The data that I propose to use come from a long-term forest inventory taken from UVM’s Jericho Research Forest. The research area is known as the Farm Woodlot and was managed as a “typical Vermont woodlot” to determine practical methods to grow high quality wood, increased growth and stocking, and uses for inferior species. Between 1948 and 1963, fifteen permanent forest plots (sizes range from 0.3 ac - 1 ac) located in five forest types were inventoried every five years. Inventories included measurements such as tree ID (coded number), species, diameter, and form (e.g., living, dead, cut). Due to funding cuts, no measurements were taken again until 2017 however the Farm Woodlot was still harvested from time to time (rough volume and temporal records exist). Despite this long gap in measurements, we are still able to use this long-term data set to ask questions such as:
How has management influenced structural and species diversity and the evolution of functional profiles over time?
Are there tradeoffs in managing for high growth/carbon (e.g., mitigation strategies) compared to stand complexity (e.g., resilience strategies).
For this project, I am interested in looking at a subset of my data to model forest structural developments over time
Specifically, I will try to use a Weibull Distribution to create a model that takes growth, mortality, and harvest intensity information from the four initial measurement periods to project stand development over time. Once I have a model, I would like to try to use use the 2017 measurement period to validate model performance.
One paper that supports the Weibull for forest structure distribution
Bayesian Estimation for the Three-Parameter Weibull Distribution with Tree Diameter Data