steering motion ahead of the storm following assimilation of the G-IV data, yielding a forecast position closer to the best track.

The forecast motion of tropical cyclones (TCs) critically depends on the evolution of the layer-averaged steering flow, which is associated with features proximate and remote to the TC. Given this, it is of interest to objectively identify the locations and aspects of the steering flow that will have the biggest impact on subsequent TC track forecasts, which in turn could be used to identify where to take supplemental observations, such as from aircraft, or extra rawinsondes. This paper describes the application of the ensemble-based sensitivity method to evaluate the sensitivity of TC track forecasts, which was used for synoptic surveillance flight planning for 55 potential missions during the 2019-21 seasons. TC track sensitivity can be calculated from either operational ECMWF or GEFS ensemble output (following the GEFS upgrade to version 12). Several automated methods are developed and described that provide sensitivity guidance that is useful and can be quickly interpreted, including a time-integrated track metric, and defining the steering wind within a coordinate framework along the axis of greatest position variability or vorticity. For the majority of cases, the sensitivity to the steering wind is maximized within 500 km of the TC center, particularly in the vicinity of nearby weaknesses in the subtropical ridge, with comparatively less sensitivity to upstream midlatitude features. SIGNIFICANCE STATEMENT: Tropical cyclone motion is primarily influenced by the steering flow, which is determined by the evolution of various atmospheric features, and can be characterized by large uncertainties due to the lack of observations and/or low predictability. This study describes an ensemble-based method of predicting where uncertainties in the steering flow have the largest impact on subsequent tropical cyclone forecasts, which can be used to identify locations where additional aircraft observations may benefit the forecast. Over a 3-yr period, tropical cyclone track forecasts were found to be most sensitive to the steering flow within 500 km of the center of the tropical cyclone, with some additional sensitivity to weaknesses in the subtropical ridge, which allows the TC to move poleward. Future work will expand this methodology toward computing sensitivity for tropical cyclone hazard forecasts.

Summer Arctic cyclones (ACs) are long-lived, synoptic-scale features associated with strong winds and precipitation that can lead to reductions in sea ice. Consequently, a better understanding of the upper-tropospheric low-level thermodynamic features that limit the predictability of these storms may lead to better sea ice predictions. Here, the position and intensity variability of an August 2013 AC are assessed using ensemble-based sensitivity and compared to a midlatitude cyclone. Results suggest that the position and intensity variability of these storms are largely associated with both upstream and downstream uncertainty of mesoscale features embedded within the larger-scale potential vorticity features associated with cyclogenesis. Further, the intensity variability of the AC is also associated with uncertainty to a thermal boundary near the cyclone, whereas the Atlantic basin cyclone's intensity variability is sensitive to low-level temperature and moisture along the polar front preceding the cyclone. SIGNIFICANCE STATEMENT: Arctic cyclones (ACs) are long-lived, large-scale storms that are a common occurrence during the summer. These storms can have large implications on sea ice concentration and extent. Therefore, it is important to understand how the uncertainty and dynamics of features at the top of the troposphere and near the surface impact the predictability of the storm position and intensity. The same methodology is applied to a midlatitude cyclone for comparison, as midlatitude predictability has been studied in depth. Results suggest that both the position and strength of these storms are dependent on the shape of the upper-level feature. Further, the strength of the AC is also dependent on the magnitude of a frontal boundary along the coast, whereas the strength of the midlatitude cyclone depends on the thermal boundary along the upper-level front merging with the surface cyclone.

