Using the METplus TC Tools to Verify Tropical Cyclones in Real Time

A major component of the Unified Forecast System Research-to-Operations (UFS-R2O) project is the implementation of a unified verification system for community numerical weather prediction model development efforts. A community verification tool supports evidence-based decision making when transitioning these models to operations. This unified verification system uses the Developmental Testbed Center (DTC)'s Model Evaluation Tools (METplus) to process model output into a common format, which is then stored in a relational or document database.

METplus includes tools for verifying tropical cyclones (TCs), generating statistics such as track error, intensity bias, and eye diameter. These tools were initially designed for post-season storm analysis, but they can be used in real-time verification as well. This presentation will examine two such real-time implementations: the US National Oceanographic and Atmospheric Administration (NOAA) Global Systems Laboratory (GSL)'s live hurricane season model verification, and the future inclusion of METplus's TC tools in the operational Hurricane Analysis and Forecast System (HAFS) workflow.

At GSL, the global model physics developers take part each year in the Hurricane Forecast Improvement Project (HFIP), a coordinated effort in the US to improve the accuracy and reliability of TC forecasts. As there are only a few months out of the year when the bulk of North Atlantic and East Pacific TCs occur, GSL's model developers need fast, real-time verification results to show which of their model physics under development are performing the best, and what improvements need to be made and subsequently tested on the next of that years' storms. To this end, GSL runs two parallel workflows with the METplus TC tools. The first verifies all the points in model TC forecasts as they become available (for example, at 00UTC on 1 September, this workflow would verify the 6hr track forecast from 18UTC on 31 August, the 12hr track forecast from 12Z on 31 August, etc). This workflow provides the fastest feedback to GSL model developers, but cannot use all of the TC tools available without full forecast tracks, and doesn't utilize later corrections to observed TC track and intensity data. The second workflow runs at a week's lag, and verifies the last seven days of all valid forecast tracks. This feedback is slower to become accessible to model developers, but takes the more complete model and observational tracks available and utilizes the full suite of TC tools to calculate after-the-fact statistics such as rapid intensification (RI). This presentation will explore both of these workflow approaches, and explain the differences that users might expect to see between them.

NOAA's UFS Hurricane Application, HAFS, began running in operations in June 2023. The HAFS is an atmosphere-ocean-wave coupled TC forecast system, featuring convection-allowing high-resolution storm-following nests, vortex initialization, inner-core data assimilation, and TC-calibrated model physics. As part of the NOAA UFS, the HAFS has a focus on transitioning TC modeling research into operations through a community modeling paradigm. In order to facilitate both rapid performance and extensive large sample evaluations, the METplus verification system is being implemented into the HAFS workflow. This presentation will focus on the initial development efforts to implement the suite of MET TC evaluation tools, including track, intensity, wind radii, and rapid intensification verification into the HAFS workflow.

The METplus TC tools, like the rest of METplus, are open source, and our hope is that, as more features are added, their use continues to expand amongst the UFS-R2O community, and to other public and private sector entities, aiding the development of new, next-generation models.