ForestERA Western Mogollon Plateau Tree Density (stems/ha) Metadata

Abstract:

This is a 90m resolution raster dataset describing tree stem densities (stems > 1" dbh / ha) across a 2 million acre area in Arizona referred to as the Western Mogollon Plateau. This data layer was created as part of the Forest Ecosystem Restoration Analysis (ForestERA) project to support landscape-scale forest restoration planning efforts by a broad group of stakeholders including federal and state agencies, academic institutions, and non-governmental entities.  These data are intended for regional analyses over spatial extents on the order of tens to hundreds of thousands of acres, and were not developed for use at finer spatial scales, although they may be useful for some applications at those finer scales.  TO VIEW ADDITIONAL SPATIAL METADATA IN ARCCATALOG, USE THE METADATA TOOLBAR. SELECT METADATA PROPERTIES> ENCLOSURES> AND DOUBLE CLICK THE METADATA TEXT FILE THERE TO OPEN IT.  FORESTERA ALSO DISTRIBUTED THIS TEXT FILE CONTAINING ADDITIONAL METADATA AS A SEPARATE DOCUMENT IN THE FOLDER CONTAINING THE SPATIAL DATA.

Purpose:

This layer was developed by the ForestERA project for use in landscape-level planning and prioritization of forest management on the western Mogollon Plateau.  It was designed for multiple purposes, including wildlife habitat modeling, fire behavior modeling, modeling forest managment applications, and forest managment planning.

Supplemental_Information:

This layer was developed using a regression tree methodology and Landsat 7 Enhanced Thematic Mapper (ETM) imagery from the fall of 1999 and summer of 2000.  We used Cubist software (Rulequest Research) for the modeling effort with the following settings: minimum number of samples for a rule = 9, 10% interpolation allowed, and a committee model built from 20 member trees.  We used 27 predictive variables in the classification tree, including slope, aspect, elevation, canopy cover, principle components layers derived from ETM imagery, a Normalized Derived Vegetation Index (NDVI) layer, and "texture" layers derived from ETM imagery and the Digital Elevation Model (DEM).  Training data came from over 800 (11.3 m radius or 0.04 ha) ground plots scattered across the region.

Accuracy assessment was undertaken using independant data collected on 567 (11.3m radius or 0.04 ha) ground plots at 63 locations across the study area.  The results of the assessment indicate that a highly significant relationship exists between the ETM-derived basal area estimates in the layer and the values for tree density measured on the ground (R2 = 0.584, P < 0.0001).  The slope of the line from this relationship is 0.99 indicating that the ETM-derived tree density estimates are nearly unbiased with relation to the actual tree densities on the ground.  

Based on the differences between the values for basal area measured on the ground and the predicted basal area values in the ETM derived layer, we have determined that values for 50% of the pixels in the ETM derived layer lie within 80 trees/ha of the actual values for canopy cover on the ground, and values for over 80% of the pixels lie within 150 trees/ha of the actual values on the ground.

These data are intended for regional analyses over spatial extents on the order of tens to hundreds of thousands of acres, and were not developed for use at finer spatial scales, although they may be useful for some applications at those finer scales.

This layer is provided for public use by the ForestERA Project in the Center for Environmental Sciences at NAU. Reports, presentations, and publications in which this layer is presented or used should acknowledge the ForestERA project.

This layer was developed by the Forest Ecosystem Restoration Analysis (ForestERA) team in the Center for Environmental Sciences and Education at Northern Arizona University. Funding was provided by the Ecological Restoration Institute (ERI) at NAU. The original data layer is considered the property of the ForestERA project, Northern Arizona University (NAU).