Purdue University Graduate School
Dow_thesis.pdf (4.1 MB)


Download (4.1 MB)
posted on 2023-05-01, 16:16 authored by Cameron David DowCameron David Dow

In 2016, chestnut oak (Quercus prinus) trees across southern Indiana began displaying symptoms of decline disease. In the years following, widespread patches of mortality appeared on slopes and along ridges, prompting the Indiana Department of Natural Resources (IDNR) to investigate. The IDNR noted the presence of Phytophthora cinnamomi on one diseased chestnut oak, leading to the initiation of this study. Our goals were to (1) determine if P. cinnamomi contributed to the widespread decline, (2) determine what site or stand variables were associated with higher rates decline, (3) examine the growth of declining trees prior to the onset of symptoms to determine if drought contributed to decline, and (4) use Forest Service Forest Inventory and Analysis (FIA) data to examine trends in regional chestnut oak mortality.

From 2021-2022, we collected thirty fine root and soil samples from declining chestnut oak trees within Morgan-Monroe State Forest, Yellowwood State Forest, and the Hoosier National Forest. Throughout these forests, we established sampling plots within declining and healthy chestnut oak stands. In each plot we recorded site, stand, and tree level variables, then collected tree cores from two or three chestnut oak trees. Fine root samples were tested for the presence of P. cinnamomi at the Purdue Pathogen and Pest Diagnostic Laboratory (PPDL) in West Lafayette, Indiana.

We used binomial linear regression to test for significant (α = 0.05) relationships between site and stand variables and decline, where a binary decline or no decline was used as the response variable. Both basal area increment (BAI) and ring width index (RWI) chronologies were built separately for healthy, declining, and dead chestnut oak trees. We used a paired t-test (α = 0.05) to test for significant differences in 10-year segments of BAI between the three chronologies. Finally, we used linear regression to test for significant (α = 0.05) effects of the standardized precipitation-evapotranspiration index (SPEI) in current and previous year growing seasons on RWI. Chestnut oak mortality volume across the eastern United States was calculated using the EVALIdator tool provided by USDA-FS FIA. Associations between chestnut oak mortality recorded by FIA and several climate and topographical variables were examined using a random forest classification.

Out of thirty fine root samples, only one tested positive for the presence of P. cinnamomi, indicating that this decline was not associated with the pathogen. The analysis of site and stand variables revealed a greater chance of decline on east and northeastern facing slopes, with a slight increase in decline likelihood with increasing stand density. There was significantly lower BAI in dead and declining trees long before decline symptoms began, a pattern consistent with previous drought induced declines. We observed a significant relationship between RWI and SPEI in the early growing season (June and 3-month June SPEI) and throughout almost all of the prior year’s growing season (May, June, 3-month June, 3-month July, and 3-month August SPEI). Chestnut oak mortality volume across the eastern US steadily increased from 2006-2020, indicating a region-wide increase in mortality. Our random forest classification indicated the importance of increased precipitation and precipitation timing on chestnut oak mortality.

Chestnut oak decline observed in southern Indiana was induced by a series of droughts in 2005, 2007, and 2012. The greater early life BAI of chestnut oak which were impacted by decline revealed that individuals which likely prioritized stem growth over root growth were predisposed to decline and mortality from these droughts. This prioritization could be brought on by genetic differences, favoring rapid height growth in developing even-aged stands, or by an abundance of moisture availability. Our FIA analysis of mortality revealed increased mortality volume across many states from 2006-2020, and that chestnut oak mortality may be related to greater precipitation compared to historic levels. Considering these results, we suspect that chestnut oak which have recently died or are currently declining are likely individuals which lack the root system to endure repeated drought.


Degree Type

  • Master of Science


  • Forestry and Natural Resources

Campus location

  • West Lafayette

Advisor/Supervisor/Committee Chair

Michael Jenkins

Additional Committee Member 2

Michael Saunders

Additional Committee Member 3

James Jacobs

Usage metrics



    Ref. manager