Hurricane Intensity Prediction Research Articles

Quantifying the Environmental Effects on Tropical Cyclone Intensity Change Using a Simple Dynamically Based Dynamical System Model

Abstract Accurate prediction of tropical cyclone (TC) intensity is quite challenging due to multiple competing processes among the TC internal dynamics and the environment. Most previous studies have evaluated the environmental effects on TC intensity change from both internal dynamics and external influence. This study quantifies the environmental effects on TC intensity change using a simple dynamically based dynamical system (DBDS) model recently developed. In this simple model, the environmental effects are uniquely represented by a ventilation parameter B, which can be expressed as multiplicative of individual ventilation parameters of the corresponding environmental effects. Their individual ventilation parameters imply their relative importance to the bulk environmental ventilation effect and thus to the TC intensity change. Six environmental factors known to affect TC intensity change are evaluated in the DBDS model using machine learning approaches with the best track data for TCs over the North Atlantic, central, eastern, and western North Pacific and the Statistical Hurricane Intensity Prediction Scheme (SHIPS) dataset during 1982–2021. Results show that the deep-layer vertical wind shear (VWS) is the dominant ventilation factor to reduce the intrinsic TC intensification rate or to drive the TC weakening, with its ventilation parameter ranging between 0.5 and 0.8 when environmental VWS between 200 and 850 hPa is larger than 8 m s−1. Other environmental factors are generally secondary, with their respective ventilation parameters over 0.8. An interesting result is the strong dependence of the environmental effects on the stage of TC development.

Applying Adaptive Neuro-Fuzzy Inference System to Improve Typhoon Intensity Forecast in the Northwest Pacific

Typhoon intensity forecast is an important issue. The objective of this study is to construct a 5-day 12-hourly typhoon intensity forecast model based on the adaptive neuro-fuzzy inference systems (ANFIS) to improve the typhoon intensity forecast in the Northwest Pacific. It analyzed the improvement of the ANFIS typhoon intensity forecast model by comparing it with the MLR model when only the atmospheric factor or both atmospheric and oceanic factors are considered. This study collected the SHIPS (Statistical Hurricane Intensity Prediction Scheme) developmental data of typhoons in the Northwest Pacific before landing from 2000 to 2012. The input factors of the ANFIS model were simplified by the stepwise regression procedure (SRP). Subtractive clustering (SC) was used to determine the number of ANFIS rules and to reduce model complexity. Model Index (MI) was taken as the clustering standard of SC to determine the network architecture of the ANFIS typhoon intensity forecast model. The simulated results show that the MI could effectively determine the radius of influence of SC. The typhoon intensity forecast was significantly improved after oceanic environmental factors were added. The improvement of RMSE of ANFIS was the highest at 84 h; the improvement of ANFIS on the underestimated ratio was primarily positive. The Typhoon Songda case study shows that the maximum bias of ANFIS is greatly improved, at 60 h of the lead time, and the improvement percentage of maximum bias is the highest (39%). Overall, the ANFIS model could effectively improve the MLR model in typhoon intensity forecast.

Lightning-Based Tropical Cyclone Rapid Intensification Guidance

Abstract With several seasons of Geostationary Lightning Mapper (GLM) data, this work revisits incorporating lightning observations into operational tropical cyclone rapid intensification guidance. GLM provides freely available, real-time lightning data over the central and eastern North Pacific and North Atlantic Oceans. A long-term lightning dataset is needed to use GLM in a statistical–dynamical operational application to capture the relationship between lightning and the rare occurrence of rapid intensification. This work uses the World Wide Lightning Location Network (WWLLN) dataset from 2005 to 2017 to develop lightning-based predictors for rapid intensification guidance models. The models mimic the operational Statistical Hurricane Intensity Prediction Scheme Rapid Intensification Index and Rapid Intensification Prediction Aid frameworks. The frameworks are averaged to form a consensus as a means to isolate the impact of the lightning predictors. Two configurations for lightning predictors are assessed: a spatial configuration with 0–100-km inner core and 200–300-km rainband area for the preceding 6-h predictors and a temporal configuration with an inner core only for the preceding 0–1, 0–6, and 6–12 h. When tested on the 2018–21 seasons, the temporal configuration adds skill primarily to the 12–48-h forecasts when compared to the no-lightning version and rapid intensification operational consensus. When WWLLN is replaced with GLM, minor changes to the prediction are observed suggesting that this approach is suitable for operational applications and provides a new baseline for tropical cyclone lightning-based rapid intensification aids. Significance Statement The forecasting of rare, yet critical, tropical cyclone rapid intensification events continues to be challenging. The current operational tools to anticipate rapid intensity changes use a combination of numerical weather prediction–derived environmental conditions and satellite-based cloud top temperature variations of deep convection. Here, we use freely available Geostationary Lightning Mapper data, which provide independent information about convection, in similar intensity guidance frameworks using temporal and spatial aspects of lightning variability. Our results show an improvement in short-term (12–48 h) rapid intensification forecasts by using temporal lightning information, and our investigation highlights that users of Geostationary Lightning Mapper lightning information should be cognizant of the influence and impact of land on these observations.