Accurate forecasting of atmospheric river (AR) landfall is vital for western U.S. water management and flood risk mitigation. To address observation gaps and improve forecasts, the AR Reconnaissance (AR Recon) program was created as a research and operations partnership (RAOP) led by the Center for Western Weather and Water Extremes (CW3E) and the National Oceanic and Atmospheric Administration (NOAA), with participants from international, federal, and state agencies, universities, and industry. This paper summarizes AR Recon planning processes, targeted sampling strategies, forecast impacts, scientific advancements, and lessons learned, and also highlights benefits for water management and flood risk reduction. Since 2016, AR Recon has used NOAA and U.S. Air Force aircraft to release dropsondes over the North Pacific, collecting meteorological data for real-time operational use and research. Complementary observations include airborne radio occultation, radiosondes launched from the West Coast, and barometer-equipped drifting buoys alleviating oceanic data gaps. From November to March, when an AR is expected to affect the West Coast within approximately a week, AR Recon initiates daily forecast meetings with a forecast briefing, quantitative tool synthesis, and flight track design. Flights are tailored to sample essential atmospheric structures (e.g., AR cores, edges, jets, troughs, vorticity anomalies, mesoscale frontal waves, and extratropical cyclones). Ensemble and adjoint sensitivity tools support flight planning. A Mission Director guides the decisions on whether to fly and where, emphasizing targeting of essential atmospheric structures and considering operational benefits and science objectives. This RAOP approach enhances forecast accuracy in the western United States and beyond, and advances science.

Global Forecast System (GFS), North American Mesoscale Forecast System (NAM), and High -Resolution Rapid Refresh (HRRR) 2-m temperature, 10-m wind speed, and precipitation accumulation forecasts initialized at 1200 UTC are verified against New York State Mesonet (NYSM) observations from 1 January 2018 through 31 December 2021. NYSM observations at 126 site locations are used to calculate standard error statistics (e.g., forecast error, root -mean -square error) for temperature and wind speed and contingency table statistics for precipitation across forecast hours, meteorological seasons, and regions. The majority of the focus is placed on the first 18 forecast hours to allow for comparison among all three models. A daily NYSM station -mean temperature error analysis identified a slight cold bias at temperatures below 25 degrees C in the GFS, a cool -to -warm bias as forecast temperatures warm in the HRRR, and a warm bias at temperatures above 30 degrees C in each model. Differences arise when considering temperature biases with respect to lead times and seasons. Wind speeds are overforecast at all ranges in each season, and forecast wind speeds $ 18 m s21 are rarely observed. Performance diagrams indicate overall good forecast performance at precipitation thresholds of 0.1-1.5 mm, but with a high frequency bias in the GFS and NAM. This paper provides an overview of deterministic forecast performance across New York State, with the aim of sharing common biases associated with temperature, wind speed, and precipitation with operational forecasters and is the first step in developing a real-time model forecast uncertainty prediction tool.

The NASA Convective Processes Experiment - Cabo Verde (CPEX-CV) field campaign took place in September 2022 out of Sal Island, Cabo Verde. A unique payload aboard the NASA DC-8 aircraft equipped with advanced remote sensing and in situ instrumentation, in conjunction with radiosonde launches and satellite observations, allowed CPEX-CV to target the coupling between atmospheric dynamics, marine boundary layer properties, convection, and the dust-laden Saharan Air Layer in the data-sparse tropical East Atlantic region. CPEX-CV provided measurements of African Easterly Wave environments, diurnal cycle impacts on convective lifecycle, and several Saharan dust outbreaks, including the highest dust optical depth observed by the DC-8 interacting with what would become Tropical Storm Hermine. Preliminary results from CPEX-CV underscore the positive impact of dedicated tropical East Atlantic observations on downstream forecast skill, including sampling environmental forcings impacting the development of several non-developing and developing convective systems such as Hurricanes Fiona and Ian. Combined airborne radar, lidar, and radiometer measurements uniquely provide near-storm environments associated with convection on various spatiotemporal scales and, with in situ observations, insights into controls on Saharan dust properties with transport. The DC-8 also collaborated with the European Space Agency to perform coordinated validation flights under the Aeolus spaceborne wind lidar and over the Mindelo ground site, highlighting the enhanced sampling potential through partnership opportunities. CPEX-CV engaged in professional development through dedicated team building exercises that equipped the team with a cohesive approach for targeting CPEX-CV science objectives and promoted active participation of scientists across all career stages.

