Using Forest Inventory Data with Landsat 8 Imagery to Map Longleaf Pine Forest Characteristics in Georgia, USA

John Hogland,David L R Affleck,Joseph St Peter,Nathaniel Anderson

doi:10.3390/rs11151803

Abstract

This study improved on previous efforts to map longleaf pine (Pinus palustris) over large areas in the southeastern United States of America by developing new methods that integrate forest inventory data, aerial photography and Landsat 8 imagery to model forest characteristics. Spatial, statistical and machine learning algorithms were used to relate United States Forest Service Forest Inventory and Analysis (FIA) field plot data to relatively normalized Landsat 8 imagery based texture. Modeling algorithms employed include softmax neural networks and multiple hurdle models that combine softmax neural network predictions with linear regression models to estimate key forest characteristics across 2.3 million ha in Georgia, USA. Forest metrics include forest type, basal area and stand density. Results show strong relationships between Landsat 8 imagery based texture and field data (map accuracy > 0.80; square root basal area per ha residual standard errors < 1; natural log transformed trees per ha < 1.081). Model estimates depicting spatially explicit, fine resolution raster surfaces of forest characteristics for multiple coniferous and deciduous species across the study area were created and made available to the public in an online raster database. These products can be integrated with existing tabular, vector and raster databases already being used to guide longleaf pine conservation and restoration in the region.

Highlights

Longleaf pine (Pinus palustris) ecosystems are some of the most critically endangered ecosystems in the world [1]
Though the negative ecological impacts of longleaf pine losses are still being studied across the southeastern United States of America (USA), it is clear that these systems provide critical habitat for many species
To select final candidate National Agriculture Imagery Program (NAIP), Landsat 8 and combined NAIP/Landsat 8 based models for comparisons, we visually evaluated the subset of selected variables from the general additive modeling (GAM) procedure for variable transformations and chose predictor variables using Akaike information criterion (AIC), deviance, residual standard errors (RSE; for continuous variables) and predictor variable p-values

Summary

Introduction

Longleaf pine (Pinus palustris) ecosystems are some of the most critically endangered ecosystems in the world [1]. What remains of these once dominant forests supports many plants and animals and provides refuge for threatened and endangered species [2]. Though the negative ecological impacts of longleaf pine losses are still being studied across the southeastern United States of America (USA), it is clear that these systems provide critical habitat for many species. By 2014, at least 558,000 ha of longleaf pine had been restored, 445,000 ha had been improved through prescribed burning, 63,000 ha had been established through planting and 30,428 ha had been improved through the removal of invasive species and the opening of the forest canopy [4]. Early success heavily leveraged expert opinion related to the condition and location of longleaf ecosystems and capitalized on programs previously established to restore longleaf pine

Objectives

Methods

Results

Conclusion