A Rapid Intensification Warning Index for Tropical Cyclones Based on the Analog Method

AbstractPrediction of the rapid intensification (RI) of tropical cyclones (TCs) remains challenging. In this paper, by using information from the early period following TC formation, the rapid intensification warning index (RIWI) is developed based on the analog method. A 10‐year cross‐validation and data from Hurricane Ida (2021) are used to verify its potential application. Results show that the RIWI can efficiently discriminate between RI and non‐RI storms and has a significant positive correlation with the lifetime maximum intensity (LMI) of the TCs. By using this index, an early warning can be issued ∼30 hr before the onset of RI, which is much earlier than the predictions made using the probabilistic Statistical Hurricane Intensity Prediction System RI index. In addition, by using the RIWI as a predictor, the prediction of LMI provides an early estimate of TC severity.

Investigating Tropical Cyclone Rapid Intensification with an Advanced Artificial Intelligence System and Gridded Reanalysis Data

Rapid Intensification (RI) in Tropical Cyclone (TC) development is one of the most difficult and still challenging tasks in weather forecasting. In addition to the dynamical numerical simulations, commonly used techniques for RI (as well as TC intensity changes) analysis and prediction are the composite analysis and statistical models based on features derived from the composite analysis. Quite a large number of such selected and pre-determined features related to TC intensity change and RI have been accumulated by the domain scientists, such as those in the widely used SHIPS (Statistical Hurricane Intensity Prediction Scheme) database. Moreover, new features are still being added with new algorithms and/or newly available datasets. However, there are very few unified frameworks for systematically distilling features from a comprehensive data source. One such unified Artificial Intelligence (AI) system was developed for deriving features from TC centers, and here, we expand that system to large-scale environmental condition. In this study, we implemented a deep learning algorithm, the Convolutional Neural Network (CNN), to the European Centre for Medium-Range Weather Forecasts (ECMWF) ERA-Interim reanalysis data and identified and refined potentially new features relevant to RI such as specific humidity in east or northeast, vorticity and horizontal wind in north and south relative to the TC centers, as well as ozone at high altitudes that could help the prediction and understanding of the occurrence of RI based on the deep learning network (named TCNET in this study). By combining the newly derived features and the features from the SHIPS database, the RI prediction performance can be improved by 43%, 23%, and 30% in terms of Kappa, probability of detection (POD), and false alarm rate (FAR) against the same modern classification model but with the SHIPS inputs only.

Estimating the tropical cyclone wind structure using physics-incorporated networks

Satellite-based remote sensing technology plays a significant role in identifying tropical cyclones (TCs), and most of the current research focuses on intensity estimation. However, analyzing the wind structure of TCs, which is directly related to the danger they bring, remains a challenge. By adding prior knowledge of TCs into the model, we propose a physics-incorporated network based on multi-task learning to estimate wind radii and intensity, whose layers can automatically extract rotation-invariant features related to the TC core from multichannel satellite imageries. In addition, we build a more comprehensive dataset, including global Statistical Hurricane Intensity Prediction Scheme (SHIPS) predictors, to tackle the structure task. We compare our model with existing methods, and it shows that our model gets better results in estimating 50-knot and 64-knot wind radii and achieves a 4.87-knot root-mean-squared error (RMSE) of intensity. By predicting probability density functions, our model quantifies the uncertainty of the result. The experimental results show that the incorporation of rotation equivariance into the layers can enhance TC structure estimation. By considering the feature importance of multi-source predictors, we find that our model pays attention to key predictors related to the TC structure. Specifically, the tangential wind speed at 500 km from the TC center and the radius of the 5-knot wind both greatly reduce the error of the estimated parameters. Finally, two case studies show that the proposed model performs well most of the time during TCs’ rapid intensification. However, when TCs’ system is not well organized, estimating the wind structure is challenging.

