Citation: Tagestad, J.D.; Saltiel, T.M.; Coleman, A.M. Rapid Spaceborne Mapping of Wildfire Retardant Drops for Active Wildfire Management. Remote Sens. 2023, 15, 342. https:// doi.org/10.3390/rs15020342 Academic Editors: Dimitris Stavrakoudis and Ioannis Z. Gitas Received: 21 October 2022 Revised: 3 December 2022 Accepted: 15 December 2022 Published: 6 January 2023 Copyright: © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). remote sensing Communication Rapid Spaceborne Mapping of Wildfire Retardant Drops for Active Wildfire Management Jerry D. Tagestad * , Troy M. Saltiel and André M. Coleman Pacific Northwest National Laboratory, Earth Systems Predictability & Resiliency Group, Richland, WA 99352, USA * Correspondence: jerry.tagestad@pnnl.gov Abstract: Aerial application of fire retardant is a critical tool for managing wildland fire spread. Retardant applications are carefully planned to maximize fire line effectiveness, improve firefighter safety, protect high-value resources and assets, and limit environmental impact. However, topogra- phy, wind, visibility, and aircraft orientation can lead to differences between planned drop locations and the actual placement of the retardant. Information on the precise placement and areal extent of the dropped retardant can provide wildland fire managers with key information to (1) adaptively manage event resources, (2) assess the effectiveness of retardant slowing or stopping fire spread, (3) document location in relation to ecologically sensitive areas; and perform or validate cost- accounting for drop services. This study uses Sentinel-2 satellite data and commonly used machine learning classifiers to test an automated approach for detecting and mapping retardant applica- tion. We show that a multiclass model (retardant, burned, unburned, and cloud artifact classes) outperforms a single-class retardant model and that image differencing (post-application minus pre-application) outperforms single-image models. Compared to the random forest and support vector machine, the gradient boosting model performed the best with an overall accuracy of 0.88 and an F1 Score of 0.76 for fire retardant, though results were comparable for all three models. Our approach maps the full areal extent of the dropped retardant within minutes of image availability, rather than linear representations currently mapped by aerial GPS surveys. The development of this capability allows for the rapid assessment of retardant effectiveness and documentation of placement in relation to sensitive environments. Keywords: wildfire; wildland fire management; remote sensing; fire retardant 1. Introduction With the increase in large and complex wildland fires, there has been a concomitant increase in firefighting infrastructure, including tactical aerial platforms. Aerial application of fire retardant is a critical tool for managing wildland fires and is carefully planned to maximize the effectiveness of each drop [1,2] while protecting the safety of flight and ground crews and minimizing environmental effects. However, visibility, wind, and turbulence can lead to differences between planned drop locations and the dispersal and final placement of the retardant. Additionally, aircraft flight operations and topography can greatly influence the coverage of the drop footprint [3]. Knowing the precise location of the retardant on the landscape is critical in determining the application’s efficacy and documenting the retardant footprint in relation to riparian areas and firefighter locations [3–6]. Fire retardant placement is commonly mapped as a linear feature via aerial GPS surveys that take place as soon as possible, but often days after the drop. Several studies have used airborne data from infrared cameras to map fire retardants to determine the effect of retardants on fire spread [3,7,8]. Notably, these studies use methods to manually delineate the footprint of the retardant drop (rather than just a centerline), noting that drop dimension is an important factor in the effect of the retardant on fire spread [3]. The ability to carefully track Remote Sens. 2023, 15, 342. https://doi.org/10.3390/rs15020342 https://www.mdpi.com/journal/remotesensing