Author

Jacob R Beard

Advisor

Matney, Thomas G.

Committee Member

Schultz, Emily B.

Committee Member

Grado, Stephen C.

Committee Member

Sabatia, Charles O.

Date of Degree

1-1-2016

Document Type

Dissertation - Open Access

Major

Forest Resources

Degree Name

Doctor of Philosophy

College

College of Forest Resources

Department

Department of Forestry

Abstract

The Mississippi Institute for Forest Inventory (MIFI) multi-product forest inventory divides Mississippi into five inventory regions with one region inventoried each year on a rotating basis. Resource analyses that overlap these temporally separated regions require adjustment to a common comparative time base by applying appropriate forest stand growth and harvest allocation models to the portions of a selected area not inventoried at the desired common time base. Currently the Mississippi Dynamic Inventory Reporter (MDIR) does not make adjustments to temporally synchronize portions of user selected working circles, polygons, or counties that occur in separate inventory regions. Separate inventory reports for each overlap area must be prepared to which growth and harvest are manually allocated to bring each area to the same point in time. The study objective was to provide an automated solution to temporal reconciliation by developing a growth and yield system that reconciles modeled timber volume growth, mortality, and harvests with known values from previous successive inventories and state tax records of harvested volumes at the county level. The modeling effort focused on constructing an optimized system for the Southwest MIFI 2004 and 2012 inventories. Species group specific, distance independent, tree-list models, including probability of survival and diameter growth equations, were developed through logistic and linear regressions, respectively. Probability of survival models were assessed for model performance using logistic regression concordant/discordant pairs. R2 and parameter p-values were used to evaluate diameter growth model performance. As the 2004 and 2012 datasets are each composed of randomly selected plots within the Southwest region, county totals were used for temporal comparison. County level Doyle volume calibration was within 150 units of tolerance for all counties in the Southwest region. The resulting growth and yield system represents a successful effort to develop a methodology for bridging temporally separated MIFI inventory analyses, while providing newly developed diameter and mortality equations for the state. The accompanying computer application allows the addition of both enhanced growth and yield and stand table projection models. System implementation will greatly reduce the time required for producing multi-temporal analyses and, thus, increase their usability and functionality.

URI

https://hdl.handle.net/11668/20308

Comments

inventory||growth and yield||forest||forestry||growth modelling||growth models||forest modelling||inventory projection||temporal adjustment

Share

COinS