Does ERA5 Mark a New Era for Resolving the Tropical Cyclone Environment?

Abstract The synoptic environment around tropical cyclones plays a significant role in vortex evolution. To capture the environment, the operational and research communities calculate diagnostic quantities. To aid with applications and research, the Tropical Cyclone Precipitation, Infrared, Microwave, and Environmental Dataset (TC PRIMED) combines disparate data sources. A key part of TC PRIMED is the environmental context. Often, environmental diagnostics come from multiple sources. However, TC PRIMED uses the European Centre for Medium-Range Weather Forecasts fifth-generation reanalysis (ERA5) product to provide a more complete representation of the storm environment from a single source. Reanalysis products usually poorly resolve tropical cyclones and their surrounding environment. To understand the uncertainty of large-scale diagnostics, ERA5 is compared to the Statistical Hurricane Intensity Prediction Scheme developmental dataset and the National Oceanic and Atmospheric Administration Gulfstream IV-SP dropwindsondes. This analysis highlights biases in the ERA5 environmental diagnostic quantities. Thermodynamic fields show the largest biases. The boundary layer exhibits a cold temperature bias that limits the amount of convective instability; also, the upper troposphere contains temperature biases and shows a high relative humidity bias. However, the upper-troposphere large-scale kinematic fields and derived metrics are low biased. In the lower troposphere, the temperature gradient and advection calculated from the thermal wind suggest that the low-level wind field is not representative of the observed distribution. These diagnostics comparisons provide uncertainty so that users of TC PRIMED can assess the implications for specific research and operational applications.

A Simple Bias and Uncertainty Scheme for Tropical Cyclone Intensity Change Forecasts

Abstract To better forecast tropical cyclone (TC) intensity change and understand forecast uncertainty, it is critical to recognize the inherent limitations of forecast models. The distributions of intensity change for statistical–dynamical models are too narrow, and some intensity change forecasts are shown to have larger errors and biases than others. The Intensity Bias and Uncertainty Scheme (IBUS) is developed in an intensity change framework, which estimates the bias and the standard deviation of intensity forecast errors. The IBUS is developed and applied to the Decay Statistical Hurricane Intensity Prediction Scheme (DSHP), the Logistic Growth Equation Model (LGEM), and official National Hurricane Center (NHC) forecasts (OFCL) separately. The analysis uses DSHP, LGEM, and OFCL forecasts from 2010 to 2019 in both the Atlantic and east Pacific basins. Each IBUS contains both a bias correction and forecast uncertainty estimate that is tested on the training dataset and evaluated on the 2020 season. The IBUS is able to reduce intensity biases and improve forecast errors beyond 120 h in each model and basin relative to the original forecasts. The IBUS is also able to communicate forecast uncertainty that explains ∼7%–11% of forecast variance at 48 h for DSHP and LGEM in the Atlantic. Better performance is found in the east Pacific at 96 h where the IBUS explains up to 30% of the errors in DSHP and 14% of the errors for LGEM. The IBUS for OFCL explains 9%–13% of the 48-h forecast uncertainty in the Atlantic and east Pacific with up to 30% variance explained for east Pacific forecasts at 96 h. IBUS for OFCL has the capability to provide intensity forecast uncertainty similar to the “cone of uncertainty” for track forecasts.

