Wind Speed Estimation and Conversion

The purpose of the Jupyter notebook presented in this project is to assign wind speeds to property locations in Florida that were damaged during hurricanes. The property locations can come from insurance data or from reconnaissance data. The notebook can process wind field input data on any grid format and for different exposure types (like marine conditions, open terrain, or urban terrain), different time averages (including 3-second gusts, 1-minute sustained winds, and 10-minute average), and different heights. The notebook performs three key functions: (1) linear interpolation from source grid points to any target location in Florida; (2) conversion of wind speeds between exposure types and time averages; and (3) conversion of wind speeds from one height to another, all using established wind engineering formulas.
Since 2017, Applied Research Associates (ARA), under contract with the National Institute of Standards and Technology (NIST), has provided wind field data on a grid, with both 10 m 3-second gusts and 1-minute sustained winds, both for open terrain, for hurricanes impacting the USA. These datasets are published in the DesignSafe reconnaissance portal. They are a possible source of input for the Jupyter notebook. Other sources of wind field data can be used if available.
The other input data needed for the Jupyter notebook are the files with the lat/long of the properties, either from insurance data or from reconnaissance data (available in the DS reconnaissance portal). In the case of insurance data where issues of confidentiality might be at play, it is left at the discretion of the user to deal with these issues.
The final piece of input data that the Jupyter notebook needs is the surface roughness coefficient to characterize the terrain as urban, suburban, coastal, etc. They are currently contained in a GeoTIFF data file provided by Dr. Steve Cocke from Florida State University. This current GeoTIFF file covers only the state of Florida, but provided that this data is available for other states, the notebook could be extended to other states. It needs to be updated as needed since terrain exposure varies over time and development.
The Jupyter notebook reads the wind field data and first interpolates wind speeds to exact property lat/lon coordinates using linear interpolation, then applies terrain adjustment formulas to get actual terrain wind speeds at each property location. If needed, a final adjustment is performed to bring the winds to the desired height. For example, given any hurricane wind field data from ARA (e.g., hurricane Michael), the notebook interpolates the ARA 10m 3-second gusts or 1-minute sustained winds to any given location, then adjusts these values based on terrain roughness (e.g., urban, coastal, or forested), to get actual terrain 10m wind speeds at that location.