The THINICE field campaign, based in Svalbard in August 2022, provided unique observations of summertime Arctic cyclones, their coupling with cloud cover, and their interactions with tropopause polar vortices and sea ice conditions. THINICE was motivated by the need to advance our understanding of these processes and to improve coupled models used to forecast weather and sea ice, as well as long-term projections of climate change in the Arctic. Two research aircraft were deployed with complementary instrumentation. The Service des Avions Fran & ccedil;ais Instrument & eacute;s pour la Recherche en Environnement (Safire) Aerei da Trasporto Regionale 42 (ATR42) aircraft, equipped with the radar-lidar (RALI) remote sensing instrumentation and in situ cloud microphysics probes, flew in the midtroposphere to observe the wind and multiphase cloud structure of Arctic cyclones. The British Antarctic Survey Meteorological Airborne Science Instrumentation (MASIN) aircraft flew at low levels measuring sea ice properties, including surface brightness temperature, albedo and roughness, and the turbulent fluxes that mediate exchange of heat and momentum between the atmosphere and the surface. Long-duration instrumented balloons, operated by WindBorne Systems, sampled meteorological conditions within both cyclones and tropospheric polar vortices across the Arctic. Several novel findings are highlighted. Intense, shallow low-level jets along warm fronts were observed within three Arctic cyclones using the Doppler radar and turbulence probes. A detailed depiction of the interweaving layers of ice crystals and supercooled liquid water in mixed-phase clouds is revealed through the synergistic

During a 6-day intensive observing period in January 2021, Atmospheric River Reconnaissance (AR Recon) aircraft sampled a series of atmospheric rivers (ARs) over the northeastern Pacific that caused heavy precipitation over coastal California and the Sierra Nevada. Using these observations, data denial experiments were conducted with a regional modeling and data assimilation system to explore the impacts of research flight frequency and spatial resolution of dropsondes on model analyses and forecasts. Results indicate that dropsondes significantly improve the representation of ARs in the model analyses and positively impact the forecast skill of ARs and quantitative precipitation forecasts (QPF), particularly for lead times . 1 day. Both reduced mission frequency and reduced dropsonde horizontal resolution degrade forecast skill. On the other hand, experiments that assimilated only G -IV data and experiments that assimilated both G -IV and C-130 data show better forecast skill than experiments that only assimilated C-130 data, suggesting that the additional information provided by G -IV data is necessary for improving forecast skill. Although this is a case study, the 6-day period studied encompassed multiple AR events that are representative of typical AR behavior. Therefore, the results indicate that future operational AR Recon missions incorporate daily mission or back-to-back flights, maintain current dropsonde spacing, support high-resolution data transfer capacity on the C -130s, and utilize G -IV aircraft in addition to C -130s.

© 2023 American Meteorological Society. This is an Author Accepted Manuscript distributed under the terms of the default AMS reuse license. For information regarding reuse and general copyright information, consult the AMS Copyright Policy (www.ametsoc.org/PUBSReuseLicenses). Corresponding author: David Lavers, david.lavers@ecmwf.int

Satellites provide the largest dataset for monitoring the earth system and constraining analyses in numerical weather prediction models. A significant challenge for utilizing satellite radiances is the accurate estimation of their biases. High-accuracy non-radiance data are commonly employed to anchor radiance bias corrections. However, aside from the impacts of radio occultation data in the stratosphere, the influence of other types of “anchor” observation data on radiance assimilation remain unclear. This study provides an assessment of impacts of dropsonde data collected during the Atmospheric River (AR) Reconnaissance program, which samples ARs over the Northeast Pacific, on the radiance assimilation using the Global Forecast System (GFS) and Global Data Assimilation System at National Centers for Environmental Prediction. The assimilation of this dropsonde dataset has proven crucial for providing enhanced anchoring for bias corrections and improving the model background, leading to an increase of ~5–10% in the number of assimilated microwave radiance in the lower/middle troposphere over the Northeast Pacific and North America. The impact on tropospheric infrared radiance is small but also beneficial. Impacts of dropsondes on the use of stratospheric channels are minimal due to the absence of dropsonde observations at certain altitudes, such as aircraft flight levels (e.g., 150 hPa). Results in this study underscore the usefulness of dropsondes, along with other conventional data, in optimizing the assimilation of satellite radiance. This study reinforces the importance of a diverse observing network for accurate weather forecasting and highlights the specific benefits derived from integrating dropsonde data into radiance assimilation processes.

