Semester of Graduation

Fall 2022

Degree

Master of Forest Resources (MFR)

Department

Renewable Natural Resources

Document Type

Thesis

Abstract

Climate change is expected to radically alter our planet’s forests, with higher frequencies of drought- and flood-induced mortality events posing a challenge for forest managers and biologists. Research into the factors underlying plant tolerance to environmental stressors is therefore gaining popularity for incorporation into projective and earth system modeling using remote sensing measures. Leaf turgor loss point (TLP) is a key trait associated with drought tolerance among plants and is defined as the water potential at which leaf turgor pressure reaches zero, causing wilting. Here, I investigated patterns of TLP across the landscape and its role as an indicator of drought in bottomland hardwood (BLH) forests, while also exploring a new method to assess TLP using reflectance spectroscopy.

The TLP of 20 common BLH tree species were measured on Richard K. Yancey Wildlife Management Area, located in Concordia Parish, Louisiana, U.S. TLP was measured in two habitats (non-flooded and seasonally flooded) and at two timepoints within the growing season (early and late). Analysis showed a positive TLP-drought tolerance relationship among BLH species, with high values of TLP found to be associated with high drought tolerance scores. A linear mixed effects model was used to assess how species, season, habitat, and tree diameter influence TLP’s plasticity within the floodplain forest system.

Building upon recent studies, I then related hyperspectral reflectance data with measured TLP values using reflectance spectroscopy among 17 BLH species within and across growing season. I used partial least squares regression (PLSR) to predict TLP using hyperspectral reflectance, creating a new and rapid method for TLP measurement. The best-fit model utilized the short-wave-infrared (SWIR) region (i.e., 1100–2400 nm) and demonstrated the capacity to rapidly estimate TLP (R2 = 0.72). For comparison, I successfully used the same approach to estimate leaf mass per area (LMA) (R2 = 0.85) using samples from the same leaves. This study demonstrates the feasibility of accurately estimating TLP with remote sensing approaches and presents the potential for the scaling up and incorporation into vegetation demographic and earth system modeling.

Date

11-3-2022

Committee Chair

Wolfe, Brett T.

DOI

10.31390/gradschool_theses.5679

Available for download on Sunday, November 02, 2025

Share

COinS