This dataset portrays 61 forest group types across the
southeastern United States. These data were derived from a product created by
the USFS Forest Inventory and Analysis (FIA) program and the Remote Sensing
Applications Center (RSAC). The original
dataset (accessible from the
US Forest Service Website)
consists of 141 forest types across the contiguous United States.
The original dataset was downloaded on 12/13/2013, clipped
to the southeastern states, and used in a
generalization procedure which involved the application of a majority filter (8
neighbors, half replacement threshold) followed by a sieving process which
removed pixels in groups of four or less by reassigning them to their nearest
neighbor value.
Abstract:
This dataset portrays 141 forest types across the
contiguous United States. These data were derived from MODIS composite
images from the 2002 and 2003 growing seasons in combination with nearly
100 other geospatial data layers, including elevation, slope, aspect,
ecoregions, and PRISM climate data.
The dataset was developed as a collaborative effort
between the USFS Forest Inventory and Analysis and Forest Health
Monitoring programs and the USFS Remote Sensing Applications Center.
Purpose:
The purpose of this dataset is to portray broad
distribution patterns of forest cover in the United States and provide
input to national scale modeling projects. The data should not be
displayed at scales smaller than 1:2,000,000. Independent accuracy
assessments were conducted using a 10% holdout of the training data, and
are available through the data source contact.
Complete metadata is available on the
USFS website in HTML format.