Improving Climate-Based Seed Selection for Forest Health and Carbon Sequestration
Principal Investigator: Joe Stewart, PhD
Project Partners: Jessica Wright, PhD; Greg O’Neill, PhD; Derek Young, PhD; James Thorne, PhD
Institution: University of California, Davis
Project Type: General Research
Grant Award: #8GG21806
Amount awarded: $299,807
Award Date: March, 2021
Trees that grow in climates that differ from the climate to which they are genetically adapted (Climate Adaptation Mismatch, CAM) experience a disrupted annual growth cycle, leading to reduced growth and survival and increased susceptibility to frost, pest, and pathogen damage. Trees and other organisms with long generation times have limited capacity for rapid evolution to changing environmental conditions. Because the climate is changing faster than tree populations evolve, CAM has already led to decreased growth and survival. The effects of CAM are measured with provenance test experiments, where seeds from different climates are planted at common locations and trees are measured over time. Across California, these data suggest that CAM has already caused declining forest growth and survival. Beyond the concerning implications of reduced carbon sequestration and reduced timber production, this mismatch also translates to increased wildfire risk because slower-growing trees are more vulnerable to ignition for a longer.
California recently joined the growing body of governments attempting to combat CAM by developing reforestation guidelines for sourcing seeds adapted to the climate of the planting location (e.g. moving seeds upslope). These guidelines are provided through California’s Climate-Adapted Seed Tool (CAST). The tool uses data from provenance tests to estimate the effects of CAM on growth and survival and to highlight the best seed sources for particular planting locations. While the current version of CAST represents meaningful progress, more advances are required to improve accuracy and ensure the best outcome for California’s forests. These advances include refining underlying statistical models, improving species-specific recommendations, removing geographic uncertainty associated with large seed zones, prioritizing areas for seed collection, and optimizing seed recommendations for dynamic climate conditions that change as the trees grow.
We estimate that widespread adoption of climate-based seed transfer in California reforestation projects—coupled with the improvements in this proposal—would reduce statewide net greenhouse gas emissions by 0.5% by 2050, and by 1% by 2060. Other benefits include increased timber production, reduced wildfire severity, and faster growth into desired habitat types. Due to non-linear tree-adaptation response curves and inertia in the climate system, benefits are poised to accelerate over time. For trees planted this decade, we estimate the advances detailed in this proposal would further increase live tree volume by age 30 by over 10%, with approximately proportionate increases to carbon sequestration and timber production. The statewide benefits of a 10% increase in forest growth are large. On private forest lands alone, we estimate it would result in a 24 Mmt/decade increase in carbon stored, and a $348,000,000/decade increase in the market value of sawlogs.
We seek to answer the following research questions. Q1: Which seed sources are best adapted to the climate of anticipated planting locations for each species? Q2: How far (multivariate climate transfer distance) should seeds be moved to maximize long-term growth and survival over climate conditions that will change over the lifespan of the forest? Q3: For species with limited or absent provenance test data, how much benefit (e.g. growth, survival, carbon sequestration) would be achieved by establishing additional provenance tests to improve decision support? We seek to test the following hypotheses. H1: The cross-validated prediction accuracy of climate transfer functions can be improved by constructing ensemble models. H2: Climate transfer functions are similar between species, but differences are large enough to impact seed-source decisions. H3: Posteriors from species with large sample size can be used as priors to improve climate transfer functions for species with limited sample size.
Presentation: Optimizing assisted gene flow for forests https://www.youtube.com/watch?v=l6aHAK9Og-Q
For more information on this project please visit:
Stewart (PI) website:
The project website:
Joe Stewart (PI)
CAL FIRE Forest Health Research Program