Tropical and Extratropical Cyclone Detection Using Deep Learning

Kumler-Bonfanti, Christina; Stewart, Jebb Q.; Hall, David; Govett, M.

doi:10.1175/jamc-d-20-0117.1

Cited by 39 publications

(18 citation statements)

References 90 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Another example of semantic segmentation is cyclone detection. In the work of Kumler et al [16], there are 3 possible classes: tropical cyclone, extratropical cyclone, and not a cyclone. Additionally, Kumler et al experiment with various loss functions for the U-net trained to solve this problem.…”

Section: Loss Functions For Semantic Segmentationmentioning

confidence: 99%

“…Vignette #1: Insights from Christina Kumler Application: Detecting tropical and extratropical cyclones We developed four U-net models to identify either tropical cyclones, or both tropical and extratropical cyclones [16]. Inputs to the U-nets consisted of two global images (covering the whole Earth): one from satellite measurements and one from the Global Forecasting System (GFS), which is a process-based atmospheric model.…”

Section: Insights From the Practical Use Of Custom Losses In Environm...mentioning

confidence: 99%

“…The pixel size indicates how large the cube around the cyclone was labeled in the input image. Adapted from Table 2 in [16]. [44] is a reliability curve with additional reference lines and histograms.…”

Section: Insights From the Practical Use Of Custom Losses In Environm...mentioning

confidence: 99%

See 2 more Smart Citations

CIRA Guide to Custom Loss Functions for Neural Networks in Environmental Sciences -- Version 1

Ebert‐Uphoff¹,

Lagerquist²,

Hilburn³

et al. 2021

Preprint

View full text Add to dashboard Cite

Neural networks are increasingly used in environmental science applications. Furthermore, neural network models are trained by minimizing a loss function, and it is crucial to choose the loss function very carefully for environmental science applications, as it determines what exactly is being optimized. Standard loss functions do not cover all the needs of the environmental sciences, which makes it important for scientists to be able to develop their own custom loss functions so that they can implement many of the classic performance measures already developed in environmental science, including measures developed for spatial model verification. However, there are very few resources available that cover the basics of custom loss function development comprehensively, and to the best of our knowledge none that focus on the needs of environmental scientists. This document seeks to fill this gap by providing a guide on how to write custom loss functions targeted toward environmental science applications. Topics include the basics of writing custom loss functions, common pitfalls, functions to use in loss functions, examples such as fractions skill score as loss function, how to incorporate physical constraints, discrete and soft discretization, and concepts such as focal, robust, and adaptive loss. While examples are currently provided in this guide for Python with Keras and the TensorFlow backend, the basic concepts also apply to other environments, such as Python with PyTorch. Similarly, while the sample loss functions provided here are from meteorology, these are just examples of how to create custom loss functions. Other fields in the environmental sciences have very similar needs for custom loss functions, e.g., for evaluating spatial forecasts effectively, and the concepts discussed here can be applied there as well. All code samples are provided in a GitHub repository.

show abstract

Section: Loss Functions For Semantic Segmentationmentioning

confidence: 99%

Section: Insights From the Practical Use Of Custom Losses In Environm...mentioning

confidence: 99%

See 1 more Smart Citation

CIRA Guide to Custom Loss Functions for Neural Networks in Environmental Sciences -- Version 1

Ebert‐Uphoff¹,

Lagerquist²,

Hilburn³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…By performing convolutional operations on an input map, CNNs identify salient features in the input space which influence the desired prediction. In meteorology, CNN models have recently been successfully applied to detect synopticscale structures such as fronts (e.g., Lagerquist et al, 2019), atmospheric rivers (e.g., Muszynski et al, 2019;Prabhat et al, 2021), extratropical cyclones (Lu et al, 2020;Kumler-Bonfanti et al, 2020), dry intrusions (Silverman et al, 2021), and tropical cyclones (e.g., Matsuoka et al, 2018;Prabhat et al, 2021). In this research, the architecture of the CNN models is based on the UNet, which was originally designed as a semantic-segmentation model for medical images (Ronneberger et al, 2015).…”

Section: Introductionmentioning

confidence: 99%

EuLerian Identification of ascending AirStreams (ELIAS 2.0) in numerical weather prediction and climate models – Part 1: Development of deep learning model

Quinting

Grams

2022

Geosci. Model Dev.