Discovering Precursors to Tropical Cyclone Rapid Intensification in the Atlantic Basin Using Spatiotemporal Data Mining

Regarded as one of the most dangerous types of natural disaster, tropical cyclones threaten the life and health of human beings and often cause enormous economic loss. However, intensity forecasting of tropical cyclones, especially rapid intensification forecasting, remains a scientific challenge due to limited understanding regarding the intensity change process. We propose an automatic knowledge discovery framework to identify potential spatiotemporal precursors to tropical cyclone rapid intensification from a set of tropical cyclone environmental fields. Specifically, this framework includes (1) formulating RI and non-RI composite environmental fields from historical tropical cyclones using NASA MERRA2 data; (2) utilizing the shared nearest neighbor-based clustering algorithm to detect regions representing relatively homogeneous behavior around tropical cyclone centers; (3) determining candidate precursors from significantly different regions in RI and non-RI groups using a spatiotemporal statistical method; and (4) comparing candidates to existing predictors to select potential precursors. The proposed knowledge discovery framework is applied separately to different factors, including 200 hPa zonal wind, 850–700 hPa relative humidity, and 850–200 hPa vertical shear, to detect potential precursors. Compared to the existing predictors manually labeled, i.e., U200 and U20C, RHLO, and SHRD in the Statistical Hurricane Intensity Prediction Scheme, our automatically discovered precursors have a comparable or better capability for estimating the probability of rapid intensification.

An Advanced Artificial Intelligence System for Identifying the Near-Core Impact Features to Tropical Cyclone Rapid Intensification from the ERA-Interim Data

Prediction of tropical cyclone (TC) intensity is one of the ground challenges in weather forecasting, and rapid intensification (RI) is a key part of that prediction. Most of the current RI studies are based on a selected variable (feature) set, which is accumulated based on expert expertise in past studies of TC intensity changes and RI. Are there any more important variables in TC intensity predictions that were not identified in past studies? A systematic and comprehensive search for those variables from vast amounts of gridded data, satellite images, and other historically collected data could be helpful for answering the above question. Artificial intelligence (AI) has the capabilities to distill features in large array data, and it is helpful in identifying new features related to TC intensity changes in general and RI in particular. Here, we leverage the local linear embedding (LLE) dimension reduction techniques to the European Centre for Medium-Range Weather Forecasts ERA-Interim reanalysis data for identifying new variables related to RI. In addition to the well-known features in the SHIPS (statistical hurricane intensity prediction scheme) database, we identified other significant features, such as 400 and 450 hPa meridional wind, 1000 hPa potential vorticity, and vertical pressure speed, that could help the understanding and prediction of RI occurrences. Furthermore, our AI system outperforms our baseline model with SHIPS data only by 26.6% and 8.4% in kappa and PSS (Peirce’s skill score), respectively.

The Effects of Environmental Wind Shear Direction on Tropical Cyclone Boundary Layer Thermodynamics and Intensity Change from Multiple Observational Datasets

Abstract The relationship between deep-layer environmental wind shear direction and tropical cyclone (TC) boundary layer thermodynamic structures is explored in multiple independent databases. Analyses derived from the tropical cyclone buoy database (TCBD) show that when TCs experience northerly component shear, the 10-m equivalent potential temperature θe tends to be more symmetric than when shear has a southerly component. The primary asymmetry in θe in TCs experiencing southerly component shear is radially outward from 2 times the radius of maximum wind speed, with the left-of-shear quadrants having lower θe by 4–6 K than the right-of-shear quadrants. As with the TCBD, an asymmetric distribution of 10-m θe for TCs experiencing southerly component shear and a symmetric distribution of 10-m θe for TCs experiencing northerly component shear was found using composite observations from dropsondes. These analyses show that differences in the degree of symmetry near the sea surface extend through the depth of the boundary layer. Additionally, mean dropsonde profiles illustrate that TCs experiencing northerly component shear are more potentially unstable between 500- and 1000-m altitude, signaling a more favorable environment for the development of surface-based convection in rainband regions. Analyses from the Statistical Hurricane Intensity Prediction Scheme (SHIPS) database show that subsequent strengthening for TCs in the Atlantic Ocean basin preferentially occurs in northerly component deep-layer environmental wind shear environments whereas subsequent weakening preferentially occurs in southerly component wind shear environments, which further illustrates that the asymmetric distribution of boundary layer thermodynamics is unfavorable for TC intensification. These differences emphasize the impact of deep-layer wind shear direction on TC intensity changes that likely result from the superposition of large-scale advection with the shear-relative asymmetries in TC structure. Significance Statement This research investigates how the direction of the winds surrounding the storm impacts the strength of a tropical cyclone. Analyses from this study illustrate that when the winds come from the south the atmospheric boundary layer has a cool and dry side along with a warm and moist side. When the large-scale winds come from the north, temperature and moisture conditions are more uniform throughout the boundary layer. Consequently, results from tropical cyclone climatology show that winds observed to come from the north favor subsequent intensification. These relationships illustrate that tropical cyclone structure and intensity are directly influenced by their surrounding environments and that knowledge of the wind environment could help to improve future forecasts of tropical cyclone intensity change.