The multiscale nature of tropical cyclone (TC) intensity change under moderate vertical wind shear was explored through an ensemble of high-resolution simulations of Hurricane Gonzalo (2014). Ensemble intensity forecasts were characterized by large short-term (36-h) uncertainty, with a forecast intensity spread of over 20 m s(-1), due to differences in the timing of rapid intensification (RI) onset. Two subsets of ensemble members were examined, referred to as early-RI and late-RI members. The two ensemble groups displayed significantly different vortex evolutions under the influence of a nearby upper-tropospheric trough and an associated dry-air intrusion. Mid-to-upper-tropospheric ventilation in late-RI members was linked to a disruption of inner-core diabatic heating, a more tilted vortex, and vortex breakdown, as the simulated TCs transitioned from a vorticity annulus toward a monopole structure. A column-integrated moist static energy (MSE) budget revealed the important role of horizontal advection in depleting MSE from the TC core, while mesoscale subsidence beneath the dry-air intrusion acted to dry a deep layer of the troposphere. Eventually, the dry-air intrusion retreated from late-RI members as vertical wind shear weakened, the magnitude of vortex tilt decreased, and late-RI members began to rapidly intensify, ultimately reaching a similar intensity as early-RI members. Conversely, the vortex structures of early-RI members were shown to exhibit greater intrinsic resilience to tilting from vertical wind shear, and early-RI members were able to fend off the dry-air intrusion relatively unscathed. The different TC intensity evolutions can be traced back to differences in the initial TC vortex structure and intensity. Significance StatementDespite recent advances, tropical cyclone intensity forecasts struggle to accurately predict episodes of rapid intensification. Such forecasts become increasingly challenging when a storm is embedded within an environment of moderate vertical wind shear. This study uses an ensemble of high-resolution simulations to examine how environmental influences can affect the tropical cyclone vortex and precipitation structure, which, in turn, modulate the intensity of the storm and the onset of rapid intensification. We propose a feedback that exists where slightly weaker and less resilient vortices are more susceptible to ventilation from dry, environmental air, aided in part by differential advection from the tilted circulation, resulting in a degradation of vortex organization and a delayed onset of rapid intensification.

Prediction of the potentially devastating impact of landfalling tropical cyclones (TCs) relies substantially on numerical prediction systems. Due to the limited predictability of TCs and the need to express forecast confidence and possible scenarios, it is vital to exploit the benefits of dynamic ensemble forecasts in operational TC forecasts and warnings. RSMCs, TCWCs, and other forecast centers value probabilistic guidance for TCs, but the International Workshop on Tropical Cyclones (IWTC-9) found that the “pull-through” of probabilistic information to operational warnings using those forecasts is slow. IWTC-9 recommendations led to the formation of the WMO/WWRP Tropical Cyclone-Probabilistic Forecast Products (TC-PFP) project, which is also endorsed as a WMO Seamless GDPFS Pilot Project. The main goal of TC-PFP is to coordinate across forecast centers to help identify best practice guidance for probabilistic TC forecasts. TC-PFP is being implemented in 3 phases: Phase 1 (TC formation and position); Phase 2 (TC intensity and structure); and Phase 3 (TC related rainfall and storm surge). This article provides a summary of Phase 1 and reviews the current state of the science of probabilistic forecasting of TC formation and position. There is considerable variability in the nature and interpretation of forecast products based on ensemble information, making it challenging to transfer knowledge of best practices across forecast centers. Communication among forecast centers regarding the effectiveness of different approaches would be helpful for conveying best practices. Close collaboration with experts experienced in communicating complex probabilistic TC information and sharing of best practices between centers would help to ensure effective decisions can be made based on TC forecasts. Finally, forecast centers need timely access to ensemble information that has consistent, user-friendly ensemble information. Greater consistency across forecast centers in data accessibility, probabilistic forecast products, and warnings and their communication to users will produce more reliable information and support improved outcomes.

