PS 32-189 - Projecting Future Drought-induced Tree Mortality Across Dry Forests in the Southwestern United States

Tuesday, August 8, 2023

5:00 PM – 6:30 PM PDT

Location: ESA Exhibit Hall

Presenting Author(s)

AS

Andrew Shannon

Center for Geospatial Analytics, North Carolina State University, United States

Coauthor(s)

RS

Robert Scheller

North Carolina State University, United States
SF

Samuel Flake

North Carolina State University, United States
LM

Lisa McCauley

The Nature Conservancy in Arizona, United States
PH

Paul Henne

U.S. Geological Survey, United States

Abstract: Drought-induced tree mortality has increased globally in recent decades. High stand density, competition for water resources, and heat-stress exacerbate rates of tree mortality, which disrupts ecosystem functions and contributes to wildfire risk, especially in forests of semi-arid and arid climates. Based on forest inventory data and climate observations, researchers have developed the statistical, species-specific relationships between stand characteristics, climate variables, and mortality for dry forest species in the southwestern United States. To project future drought-induced tree mortality, we integrated the statistical relationships for six dry forest species groups into LANDIS-II, a forest landscape simulation model. LANDIS-II dynamically simulates forest succession and accounts for interactions with climate and disturbances, such as a wildland fire, in a spatially-explicit framework. We simulated a ponderosa pine-dominated landscape across national forest land in Arizona where a history of fire exclusion has allowed stand density to increase in the present day. There, managers are interested in understanding the long-term effects of drought and fire behavior on future tree-mortality in the absence of a fire-adapted restoration scenario through the 21st century. Integrating empirical species-drought relationships into LANDIS-II will demonstrate how future tree mortality changes in response to the dynamic, interrelated patterns of drought and fire behavior.

We ran the model under the conditions of multiple climate projections that varied by drought intensity and frequency. We parameterized a fire behavior model to simulate future fire regimes based on fuel availability, weather conditions, and probability of ignitions and spread. Integrating the statistical drought relationships improved how the model represents tree mortality in response to climate conditions, and based on the fire and climate scenarios, we analyzed how tree mortality varied by intensity and frequency of fire and drought events. Preliminary model outputs indicate that biomass increases in near-term decades without management activities, and the drought intensity varies by selected climate projection. Future drought-induced tree mortality varied by species cohort, and high intensity drought events increased mortality for all species. Prolonged drought periods resulted in greater fuel availability that increased fire intensity and extent. Under conditions of low intensity fire, fire events reduced biomass, which decreased the effects of drought-induced mortality for post-fire drought periods. These results highlight how forests in this landscape change in response to the feedback between climate and fire-driven disturbances under the continuation of a business-as-usual management scenario.