Preliminary Analysis of Forest Stand Disturbances in Coastal Georgia (USA) Using Landsat Time Series Stacked Imagery

Abstract

A determination of forest characteristics across broad areas is of great concern to the forest industry in the southern United States, as timber supply decisions can be based on opportunities, or lack of thereof, across all wood procurement areas. This is important in areas such as the southern United States, where the land ownership distribution is highly fragmented and where no general comprehensive source of forest data exists other than the low-intensity USDA Forest Service FIA surveys. In an effort to describe forest characteristics along the lower Coastal Plain of the State of Georgia (USA), we utilized a time series of Landsat data and an algorithm that assesses an integrated forest Z score. The methodology was used to create disturbance maps for over 30 years that represent the year of disturbance for specific locations. The overall accuracy was 52% when all years were considered, and approximately 70% from 1991 forward. Preliminary findings showed moderate levels of accuracy when determining ages for current forests, most of which are even-aged nature stands. Further modifications to the process were necessary to adapt to the unique conditions of study region. The modeling process also prompted several areas for future refinement, including improvement of the temporal resolution of the analysis by using all the available Landsat imagery and detection of the regeneration that normally occurs several years after disturbances.

Mapping Forest Disturbances between 1987–2016 Using All Available Time Series Landsat TM/ETM+ Imagery: Developing a Reliable Methodology for Georgia, United States

“…Although the results of the accuracy assessment suggest that this new method is better than our previous analysis, which attempted to detect the disturbance history of seven counties in Coastal Georgia [20], our new method still contained several areas of concern. For instance, on the commission error side of the analysis, we found that some forest thinnings could have been misclassified as a final harvest disturbance.…”

“…However, as the GFC dataset provides the disturbance history starting only after the year 2000, it was necessary to develop our own model to detect the disturbances occurring from the 1980s onward. Although some research has been dedicated to partially tracking the forest disturbance history in Georgia [20][21][22], with the exception of the GFC, no research has covered the disturbance history of the entire state of Georgia.…”

Random Forest Regression Model for Estimation of the Growing Stock Volumes in Georgia, USA, Using Dense Landsat Time Series and FIA Dataset

“…The Piedmont ecoregion is located between the Appalachian Mountains and the Southeastern Plains, and it includes the Atlanta metropolitan area, where more than 50% of the Georgia population resides. In the Appalachian mountain area, most of the forests are hardwood or mixed forest that are less frequently disturbed [35].…”