The progress of research and forecast techniques for tropical cyclone (TC) unusual tracks (UTs) in recent years is reviewed. A major research focus has been understanding which processes contribute to the evolution of the TC and steering flow over time, especially the reasons for the sharp changes in TC motion over a short period of time. When TCs are located in the vicinity of monsoon gyres, TC track forecast become more difficult to forecast due to the complex interaction between the TCs and the gyres. Moreover, the convection and latent heat can also feed back into the synoptic-scale features and in turn modify the steering flow. In this report, two cases with UTs are examined, along with an assessment of numerical model forecasts. Advances in numerical modelling and in particular the development of ensemble forecasting systems have proved beneficial in the prediction of such TCs. There are still great challenges in operational track forecasts and warnings, such as the initial TC track forecast, which is based on a poor pre-genesis analysis, TC track forecasts during interaction between two or more TCs and track predictions after landfall. Recently, artificial intelligence (AI) methods such as machine learning or deep learning have been widely applied in the field of TC forecasting. For TC track forecasting, a more effective method of center location is obtained by combining data from various sources and fully exploring the potential of AI, which provides more possibilities for improving TC prediction.

The term jet stream generally refers to a narrow region of intense winds near the top of the midlatitude or subtropical troposphere. It is in the midlatitude jet stream where instabilities and waves may develop into synoptic-scale systems, which in turn makes accurately resolving the structure of the jet stream and associated features critical for atmospheric development, predictability, and impacts, such as extreme precipitation and winds. Using dropwindsonde observations collected during the Atmospheric River Reconnaissance (AR Recon) campaign from 2020 to 2022, this study assesses the North Pacific jet stream structure in the European Centre for Medium-Range Weather Forecasts (ECMWF) Integrated Forecasting System (IFS). Results show that the IFS has a slow-wind bias on the lead times assessed, with the strongest winds (& GE;50 m & BULL;s(-1)) having a bias of up to -1.88 m & BULL;s(-1) on forecast day 4. Also, the IFS cannot resolve the sharp potential vorticity (PV) gradient across the jet stream and tropopause, and this PV gradient weakens with forecast lead time. Cases with larger wind biases are characterized by higher PV biases and PV biases tend to be larger for cases with a higher horizontal PV gradient. These results suggest that further model-based experiments are needed to identify and address these biases, which could ultimately yield increased forecast accuracy.

