Probabilistic Prediction of Tropical Cyclone Intensity with an Analog Ensemble

Abstract

An analog ensemble (AnEn) technique is applied to the prediction of tropical cyclone (TC) intensity (i.e., maximum 1-min averaged 10-m wind speed). The AnEn is an inexpensive, naturally calibrated ensemble prediction of TC intensity derived from a training dataset of deterministic Hurricane Weather Research and Forecasting (HWRF; 2015 version) Model forecasts. In this implementation of the AnEn, a set of analog forecasts is generated by searching an HWRF archive for forecasts sharing key features with the current HWRF forecast. The forecast training period spans 2011–15. The similarity of a current forecast with past forecasts is estimated using predictors derived from the HWRF reforecasts that capture thermodynamic and kinematic properties of a TC’s environment and its inner core. Additionally, the value of adding a multimodel intensity consensus forecast as an AnEn predictor is examined. Once analogs are identified, the verifying intensity observations corresponding to each analog HWRF forecast are used to produce the AnEn intensity prediction. In this work, the AnEn is developed for both the eastern Pacific and Atlantic Ocean basins. The AnEn’s performance with respect to mean absolute error (MAE) is compared with the raw HWRF output, the official National Hurricane Center (NHC) forecast, and other top-performing NHC models. Also, probabilistic intensity forecasts are compared with a quantile mapping model based on the HWRF’s intensity forecast. In terms of MAE, the AnEn outperforms HWRF in the eastern Pacific at all lead times examined and up to 24-h lead time in the Atlantic. Also, unlike traditional dynamical ensembles, the AnEn produces an excellent spread–skill relationship.

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

“…3), with aggregate forecast RI probabilities that are lower than the verifying observations. Therefore, RI probabilities are subjectively calibrated using probability matching, which is akin to the quantile mapping approach described by Alessandrini et al (2018). This is best illustrated by an example.…”

“…To date, consensus models have outperformed the other intensity model types in the Atlantic and eastern and central North Pacific basins, but they are followed closely by statistical-dynamical models and recently by the best-performing dynamical models (Stewart 2014(Stewart , 2016Pasch 2015). Probabilistic TC intensity forecasts are almost exclusively derived from statistical-dynamical models (e.g., Kaplan and DeMaria 2003;Kaplan et al 2010Kaplan et al , 2015Rozoff and Kossin 2011;Cloud et al 2019), although recent attempts to use dynamical-model ensembles to predict RI shown promise (e.g., Alessandrini et al 2018).…”

Forecasting Severe Weather with Random Forests

“…Many of these studies have applied machine learning (ML) to the prediction task; in general, ML techniques have demonstrated great promise in applications to high-impact weather prediction (e.g., McGovern et al 2017McGovern et al , 2019. In addition to severe weather, ML has demonstrated success in forecasting heavy precipitation (e.g., Gagne et al 2014;Herman and Schumacher 2018a,b;Whan and Schmeits 2018;Loken et al 2019), cloud ceiling and visibility (e.g., Herman and Schumacher 2016;Verlinden and Bright 2017), and tropical cyclones (Loridan et al 2017;Alessandrini et al 2018;Wimmers et al 2019). Furthermore, automated probabilistic guidance, including ML algorithms, have been identified as a priority area for integrating with the operational forecast pipeline (e.g., Rothfusz et al 2014;Karstens et al 2018).…”

Improving Wind Forecasts in the Lower Stratosphere by Distilling an Analog Ensemble Into a Deep Neural Network

“…Forecast improvements are demonstrated with only 2 years of previous forecasts. Versions of the AnEn have been applied successfully for the prediction of weather parameters (Delle Monache et al., 2013; Eckel & Delle Monache, 2016; Frediani et al., 2017; Keller et al., 2017; Nagarajan et al., 2015; Plenkovi et al., 2018; Sperati et al., 2017; Yang et al., 2018), tropical cyclone intensity (Alessandrini et al., 2018), air quality (Delle Monache et al., 2020; Djalalova et al., 2015; Huang et al., 2017), and renewable energy (Alessandrini, Delle Monache, Sperati, & Nissen, 2015; Alessandrini, Delle Monache, Sperati, & Cervone, 2015; Cervone et al., 2017; Davò et al., 2016; Ferruzzi et al., 2016; Junk et al., 2015; Mahoney et al., 2012; Shahriari et al., 2020; Vanvyve et al., 2015), but this is the first application of the approach to stratospheric winds.…”