On the Sensitivity of Hurricane Intensity and Structure to Horizontal Tracer Advection Schemes in FV3

AbstractWe investigate the sensitivity of hurricane intensity and structure to the horizontal tracer advection in the Geophysical Fluid Dynamics Laboratory (GFDL) Finite-Volume Cubed-Sphere Dynamical Core (FV3). We compare two schemes, a monotonic scheme and a less diffusive positive-definite scheme. The positive-definite scheme leads to significant improvement in the intensity prediction relative to the monotonic scheme in a suite of 5-day forecasts that mostly consist of rapidly intensifying hurricanes. Notable storm structural differences are present: the radius of maximum wind (RMW) is smaller and eyewall convection occurs farther inside the RMW when the positive-definite scheme is used. Moreover, we find that the horizontal tracer advection scheme affects the eyewall convection location by affecting the moisture distribution in the inner-core region. This study highlights the importance of dynamical core algorithms in hurricane intensity prediction.

Deep Learning Experiments for Tropical Cyclone Intensity Forecasts

AbstractReducing tropical cyclone (TC) intensity forecast errors is a challenging task that has interested the operational forecasting and research community for decades. To address this, we developed a deep learning (DL)-based Multilayer Perceptron (MLP) TC intensity prediction model. The model was trained using the global Statistical Hurricane Intensity Prediction Scheme (SHIPS) predictors to forecast the change in TC maximum wind speed for the Atlantic Basin. In the first experiment, a 24-hour forecast period was considered. To overcome sample size limitations, we adopted a Leave One Year Out (LOYO) testing scheme, where a model is trained using data from all years except one and then evaluated on the year that is left out. When tested on 2010–2018 operational data using the LOYO scheme, the MLP outperformed other statistical-dynamical models by 9-20%. Additional independent tests in 2019 and 2020 were conducted to simulate real-time operational forecasts, where the MLP model again outperformed the statistical-dynamical models by 5-22% and achieved comparable results as HWFI. The MLP model also correctly predicted more rapid intensification events than all the four operational TC intensity models compared. In the second experiment, we developed a lightweight MLP for 6-hour intensity predictions. When coupled with a synthetic TC track model, the lightweight MLP generated realistic TC intensity distribution in the Atlantic Basin. Therefore, the MLP-based approach has the potential to improve operational TC intensity forecasts, and will also be a viable option for generating synthetic TCs for climate studies.

Validation of Ensemble-Based Probabilistic Tropical Cyclone Intensity Change

Although there has been substantial improvement to numerical weather prediction models, accurate predictions of tropical cyclone rapid intensification (RI) remain elusive. The processes that govern RI, such as convection, may be inherently less predictable; therefore a probabilistic approach should be adopted. Although there have been numerous studies that have evaluated probabilistic intensity (i.e., maximum wind speed) forecasts from high resolution models, or statistical RI predictions, there has not been a comprehensive analysis of high-resolution ensemble predictions of various intensity change thresholds. Here, ensemble-based probabilities of various intensity changes are computed from experimental Hurricane Weather Research and Forecasting (HWRF) and Hurricanes in a Multi-scale Ocean-coupled Non-hydrostatic (HMON) models that were run for select cases during the 2017–2019 seasons and verified against best track data. Both the HWRF and HMON ensemble systems simulate intensity changes consistent with RI (30 knots 24 h−1; 15.4 m s−1 24 h−1) less frequent than observed, do not provide reliable probabilistic predictions, and are less skillful probabilistic forecasts relative to the Statistical Hurricane Intensity Prediction System Rapid Intensification Index (SHIPS-RII) and Deterministic to Probabilistic Statistical (DTOPS) statistical-dynamical systems. This issue is partly alleviated by applying a quantile-based bias correction scheme that preferentially adjusts the model-based intensity change at the upper-end of intensity changes. While such an approach works well for high-resolution models, this bias correction strategy does not substantially improve ECMWF ensemble-based probabilistic predictions. By contrast, both the HWRF and HMON systems provide generally reliable predictions of intensity changes for cases where RI does not take place. Combining the members from the HWRF and HMON ensemble systems into a large multi-model ensemble does not improve upon HMON probablistic forecasts.