View full text Add to dashboard Cite

Abstract. Physical processes on the synoptic scale are important modulators of the large-scale extratropical circulation. In particular, rapidly ascending airstreams in extratropical cyclones, so-called warm conveyor belts (WCBs), modulate the upper-tropospheric Rossby wave pattern and are sources and magnifiers of forecast uncertainty. Thus, from a process-oriented perspective, numerical weather prediction (NWP) and climate models should adequately represent WCBs. The identification of WCBs usually involves Lagrangian air parcel trajectories that ascend from the lower to the upper troposphere within 2 d. This requires expensive computations and numerical data with high spatial and temporal resolution, which are often not available from standard output. This study introduces a novel framework that aims to predict the footprints of the WCB inflow, ascent, and outflow stages over the Northern Hemisphere from instantaneous gridded fields using convolutional neural networks (CNNs). With its comparably low computational costs and relying on standard model output alone, the new diagnostic enables the systematic investigation of WCBs in large data sets such as ensemble reforecast or climate model projections, which are mostly not suited for trajectory calculations. Building on the insights from a logistic regression approach of a previous study, the CNNs are trained using a combination of meteorological parameters as predictors and trajectory-based WCB footprints as predictands. Validation of the networks against the trajectory-based data set confirms that the CNN models reliably replicate the climatological frequency of WCBs as well as their footprints at instantaneous time steps. The CNN models significantly outperform previously developed logistic regression models. Including time-lagged information on the occurrence of WCB ascent as a predictor for the inflow and outflow stages further improves the models' skill considerably. A companion study demonstrates versatile applications of the CNNs in different data sets including the verification of WCBs in ensemble forecasts. Overall, the diagnostic demonstrates how deep learning methods may be used to investigate the representation of weather systems and their related processes in NWP and climate models in order to shed light on forecast uncertainty and systematic biases from a process-oriented perspective.

show abstract

“…Additionally, deep learning models have been introduced in TC detection as well, for example, the use of deep neural networks (DNN) for existing TC detection [30], precursor detection of TCs [31], tropical and extratropical cyclone detection [32], TC track forecasting [33], and TC precursor detection by a cloud-resolving global nonhydrostatic atmospheric model [34]. However, deep learning models usually require a large number of training samples, because it is difficult to achieve high accuracy in case of finite training samples in computer vision and other fields [35][36][37].…”

Section: Introductionmentioning

confidence: 99%

NDFTC: A New Detection Framework of Tropical Cyclones from Meteorological Satellite Images with Deep Transfer Learning

Pang

Xie

et al. 2021

Remote Sensing

View full text Add to dashboard Cite

Accurate detection of tropical cyclones (TCs) is important to prevent and mitigate natural disasters associated with TCs. Deep transfer learning methods have advantages in detection tasks, because they can further improve the stability and accuracy of the detection model. Therefore, on the basis of deep transfer learning, we propose a new detection framework of tropical cyclones (NDFTC) from meteorological satellite images by combining the deep convolutional generative adversarial networks (DCGAN) and You Only Look Once (YOLO) v3 model. The algorithm process of NDFTC consists of three major steps: data augmentation, a pre-training phase, and transfer learning. First, to improve the utilization of finite data, DCGAN is used as the data augmentation method to generate images simulated to TCs. Second, to extract the salient characteristics of TCs, the generated images obtained from DCGAN are inputted into the detection model YOLOv3 in the pre-training phase. Furthermore, based on the network-based deep transfer learning method, we train the detection model with real images of TCs and its initial weights are transferred from the YOLOv3 trained with generated images. Training with real images helps to extract universal characteristics of TCs and using transferred weights as initial weights can improve the stability and accuracy of the model. The experimental results show that the NDFTC has a better performance, with an accuracy (ACC) of 97.78% and average precision (AP) of 81.39%, in comparison to the YOLOv3, with an ACC of 93.96% and AP of 80.64%.

show abstract

Tropical and Extratropical Cyclone Detection Using Deep Learning

Cited by 39 publications

References 90 publications

CIRA Guide to Custom Loss Functions for Neural Networks in Environmental Sciences -- Version 1

CIRA Guide to Custom Loss Functions for Neural Networks in Environmental Sciences -- Version 1

EuLerian Identification of ascending AirStreams (ELIAS 2.0) in numerical weather prediction and climate models – Part 1: Development of deep learning model

NDFTC: A New Detection Framework of Tropical Cyclones from Meteorological Satellite Images with Deep Transfer Learning

Contact Info

Product

Resources

About