PS 12-108 - Relating forest resistance to interacting drought and defoliation to pre-disturbance growth and memory of prior defoliation

Monday, August 15, 2022

5:00 PM – 6:30 PM EDT

Located in: ESA Exhibit Hall

Presenting Author(s)

Jillian Dyer

Graduate Student
University of Connecticut, United States

Coauthor(s)

RF

Robert Fahey

Assistant Professor
University of Connecticut, United States
AB

Audrey Barker Plotkin

Harvard Forest, Harvard University
Petersham, Massachusetts, United States
DT

Danielle Tanzer

MS Student
University of Connecticut
Storrs, CT, United States
KR

Keenan Rivers

SUNY Environmental Science and Forestry, United States

Background/Question/Methods

Forest defoliation events result from a variety of causal agents, which can produce differential severity, patterns, and timing of defoliation outcomes. Severe forest defoliation can reduce tree radial growth and promote tree mortality. With climate change, interactions between defoliation and other disturbances may become more frequent, potentially leading to greater overall disturbance severity. To understand forest resistance to defoliation and drought events, research must focus on how pre-disturbance forest health and legacies of prior defoliation affect disturbance outcomes. The objectives of this study are to: (1) Evaluate factors that might affect tree resistance to defoliation (2) Determine how tree or population prior growth rate affects resistance to defoliation and drought (3) Determine whether ecological memory of previous defoliation is associated with growth and survival of individual trees and populations affected by a subsequent defoliation. We studied forest stands affected by Lymantria dispar dispar defoliation in 2015-2018 in Southern New England. Cores were collected from trees over 20cm in diameter across 35 sites in 2019, and used to develop a growth chronology to assess how historic growth affects tree and population resistance and resilience to defoliation and drought. Plot-level defoliation severity was measured using a defoliation product created from Landsat imagery.

Results/Conclusions

Preliminary random forest analysis identified the most important predictor variables for tree resistance to defoliation in order of relative importance as tree genus, plot-level basal area (a proxy for competition), and plot-level defoliation severity in 2016 (training Pbias = -0.017, validation Pbias = 0.186, test Pbias = -0.206). We observed that as defoliation severity increased, more oak (Quercus sp.) trees had defoliation resistance index values below 1, suggesting they were less resistant to defoliation. Genus is the most important predictor so further analyses focused solely on oak trees. For our second objective, contrary to our hypothesis, our preliminary analysis suggests that there is a significant negative relationship between prior growth and resistance to defoliation among individual trees (R2 = 0.290, F(1,25) = 10.2, p< 0.01). Trends regarding ecological memory of a historic defoliation event are currently being explored. The outcomes of defoliation and drought events are affected by multiple factors including tree genus, competition, disturbance severity, and prior growth. Awareness of these factors and their effects can help managers anticipate forest response to interacting disturbances.