Contrasting Behaviors between the Rapidly Intensifying and Slowly Intensifying Tropical Cyclones in the North Atlantic and Eastern Pacific Basins

AbstractBased on 35-yr (1982–2016) best track and Statistical Hurricane Intensity Prediction Scheme data, this study examined climatology of rapidly intensifying (RI) and slowly intensifying (SI) events as well as their time evolutions of storm-related and environmental parameters for tropical cyclones (TCs) in both North Atlantic (AL) and eastern North Pacific (EP) basins. Major hurricanes were intensified mainly through RI while tropical depression and tropical storms were intensified through SI. The percentage of TCs that underwent RI peaks in the late hurricane season whereas the percentage of TCs that underwent SI peaks early. For the first time in the literature, this study found that RI events have significantly different storm-related and environmental characteristics than SI events for before-, during-, and after-event stages. In both AL and EP basins, RI events always intensify significantly faster during the previous 12 h, are located farther south, and have warmer sea surface and 200-hPa temperatures, greater ocean heat content, larger 200-hPa divergence, weaker vertical wind shear, and weaker 200-hPa westerly flow than SI events for all event-relative stages. In the AL basin, RI events have larger low-level and midlevel relative humidity and larger 850-hPa relative vorticity than SI events for all event-relative stages in the AL and most event-relative stages in the EP. RI events are associated with more convectively unstable atmosphere and are farther away from their maximum potential intensities than SI events for most event-relative stages in the AL and for all event-relative stages in the EP.

A Satellite‐Derived Upper‐Ocean Stratification Data Set for the Tropical North Atlantic With Potential Applications for Hurricane Intensity Prediction

AbstractUpper‐ocean stratification strongly impacts vertical mixing and the heat flux between the ocean and atmosphere, especially under extreme conditions of tropical cyclones (TCs). Knowledge of prestorm stratification is important for accurate TC intensity prediction. In situ observations of the tropical ocean have significantly increased in the past decade. However, they are still too sparse to resolve ocean stratification variability in near‐real time and on small spatial scales. In this study, based on long‐term observations and an ocean reanalysis data set from 2004–2017, we investigate the possibility of retrieving upper‐ocean stratification from sea surface temperature (SST), sea surface salinity (SSS), and sea surface height (SSH) using a simple regression method. It is found that more than 90% of the mean seasonal cycle and about 30% to 80% of temperature and salinity stratification anomalies can be reconstructed using surface data from either observations or an ocean reanalysis. Simple regression can be used with satellite observations to create a high‐resolution, near‐real‐time‐gridded ocean stratification data set that successfully reproduces both the large and mesoscale variability of ocean stratification. When used in a simple expression for TC‐induced SST cooling, the satellite‐derived stratification shows improvements over an ocean analysis in terms of variance explained of SST cooling, offering promise as a near‐real‐time indicator of the ocean's impact on TC intensification.

Understanding Error Distributions of Hurricane Intensity Forecasts during Rapid Intensity Changes

AbstractThe characteristics of official National Hurricane Center (NHC) intensity forecast errors are examined for the North Atlantic and east Pacific basins from 1989 to 2018. It is shown how rapid intensification (RI) and rapid weakening (RW) influence yearly NHC forecast errors for forecasts between 12 and 48 h in length. In addition to being the tail of the intensity change distribution, RI and RW are at the tails of the forecast error distribution. Yearly mean absolute forecast errors are positively correlated with the yearly number of RI/RW occurrences and explain roughly 20% of the variance in the Atlantic and 30% in the east Pacific. The higher occurrence of RI events in the east Pacific contributes to larger intensity forecast errors overall but also a better probability of detection and success ratio. Statistically significant improvements to 24-h RI forecast biases have been made in the east Pacific and to 24-h RW biases in the Atlantic. Over-ocean 24-h RW events cause larger mean errors in the east Pacific that have not improved with time. Environmental predictors from the Statistical Hurricane Intensity Prediction Scheme (SHIPS) are used to diagnose what conditions lead to the largest RI and RW forecast errors on average. The forecast error distributions widen for both RI and RW when tropical systems experience low vertical wind shear, warm sea surface temperature, and moderate low-level relative humidity. Consistent with existing literature, the forecast error distributions suggest that improvements to our observational capabilities, understanding, and prediction of inner-core processes is paramount to both RI and RW prediction.

