Publication Date

8-2023

Date of Final Oral Examination (Defense)

April 2023

Type of Culminating Activity

Thesis

Degree Title

Master of Science in Geoscience

Department

Geosciences

Supervisory Committee Chair

Nancy F. Glenn, Ph.D.

Supervisory Committee Member

Anna Bergstrom, Ph.D.

Supervisory Committee Member

Jeffrey Warren, Ph.D.

Abstract

Northern peatlands are major terrestrial carbon sinks, storing 415 ± 150 Gt of carbon. The composition of peatland vegetation affects this carbon storage capacity, and thus quantifying the vegetation helps to constrain uncertainty in peatland carbon storage estimates. Ground layer vegetation, such as Sphagnum sp. moss contributes greatly to carbon storage capacity. In forested peatlands, the tree canopy structure directly influences peatland solar insolation, soil temperature, and water table levels. Each of these factors impacts the ground layer vegetation. Currently, there is uncertainty about how the peatland tree canopy structure is influenced by elevated levels of carbon dioxide (CO2) and temperature. Providing canopy structural metrics in a nondestructive, spatially comprehensive way across different temperature and CO2 treatments is challenging for traditional methods such as destructive harvesting, Digital Hemispherical Photography (DHP), and allometric regressions. Terrestrial Laser Scanning (TLS) is well-suited to provide non-destructive detailed horizontal and vertical canopy structural information.

As part of the Spruce and Peatland Responses Under Changing Environments (SPRUCE) study located in northern Minnesota, USA, we use TLS to evaluate leaf area index (LAI), leaf area density, and leaf inclination angle over time (2015 - 2022) and space of two conifer species, Picea mariana (black spruce) and Larix laricina (eastern larch). The SPRUCE site is in a treed peatland bog under elevated CO2 and temperature conditions. The research questions in this study are 1) How accurately can we predict the LAI of the spruce and larch trees using TLS data? 2) How are the spruce and larch tree canopy structures within 12 SPRUCE plots changing from 2015 - 2022? We expected 1) A volumetric pixel-based model (VCP) will predict LAI with an accuracy of 90% as validated by destructively harvested and DHP LAI estimates 2) Under elevated CO2 and temperature, LAI will increase, leaf area density will decrease in lower canopies, and leaf inclination angles will become more vertical. At the species level, we expected the spruce and larch trees to respond with opposing trends for each metric under the same treatment. Using TLS data, we developed a modified VCP model that uses measures of point contact frequency to estimate LAI, leaf inclination angles, and leaf area density. The results indicate that the model predicts LAI with a coefficient of determination of 0.89 (R2 = 0.89), an RMSE of 0.98, and a normalized RMSE = 0.17. We also found that the model maintains moderate accuracy across voxel size input parameters, suggesting it may maintain accuracy in different treatment conditions where tree structural relationships can change. Our canopy structural results supported the hypothesis that LAI increases more significantly over time under warmer conditions when compared to control plots. Lower canopy leaf area density trends did not support the hypothesis as they showed no statistically significant trends across time. Leaf inclination angle trends through time did not support the hypothesis as they tended to decrease. As temperatures increased across the temperature gradient, leaf angles became more vertical in upper canopies under elevated CO2, leading to inconclusive support for or against the hypothesis. Species data did not support the hypothesis that spruce and larch canopy structures would differ significantly under the same treatments. The larch LAI, however, did not increase as significantly through time as the spruce under elevated CO2 conditions. Additionally, we identified anomalous fluctuations in time series data and proposed potential temperature thresholds where LAI differed the most under ambient or elevated CO2 conditions.

The findings from this study suggest that accurately quantifying canopy structure through time may be possible in different environmental conditions and species using TLS. We add support to previous findings that LAI increases more significantly through time under warming conditions compared to control conditions. These results demonstrate TLS’s utility for making species-level canopy structural estimates across horizontal and vertical profiles. Incorporating vertical canopy profile metrics such as leaf area density with LAI data can assist in better explaining how LAI is changing across time and temperature gradients.

DOI

https://doi.org/10.18122/td.2132.boisestate

Download

Share

COinS