A variety of software has been developed for classroom instruction, operational use, and for research. A list is provided below. Please email me for more information and availability. Some is open source.
Hurricane parametric wind model, known generically as "Fitz Winds.". This is designed to fit eyewall and gale-force winds with the proper asymmetry. Information is here and here. These are often used for wind input into storm surge models (see animation of Hurricane Gustav's storm surge using this software
1D storm surge modeling --- the PACSURGE system
Parallel Coordinates Analytic Toolkit for Multivariate Data Analysis. This uses parallel coordinates visualization with stepwise regression. Patent number US8346682. Information is here and here
Barotropic model
Akima, Steffen, PCHIP, and Catmull-Rom splines --- for monotonic data-fitting appropriate for meteorology interpolation
Validating ocean and weather models combining multiple metrics into a ranking system
Classroom exercises for synoptic meteorology, dynamics, tropical meteorology, oceanography, and numerical methods: see my teaching website weatherclasses.com
Enhanced HURDAT2 with hourly data, storm speed, and track direction. HURDAT2 provides historical hurricane track and intensity data. Information on HURDAT2 is here and here.
Code to process HURDAT2
Enhanced extended best track with the Holland B wind profile parameter. Extended best track contains Radius of Maximum Winds data as well as outer closed isobar which can be used to infer the environmental pressure. Information on the extended best track is here.
Code to process extended best track.
Code to process HWINDS. HWINDS is gridded hurricane wind surface analyses of individual tropical cyclones. Information on HWINDS is
Forecast validation code based on multi-metric ranking method on GitHub here.
Below are also some related links of open source we've used in our research:
Parallel coordinates visualization software by Dr. Chad Steed. See: //csteed.com/projects/ and sublinks to github. Similar to the patent, but does not include the multiple regression component.
