Forecasting Plume-Dominated Wildfires: Environmental Ingredients Governing Depth of Pyroconvection
Loading...
Authors
Salas, Tyler Michael
Issue Date
2025
Type
Thesis
Language
en_US
Keywords
Alternative Title
Abstract
Plume-dominated wildfires pose a major challenge to fire suppression efforts in preserving life and property, especially in the western United States. The pinnacle of pyroconvection, pyrocumulonimbus (PyroCb), can intensify wildfire behavior by inducing strong inflow and outflow winds, which often leads to rapid fire spread and long range spotting. PyroCb is capable of producing lightning and tornadoes, which makes the wildfire even more dangerous and difficult to control under these extreme circumstances. This problem can be further exacerbated by the straining of resources during an active fire season when allocating assets efficiently is critical in mitigating property loss and casualties. Fire weather forecasters attempt to predict and disseminate the potential of deep, vigorous pyroconvection to key fire management partners to assist in resource allocation and tactical planning to successfully combat intense wildfires. Techniques similar to dry thunderstorm forecasting and the use of the Haines Index are often employed to predict the atmosphere’s propensity to support plume-dominated wildfires. However, these techniques are not without their limitations and may result in under/over forecasting of impactful pyroconvection when employed. To address these shortcomings, the parcel-based “Blow-Up” model and the Pyrocumulonimbus Firepower Threshold (PFT) have been developed to assess the potential for sudden, large vertical plume growth and formation of pyroCb, respectively. Both of these models, along with the Haines Index and Hot-Dry-Windy Index, are evaluated in their ability to predict deep pyroconvection. The goal of this thesis is to improve short and near term forecasting capabilities of plume-dominated wildfires by identifying environmental discriminators and exploring the versatility of the PFT when juxtaposed with plume depth observations and Fire Radiative Power (FRPx10) data from the 2020 Creek and Bear fires. The former analysis is intended to help forecasters predict plume-dominated wildfires in the short term (24-48 hours lead time) by refining an ingredients-based forecast approach while the latter assists with near-term (up to ~6 hours lead time) forecasting precision should the PFT be deemed viable in an operational setting. Twenty wildfires in the western United States that occurred between 2018 and 2022 are investigated in this study and are selected to represent a diversity in cases with respect to location and timing, background environment, and magnitude of plume depth. The study proves the Haines Index to be a poor predictor of deep pyroconvection, distinguishes favorable environmental ingredients, and highlights the usefulness and limitations of the PFT model when coupled with FRPx10 and plume observations.