NH008-0006
Correcting Area Burned in Earth System Models Using a Probabilistic Representation of Sub-Grid Topography

Tuesday, 8 December 2020
Poster
Julia Oliveto1, Alexandra K Jonko1, Daniel Livingston2 and Rodman Linn1, (1)Los Alamos National Laboratory, Los Alamos, NM, United States, (2)Los Alamos National Laboratory, Los Alamos, United States
Abstract:
Earth System Models (ESMs) are increasingly being used to study not only global climate change but also the effects of global change on ecosystem processes, including wildfire. Incorporating wildfire representations in ESMs also allows us to investigate the impact of changing wildfire activity on climate. Current ESMs implement relatively simple representations of wildfire, which over and underestimate global burned area. One reason is fire modules used by ESMs, such as SPITFIRE used in the DOE Energy Exascale Earth System Model (E3SM), do not account for topography or heterogeneous distribution of vegetation across the landscape. Our project leverages LANL's physics-based fire behavior model FIRETEC to develop a representation of subgrid impacts of topography and vegetation heterogeneity on wildfire.

Our ultimate goal is to obtain a correction factor for area burned using FIRETEC simulations and high-resolution elevation and vegetation data. This presentation will focus on developing the workflow for the state of California. The workflow consists of: calculations of slope, aspect, curvature, and vegetation patch size from 30m-resolution elevation data; creation and sampling from a joint distribution of the variables listed; use of the aforementioned samples to populate topography and vegetation parameters for FIRETEC; FIRETEC simulations of area burned which corresponds to a topography/vegetation distribution representative of a part of CA. These burned areas will serve as correction factors for SPITFIRE over CA, with the goal to expand the workflow to the contiguous US.