Hurricane Dorian (2019), a category-5 tropical cyclone (TC), was characterized by a large spread in track forecasts as it moved northwest. A set of 80 ensemble forecasts from the Hurricane Analysis and Forecast System (HAFS) was produced to evaluate Dorian's track spread and the factors that contributed to it. Track spread was particularly critical at long lead times (5-7 days after initialization near the Lesser Antilles), because of the uncertainty in the location of landfall and hazards. Four clusters of members were analyzed based on the 7-day track, characterized by Dorian moving: 1) slowly near the northern Bahamas (closest to reality), 2) across the Florida Peninsula, 3) slowly into Florida's east coast, and 4) quickly north of the Bahamas. Ensemble sensitivity techniques were applied to identify areas that were most critical for Dorian's track. Key differences were found in the strength of the subtropical ridge over the western Atlantic Ocean with a weaker ridge and slower easterly steering flow in the offshore groups. Subtle differences in the synoptic pattern over the United States also appeared to affect the timing of Dorian's northward turn, specifically the strength of a shortwave trough moving over the Ohio Valley. Despite some early track differences, the correlation between early and late track errors was not significant. An examination of four members further highlights the differences in steering and the strength of the subtropical ridge. This study demonstrates the utility of ensemble datasets for studying TC forecast uncertainty and the importance of medium-range modeling of synoptic-scale steering features to accurately predict the track of tropical cyclones. Significance StatementHurricane Dorian was a catastrophic hurricane for the Bahamas and got very close to Florida without directly impacting the state. Some early forecasts showed the storm moving directly into or across Florida; others correctly showed the storm stalling over the Bahamas and then turning northward. This track forecast uncertainty made preparations in Florida challenging; therefore, we wanted to better understand why Dorian took the track that it did, to see what this tells us about the factors that affect hurricane tracks, and learn for future storms. We looked at an ensemble of 80 runs of a hurricane model, initiated at the same time. Some runs showed a Florida landfall; others showed Dorian stalling over the Bahamas. The strength of the subtropical ridge over the Atlantic north of Dorian and an upper-level trough of low pressure over the United States were key influences on storm path. These two large-scale features were better forecast in the ensemble members that correctly showed Dorian stalling and turning northward. This study shows how useful ensembles can be for understanding the processes driving hurricane motion and also shows that it is critical to forecast multiple synoptic-scale features correctly to accurately predict a hurricane's track 5-7 days in advance.

Tropical cyclone (TC) intensity has been shown to have limited predictability in numerical weather prediction models; therefore, ensemble forecasting may be critical. An ensemble prediction system (EPS) should ideally cover all sources of uncertainty; however, most meso- and convective-scale EPSs typically consider initial-condition uncertainty alone, with limited treatment of model uncertainty, even though the evolution of mesoscale features is highly dependent on uncertain parameterization schemes. The role of stochastic treatment of model error in the Hurricane Weather Research and Forecasting (HWRF) EPS is evaluated by applying independent stochastically perturbed parameterization (iSPPT) scheme to individual parameterization schemes for four TCs from 2017 to 2018. Experiments with Hurricane Irma (2017) indicate that TC intensity ensemble standard deviation is most sensitive to the amplitude of the stochastic perturbation field, with smaller impact from adjusting the decorrelation time scale and spatial length scale. Results from all four TC cases show that stochastic perturbations to the turbulent mixing scheme can increase the ensemble standard deviation in intensity metrics over a 72-h simulation without introducing significant differences in mean error or bias. By contrast, stochastic perturbations to the microphysics, radiation, and cumulus tendencies have negligible effects on intensity standard deviation.

To better understand the conditions that favor tropical cyclone (TC) rapid intensification (RI), this study assesses environmental and storm-scale characteristics that differentiate TCs that undergo RI from TCs that undergo slow intensification (SI). This comparison is performed between analog TC pairs that have similar initial intensity, vertical wind shear, and maximum potential intensity. Differences in the characteristics of RI and SI TCs in the North Atlantic and western North Pacific basins are evaluated by compositing and comparing data from the fifth-generation European Centre for Medium-Range Weather Forecasts (ECMWF) Re-Analysis (ERA5) and the Gridded Satellite (GridSat) dataset. In the period leading up to the start of RI, RI TCs tend to have a stronger and deeper vortex that is more vertically aligned than SI TCs. Additionally, surface latent heat fluxes are significantly larger in RI TCs prior to the intensity change period, compared to SI TCs. The largest surface latent heat flux differences are initially located to the left of shear; subsequently, upshear and right-of-shear differences amplify, resulting in a more symmetric distribution of surface latent heat fluxes in RI TCs. Increasing azimuthal symmetry of surface latent heat fluxes in RI TCs, together with an increasing azimuthal symmetry of horizontal moisture flux convergence, promote the upshear migration of convection in RI TCs. These differences, and their evolution before and during the intensity change period, are hypothesized to support the persistence and invigoration of upshear convection and, thus, a more symmetric latent heating pattern that favors RI.