Uncrewed Aircraft System Data Assimilation for Improved Wildland Fire Fighting Decision Support

Aerial assets are a critical tool used to fight wildland fires in areas of complex terrain. In late 2021, the first known attempt to use an air tanker at night resulted in a fatal crash that was potentially caused by strong low level turbulence. This tragedy and others like it demonstrate the need for accurate estimates of the current low-level wind field and its evolution over 3-12 hours in order ensure safe and efficient operations and to expand the operational envelope to include nighttime operations. nbsp;We propose a pilot study to demonstrate the value of collecting and assimilating targeted UAS observations to improve the representation of low-level winds and turbulence within regions impacted by active wildland fires. Black Swift Technologies (BST) will lead a partnership with NCAR/RAL and Coloradorsquo;s Center of Excellence (CoE) for Advanced Technology Aerial Firefighting to tackle this challenging problem. BST will work with the CoE to obtain permission to collect UAS observations near wildland fires and the resulting data will be employed by NCAR to perform data quality assessments and data assimilation (DA) experiments using an Ensemble Kalman Filter approach which has been tailored to assimilate UAS observations. Not only will this allow for optimizing the impact of UAS borne measurements, it will also allow for determination of critical wind and turbulence criteria that are needed to support the safe operation of airborne wildland firefighting assets. Data quality assessments will be used to quantify a key outcome of this proposed effort which will be to demonstrate the degree to which UAS observations can be used to improve the accuracy of low-level winds and turbulence in the vicinity of a wildland fire and to determine how this enhanced guidance improves the safety of mixed (crewed and uncrewed) flights performed during wildland firefighting operations.