Development and Evaluation of an Evolutionary Programming-Based Tropical Cyclone Intensity Model

AbstractA statistical–dynamical tropical cyclone (TC) intensity model is developed from a large ensemble of algorithms through evolutionary programming (EP). EP mimics the evolutionary principles of genetic information, reproduction, and mutation to develop a population of algorithms with skillful predictor combinations. From this evolutionary process the 100 most skillful algorithms as determined by root-mean square error on validation data are kept and bias corrected. Bayesian model combination is used to assign weights to a subset of 10 skillful yet diverse algorithms from this list. The resulting algorithm combination produces a forecast superior in skill to that from any individual algorithm. Using these methods, two models are developed to give deterministic and probabilistic forecasts for TC intensity every 12 h out to 120 h: one each for the North Atlantic and eastern and central North Pacific basins. Deterministic performance, as defined by MAE, exceeds that of a “no skill” forecast in the North Atlantic to 96 h and is competitive with the operational Statistical Hurricane Intensity Prediction Scheme and Logistic Growth Equation Model at these times. In the eastern and central North Pacific, deterministic skill is comparable to the blended 5-day climatology and persistence (CLP5) track and decay-SHIFOR (DSHF) intensity forecast (OCD5) only to 24 h, after which time it is generally less skillful than OCD5 and all operational guidance. Probabilistic rapid intensification forecasts at the 25–30 kt (24 h)−1 thresholds, particularly in the Atlantic, are skillful relative to climatology and competitive with operational guidance when subjectively calibrated; however, probabilistic rapid weakening forecasts are not skillful relative to climatology at any threshold in either basin. Case studies are analyzed to give more insight into model behavior and performance.

Deep-PHURIE: deep learning based hurricane intensity estimation from infrared satellite imagery

Hurricanes are among the most destructive natural phenomena on Earth. Timely prediction and tracking of hurricane intensity is important as it can help authorities in emergency planning. Several manual, semi and fully automated techniques based on different principles have been developed for hurricane intensity estimation. In this paper, a deep convolutional neural network architecture is proposed for fully automated hurricane intensity estimation from satellite infrared (IR) images. The proposed architecture is robust to errors in annotation of the storm center with a smaller root-mean-squared error (RMSE) (8.82 knots) in comparison to the previous state of the art methods. A web server implementation of Deep-PHURIE and its pre-trained neural network model are available at the URL: http://faculty.pieas.edu.pk/fayyaz/software.html#Deep-PHURIE.

A Feed Forward Neural Network Based on Model Output Statistics for Short-Term Hurricane Intensity Prediction

Abstract A feed forward neural network (FFNN) is developed for tropical cyclone (TC) intensity prediction, where intensity is defined as the maximum 1-min average 10-m wind speed. This deep learning model incorporates a real-time operational estimate of the current intensity and predictors derived from Hurricane Weather Research and Forecasting (HWRF; 2017 version) Model forecasts. The FFNN model is developed with the operational constraint of being restricted to 6-h-old HWRF data. Best track intensity data are used for observational verification. The forecast training data are from 2014 to 2016 HWRF reforecast data and cover a wide variety of TCs from both the Atlantic and eastern Pacific Ocean basins. Cross validation shows that the FFNN increasingly outperforms the operational observation-adjusted HWRF (HWFI) in terms of mean absolute error (MAE) at forecast lead times from 3 to 57 h. Out-of-sample testing on real-time data from 2017 shows the HWFI produces lower MAE than the FFNN at lead times of 24 h or less and similar MAEs at later lead times. On the other hand, the 2017 data indicate significant potential for the FFNN in the prediction of rapid intensification (RI), with RI defined here as an intensification of at least 30 kt (1 kt ≈ 0.51 m s−1) in a 24-h period. The FFNN produces 4 times the number of hits in HWFI for RI. While the FFNN has more false alarms than the HWFI, Brier skill scores show that, in the Atlantic, the FFNN has significantly greater skill than the HWFI and probabilistic Statistical Hurricane Intensity Prediction System RI index.