2024 Theses Doctoral
Do past winds protect forests from future storms? A multi-scale assessment of chronic wind-exposure and canopy structure impacts on hurricane damage in tropical forests
๐๐ง๐ญ๐ซ๐จ๐๐ฎ๐๐ญ๐ข๐จ๐ง) Tropical forests are the worldโs most structurally complex ecosystems, providing key functions like biomass accumulation and harboring biodiversity. Yet climate-change poses a potential threat to the stability of these forests โ tropical cyclones in the North Atlantic are projected to increase in intensity, leading to higher forest damage rates, potentially reducing their carbon sequestration and biodiversity potential. Hurricane Maria in 2017 was a possible portent of this dynamic, causing widespread devastation in Puerto Rico. How do forests resist such severe disturbances? Forests ecosystems contain ecological memory โ physical and biological legacies from past natural disturbances like fires and windstorms โ that can increase their resilience to future disturbances. In fire-prone forests, for example, prior exposure to non-severe fires has been shown to increase resistance to severe wildfires. Does the same mechanism apply in cyclone-prone tropical forests?
In this dissertation, I examine how chronic exposure to non-hurricane winds impacts hurricane damage at the tree, stand, and landscape scales in Puerto Rico. Specifically, I ask โ 1) Do chronic winds alter tree architecture to reduce their risk of stem-breaks? 2) Do chronic winds reduce forest stand structural complexity? 3) Do chronic winds and lower canopy structural complexity reduce individual tree and forest stand damage from Hurricane Maria?
๐๐๐ญ๐ก๐จ๐๐ฌ) I used a novel combination of remote sensing, fieldwork, and high-resolution Light Detection and Ranging (LiDAR) data collected in 2016 to address the above questions. In Chapter 1, I connected sub-meter resolution GPS data and 30 years of forest inventory with 0.03m resolution airborne LiDAR data to evaluate chronic wind impacts on the tree architecture and wind-risk of 124 forest trees of four key species. In Chapter 2, I used machine learning, remote-sensing and LiDAR data to predict the chronic wind impacts on the canopy height and structural complexity of ~20,000 0.28 ha forested sites across climatic, forest age and topographic gradients. In Chapter 3, I used pre-storm size and damage assessment field data for ~7,000 trees of 160 species across 14, 0.25 ha sites spanning an 800 m elevation gradient, alongside a remote-sensing dataset of ~12,000 forests to evaluate multiscale drivers โ including canopy structural complexity โ of individual, stand and landscape level hurricane damage.
๐๐๐ฌ๐ฎ๐ฅ๐ญ๐ฌ ๐๐ง๐ ๐๐จ๐ง๐๐ฅ๐ฎ๐ฌ๐ข๐จ๐ง๐ฌ) At the individual tree scale, I found that long-lived species grew ~3.5 m shorter and ~ 4 m2 smaller crowns on average due to chronic wind-exposure, substantially reducing their estimated wind-risk, whereas short-lived species did not respond architecturally to chronic winds. At the stand and landscape scales, I found that chronic winds reduced canopy height by 2.12 m on average, and that structural complexity decreased substantially with forest age. I found that stand-level hurricane damage was primarily a function of increased canopy structural complexity, which in turn decreased with elevation; and that individual tree damage increased with stem size and varied only slightly by species, with short-lived species much more susceptible to damage.
My findings suggest that tropical forest resistance to increasingly severe hurricanes depends largely on the physical structure of their canopies, and only then on adapted species-level life-history traits. The physical structure of forest canopies, in turn, changes substantially with exposure to non-hurricane winds. In old-growth forests in Puerto Rico, there is therefore evidence that ecological memory driven by exposure to non-hurricane winds can protect forests from severe wind disturbances. However, younger, more structurally complex forests may be potentially increasingly more vulnerable in a changing climate.
Geographic Areas
Subjects
Files
-
AnkoriKarlinsky_columbia_0054D_18867.pdf application/pdf 3.91 MB Download File
More About This Work
- Academic Units
- Ecology, Evolution, and Environmental Biology
- Thesis Advisors
- Uriarte, Maria
- Degree
- Ph.D., Columbia University
- Published Here
- October 16, 2024