Performing Department
Forestry and Wildland Resources
Non Technical Summary
Across much of the western U.S., widespread tree mortality related to drought has reached epidemic proportions, with well over 100 million trees dying in California alone since 2010 (Fettig et al. 2019). This loss of a natural renewable resource has serious national consequences including reduced timber production, reduced carbon sequestration to mitigate atmospheric greenhouse gases contributing to global warming, and increased risks of fires devastating to urban and wildland areas alike. In a climate predicted to continue warming and drying for many decades to come, the successful establishment and regeneration of these forestlands is dubious. Further, even if forests are able to successfully regenerate in the wake of widespread mortality and catastrophic stand-replacing wildland fires, the species composition of these forests may look very different from the ancestral forests being replaced. Our study will evaluate and quantify drought-related tree mortality and subsequent regeneration across a rural area deeply tied to the forestry industry culturally and economically. This work is needed and timely, as it will answer many questions related to the future landscape and economy of this region and will also identify areas with dangerously high fuel loads. Our ambitious, multidisciplinary, and creative plans to widely share findings with a broad target audience including academics, forest ecologists, land managers, Native American tribal members, land conservation groups, non-profit organizations, college students, timber companies, and local community members will improve public awareness about the need for and importance of active forest management in this densely timbered region. Improved public support of active management is especially important in northern California, as there is a strong legacy of public resistance to actions such as mechanical thinning and prescribed burning due to historically aggressive industrial timber harvesting practices that left behind marred and deteriorated landscapes. Improved understanding of forest demographics in the face of changing climate and dangerously high fuel loads will help land stewards maximize forest timber production, carbon sequestration, resistance, and resilience. Healthy, productive, and sustainable forests in this era of natural and sociopolitical change are critical to rural areas like northern California, as these lands are historically, culturally, and economically tied to the trees.
Animal Health Component
0%
Research Effort Categories
Basic
85%
Applied
15%
Developmental
0%
Goals / Objectives
The goals of this project are to 1) examine how competition, species, and habitat interact to influence forest mortality and regeneration, and 2) improve management efforts to mitigate future forest die-back and wildfire risk and maximize natural forest regeneration. Since 2010, over 52 million hectares of trees have died in California's forests due to climate change, drought, and drought-induced disturbances. Forest loss and wildfire trends have important implications for the health, safety, and well-being of communities in both urban and rural areas throughout the American West. Our project will study the interactive influences of competition, habitat, and geography on tree mortality due to drought and regeneration across a richly speciated bioregion. This work on drought-induced forest mortality and regeneration trends will complement and broaden our ongoing three-year study funded by NSF that is investigating the influences of competition, habitat, and geography on radial growth and drought resistance in a diverse group of conifer species. This added research on mortality and regeneration funded by McIntire-Stennis will benefit society by providing valuable input data for models incorporating aspects of habitat type, forest density, biogeographic region, forest mortality, forest regeneration, forest productivity, and terrestrial carbon sequestration. The project will also mentor young scientists at an undergraduate institution (HSU) and provide knowledge to land managers and regional tribes through public and educational outreach.Using the recent drought (2012 - 2015) and post-drought (2016 - 2020) periods as a natural experiment, this study will investigate tree mortality and regeneration across mesic to xeric habitats and low to high competitive gradients. Our specific objectives are to 1) quantify recent tree mortality and natural tree regeneration across a wide range of competitive environments, habitat types, and species, and to 2) inform land managers about drivers of and trends in forest mortality and regeneration through publications, field site visits, and agency presentations. Our findings will advance the literature on recent forest responses to drought and climate variability, while the project itself will provide interdisciplinary student research opportunities. Focused in northern California, this study will yield essential information about forest health in a diverse and culturally important region of the Pacific West. Given the epidemic drought-induced forest mortality and resulting catastrophic wildland fires currently sweeping the state, these findings will be of particular interest to a broad audience. And, as forest mortality and drought are not issues unique to this region, findings from this work will benefit management efforts in many forest ecosystems, with implications for fire risk, human health, and the well-being of many communities.
