An official website of the United States government
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.
more »
« less
This content will become publicly available on May 15, 2026
Identifying data sources and physical strategies used by neural networks to predict TC rapid intensification
Abstract The rapid intensification (RI) of tropical cyclones (TC), defined here as an intensity increase of ≥ 30 kt in 24 hours, is a difficult but important forecasting problem. Operational RI forecasts have considerably improved since the late 2000s, largely thanks to better statistical models, including machine learning (ML). Most ML applications use scalars from the Statistical Hurricane Intensity Prediction Scheme (SHIPS) development dataset as predictors, describing the TC history, near-TC environment, and satellite presentation of the TC. More recent ML applications use convolutional neural networks (CNN), which can ingest full satellite images (or time series of images) and freely “decide” which spatiotemporal features are important for RI. However, two questions remain unanswered: (1) Does image convolution significantly improve RI skill? (2) What strategies do CNNs use for RI prediction – and can we gain new insights from these strategies? We use an ablation experiment to answer the first question and explainable artificial intelligence (XAI) to answer the second. Convolution leads to only a small performance gain, likely because, as revealed by XAI, the CNN’s main strategy uses image features already well described in scalar predictors used by pre-existing RI models. This work makes three additional contributions to the literature: (1) NNs with SHIPS data outperform pre-existing models in some aspects; (2) NNs provide well calibrated uncertainty quantification (UQ), while pre-existing models have no UQ; (3) the NN without SHIPS data performs surprisingly well and is fairly independent of pre-existing models, suggesting its potential value in an operational ensemble.
more »
« less
- Award ID(s):
- 2019758
- PAR ID:
- 10596367
- Publisher / Repository:
- AMS
- Date Published:
- Journal Name:
- Weather and Forecasting
- ISSN:
- 0882-8156
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
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.more » « less
-
Abstract The parametric hurricane rainfall model (PHRaM), firstly introduced in 2007, has been widely used to forecast and quantify tropical-cyclone-induced rainfall (TC rainfall). The PHRaM is much more computationally efficient than global climate models, but PHRaM cannot be effectively utilized in the context of climate change because it does not have any parameters to capture the increase of tropospheric water vapor under the warming world. This study develops a new model that incorporates tropospheric water vapor to the PHRaM framework, named as the PHRaM with moisture (PHRaMM). The PHRaMM is trained to best fit the TC rainfall over the western North Pacific (WNP) unlike the PHRaM trained with the TCs over the continental US. The PHRaMM reliably simulates radial profile of TC rainfall and spatial distribution of accumulated rainfall during landfall in the present climate with the better prediction skills than existing statistical and operational numerical models. Using the PHRaMM, we evaluated the impacts of TC intensity and environmental moisture increase on TC rainfall change in a future climate. An increased TC intensity causes TC rainfall to increase in the inner-core region but to decrease in the outer region, whereas an increased environmental moisture causes the TC rainfall to increase over the entire TC area. According to the both effects of increased TC intensity and environmental moisture, the PHRaMM projected that the WNP TC rainfall could increase by 4.61–8.51% in the inner-core region and by 17.96–20.91% over the entire TC area under the 2-K warming scenario.more » « less
-
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 parameterB, 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.more » « less
-
Abstract The phenomenon that rapid contraction (RC) of the radius of maximum wind (RMW) could precede rapid intensification (RI) in tropical cyclones (TCs) has been found in several previous studies, but it is still unclear how frequently and to what extent RC precedes RI in rapidly intensifying and contracting TCs in observations. In this study, the statistical relationship between RMW RC and TC RI is examined based on the extended best track dataset for the North Atlantic and eastern North Pacific during 1999–2019. Results show that for more than ∼65% of available TCs, the time of the peak contraction rate precedes the time of the peak intensification rate, on average, by ∼10–15 h. With the quantitatively defined RC and RI, results show that ∼50% TCs with RC experience RI, and TCs with larger intensity and smaller RMW and embedded in more favorable environmental conditions tend to experience RI more readily following an RC. Among those TCs with RC and RI, more than ∼65% involve the onset of RC preceding the onset of RI, on average, by ∼15–25 h. The preceding time tends to be longer with lower TC intensity and larger RMW and shows weak correlations with environmental conditions. The qualitative results are insensitive to the time interval for the calculation of intensification/contraction rates and the definition of RI. The results from this study can improve our understanding of TC structure and intensity changes.more » « less
