A Customized Image Classification Framework to Develop Regional-scale, High-resolution Conifer Maps

Submitted by tadamson on

The substantial distributional expansion and infill of pinyon (Pinus monophylla) and juniper (Juniperus sp.) trees (hereafter, "conifer") into sagebrush (Artemisia spp.) ecosystems since the late 1800s threatens the ecological function and economic viability of these ecosystems and represents a major contemporary challenge facing land and wildlife managers. The ecological consequences of the expansion are influenced by the proportion of coniferous cover within the ecosystem. Conifer-dominated systems decrease resilience to disturbance and resistance to invasion, reduce forage for cattle, increase soil erosion from water runoff, and promote wildfire activity. However, even relatively low conifer cover represents a primary threat to obligate wildlife populations such as the greater sage-grouse (Centrocercus urophasianus; hereafter, "sage-grouse"). 

Managers require accurate and high-resolution maps of conifer distribution and abundance across broad geographic extents to help guide land management decisions that better target areas for conifer treatment projects, especially for sage-grouse habitat restoration in sites characterized by scattered, isolated trees. However, available remotely sensed layers lacked the spatial resolution or accuracy to meet this need. Researchers developed a framework to map conifers at a high resolution (1 meter) across the majority of Nevada and northeastern California using image classification. The team used digital orthophoto quad tiles from the U.S. Department of Agriculture’s National Agriculture Imagery Program (2010 and 2013) to classify conifers using automated feature extraction tools. Overall accuracy was more than 86% across all mapped areas for both image validation and ground referencing methods. Four sets of full-extent maps were provided for land managers: (1) a shapefile representing accuracy results linked to mapping subunits, (2) binary rasters representing conifer presence or absence at 1-meter resolution, (3) a 900-meter resolution raster representing percentages of conifer canopy cover within each cell, and (4) 1-meter resolution canopy cover classification rasters derived from a 50-meter radius moving window analysis. These products improve upon or complement existing conifer maps for the western U.S. and may facilitate sagebrush ecosystem restoration planning through an accurate understanding of conifer distribution and abundance at multiple spatial scales. Associated manuscripts are available here, here, and here.

Spatial layers produced using automated feature extraction methods across greater sage-grouse habitat in Nevada and California, depicting (left) conifer presence or absence at a 1-meter resolution, (center) continuous conifer canopy cover at 900-meter resolution, and (right) an example of using canopy cover bins at 1-meter resolution produced from a 50-meter radius moving window to depict progressive phases of conifer expansion. The associated manuscript is available here.

Disclaimer: Any use of trade, firm, or product names is for descriptive purposes only and does not imply endorsement by the U.S. Government.

Author Name
Pete Coates
Author Email