Project Methods
Study Area - Our study area lies at a transition zone between the southern edge of the temperate coniferous rainforests of the Pacific Northwest and the southwestern Mediterranean region of California. The study area will also include the sliver of land between the Pacific Ocean and the Klamath Mountains that is locally called the North Coast Range to the southern Cascade Range east of the Klamath Mountains (Fig. 1). The climate is characterized by hot and dry summers with cool moist winters, typical of Mediterranean climate types. Three distinct climatic gradients exist in the region: a north-to-south trend of decreasing winter precipitation and warmer summer temperatures; a west-to east trend away from cooler, moister summers to a warmer, drier environment; and a montane gradient in which temperature decreases and precipitation increases with elevation. Species distributions vary in relation to these distinct climate gradients.This Klamath Mountain Ecoregion is a crossroads of mountains where five biotic regions (the Sierra Nevadas, the Cascades, the Coast Range, the California Valley, and the Great Basin) intersect. As a result of these variable botanical influences, the area is world renowned for biodiversity (Sawyer 2006). Per unit area, the Klamath Mountains (and the southern Appalachian Mountains) hold more plant taxa than any other regions in North America. Thus, this region with its mosaic of habitats and botanical diversity is an ideal study area to evaluate post-drought mortality and regeneration in western forests. Additionally, unlike other regions of the state where entire landscapes suffered nearly complete forest mortality, the majority of forests in this region are still living and can therefore serve as a valuable location to evaluate and monitor forest responses to drought now and into the future.Field Sampling - To meet our objectives to quantify tree mortality and regeneration in sample locations throughout the Klamath Mountain Ecoregion, we will use 43 sampling locations (Fig. 1B) already selected for a complementary study (NSF project; PIs Sherriff and Kerhoulas). These sites range from xeric to mesic based on 30-year mean precipitation inputs (PRISM), experienced the recent severe drought based on 2014 and 2015 Palmer Drought Severity Index (PDSI) values, and have experienced no recent (< 30 year) fire activity. Sites were chosen to maintain relatively consistent aspect, slope (< 30 degrees), soil-type (non-serpentine when possible), and be within two miles of roads. We recognize 43 sampling locations do not equate to full representation of the forest diversity throughout the Klamath Mountain Ecoregion, but instead provide an important representation of locations with diverse conifer species across a broad gradient of site-conditions (e.g., xeric to mesic and dense to open). This study will substantially and broadly inform forest management by enriching our existing indicators (annual growth, δ13C, and tree-level competition) of forest response to drought by including valuable information on tree mortality and regeneration at these same sites.At the center (GPS) point of each of the 43 study sites, we will place 0.1-ha plots to evaluate tree mortality and regeneration. To measure tree mortality, within each plot, species, diameter at breast height (dbh), status (live/dead) and decay class (if dead) will be recorded for all trees (height ³ breast height, 1.37 m). All plot trees will also be categorized into dominant, co-dominant, intermediate, and overtopped canopy classes using a classification system modified from other tree competition studies (i.e., Hunter and Barbour 2001; Schriver et al. 2018; Wright et al. 2018). To quantify local competition, basal area density will be measured using a Cruz-All around each tree within each plot. Finally, the physical environment (e.g., slope, aspect, slope position, soil type, elevation), signs of disturbance (e.g., windthrow, fire, bark beetle, or other damage), and shrub/ground cover within 5 m will also be recorded for each individual tree. To characterize the age of recently dead trees and classify as mortality that occurred pre-, during, or post-drought, we will subsample and core at least two trees from each decay class sampled per species in each plot to date the age of death. Standard dendrochronological techniques using a stereoscope, WinDendro software (Regent Instruments Inc., Québec, Canada), and crossdating procedures (Holmes 1983) will be used to date dead tree cores using existing cores from live trees previously sampled at the same sites for our complementary NSF project.To characterize current tree regeneration, all seedlings (< 30 cm tall) and saplings (> 30 cm tall and < 5 cm dbh) within each 0.1-ha plot will be tallied by species. Seedlings will be further separated into two categories, those < 15 cm tall and those > 15 cm tall. As available, basal disks from the root crown will be collected for five seedlings per species per plot to classify establishment years as pre-, during, or post-drought using standard dendroecological methods (e.g., Sherriff and Veblen 2006, Miller et al. 2017). The physical environment (e.g., slope, aspect, slope position, soil type, elevation), signs of disturbance (e.g., windthrow, fire, bark beetle, or other damage), and shrub/ground cover will also be recorded for sampling locations of seedlings and saplings.Analysis - We will use total counts of live seedlings (< 15 cm tall; > 15 cm tall; and total seedling counts), saplings, and dead trees to evaluate whether location (lat/long), climate (e.g., PDSI, climatic water deficit, actual evapotranspiration, maximum temperature, maximum vapor pressure deficit, annual precipitation), and plot covariates (e.g., elevation, soil type, aspect, slope, local competition, stand structure variables, shrub/ground cover) predict the abundance of each life stage or dead trees. We will use generalized linear models (GLM) with a log link to model the effects of covariates on counts of seedlings, saplings, and dead trees separately, using the MASS package in R (e.g., see Miller et al. 2017). Given that we are not quantitatively accounting for past land use history or prior disturbance history, we limit our analysis to recent tree regeneration and mortality, but use the local competition and stand structure data of live trees as covariates. Because canopy conditions and/or canopy thinning due to disturbance could affect seedling and sapling counts, we will use a Mann-Whitney rank sum test to analyze for the presence of disturbance in the last ca. 100 years (e.g., U.S. Forest Service aerial flight data for insect activity, CalFire data for fires) across all plots. We will evaluate pairwise covariate interactions using likelihood ratio tests. Akaike Information Criterion with a correction for finite sample sizes (AICc; Burnham and Anderson 2002) will be used to rank models according to their information content, and coefficient p-values to infer the importance of covariate-response relationships. If the data allow, we will use similar methods to evaluate and compare tree regeneration and mortality for pre-, during, and/or post-drought periods.Outcome Evaluation - This research will advance scientific understanding of forest responses to drought and recent climate changes in California while broadening student opportunities through active participation with a collaborative research team. This project will produce at least one undergraduate thesis, a master's thesis, a final report, and a peer-reviewed publication by the end of Year 3. In Years 2 and 3, we will specifically outreach the research findings and implications for forest management to diverse audiences through a diverse collection of educational efforts including field visits, professional networks, and local communities (see above for more specific examples).