Status and Trends of Land Change in
Selected U.S. Ecoregions - 2000 to 2011
Kristi L. Sayler, William Acevedo, and Janis L. Taylor
Abstract
U.S. Geological Survey scientists developed a dataset of 2006
and 2011 land-use and land-cover (
LULC
) information for select-
ed 100-km
2
sample blocks within 29 U.S. Environmental Protec-
tion Agency (EPA) Level III ecoregions across the conterminous
United States. The data can be used with the previously pub-
lished Land Cover Trends Dataset: 1973 to 2000 to assess land-
use/land-cover change across a 37-year study period. Results
from analysis of these data include ecoregion-based statistical
estimates of the amount of
LULC
change per time period, ranking
of the most common types of conversions, rates of change, and
percent composition. Overall estimated amount of change per
ecoregion from 2001 to 2011 ranged from a low of 370 km
2
in
the Northern Basin and Range Ecoregion to a high of 78,782 km
2
in the Southeastern Plains Ecoregion. The Southeastern Plains
continues to encompass one of the most intense forest harvesting
and regrowth regions in the country, with 16.6 percent of the
ecoregion changing between 2001 and 2011. These
LULC
change
statistics provide a new, valuable resource that complements oth-
er reference data and field-verified
LULC
data. Researchers can
use this resource to independently validate other land change
products or to conduct regional land change assessments.
Introduction/Background
The land-use and land-cover (
LULC
) of the United States is
constantly changing. Forests are clear-cut, urban growth and
development replaces agriculture, wildfire converts shrubland
to grassland, and flooding sometimes changes agriculture
to wetlands. Mapping and monitoring land-cover change is
crucial for understanding how the world’s resources will adapt
to these changes. Fortunately, national-scale
LULC
datasets
available for the United States have increased dramatically, es-
pecially since the advent of free Landsat data. However, these
land-cover maps provide land change statistics that vary in the
amount of change and composition for the same geographic
areas. Accurate land-cover data are therefore needed to confirm
through independent approaches that
LULC
products are in gen-
eral agreement with the prevailing convergence of evidence.
Land-cover change has long been regarded as one of the
most important variables of global change affecting ecologi-
cal systems (Vitousek, 1994; Giri
et al
., 2013), and it has also
been shown to be an important factor in predicting species
distributions (Sohl, 2014). Land-cover change is also consid-
ered to be both a cause and consequence of climate change
(Turner
et al
., 2007), and thus monitoring land-cover change
is crucial for understanding how the world’s resources will
adapt to climate change (Hibbard
et al
., 2010). The need for
accurate land-cover data by the land-cover change monitoring
community continues to grow (Foody, 2002; Foody, 2015).
Researchers need better validation data to evaluate land-use
and land-cover maps; unfortunately, the data sources dating
back in time are rare (Strahler
et al
., 2006). Validation and
map accuracy assessment usually involves the comparison of
map classifications to a higher quality reference classification
on a location-by-location basis, where the reference classifica-
tion is the best available assessment of the ground conditions
(Stehman, 2012).
This paper describes the 2006 and 2011
LULC
data and land
change statistics compiled for selected U.S. Environmental Pro-
tection Agency (
EPA
) Level III (
EPA
, 1999) ecoregions across the
conterminous United States. The dataset extends the document-
ed record of land-cover change previously developed by the
U.S. Geological Survey (
USGS
) Land Cover Trends (
LCT
) project,
which established a
LULC
change database from 1973 to 2000 for
the conterminous United States (Loveland
et al
., 2002; Sleeter
et al
., 2013, Soulard
et al
., 2014). The manually delineated
LULC
data can be used in validation and accuracy assessment of land
change products derived from remotely sensed data sources and
for continuing regional assessments of
LULC
change.
Methods
For the initial 1973 to 2000
LCT
assessment, geographers
across the country manually interpreted 100 km² sample
blocks using Landsat Multispectral Scanner (
MSS
), Thematic
Mapper (
TM
), and Enhanced Thematic Mapper Plus (
ETM+
)
satellite imagery spanning five dates (nominally 1973, 1980,
1986, 1992, and 2000). The sample blocks were chosen based
on a statistical sampling approach (Stehman
et al
., 2003) that
used the 84
EPA
Level III ecoregions as the regions of stratifica-
tion.
LULC
change was estimated on an ecoregion-by-ecoregion
basis using a classification legend that consisted of 11 modi-
fied Anderson Level I land-cover classes (Water, Developed,
Mechanically disturbed, Mining, Barren, Forest, Grassland/
Shrubland, Agriculture, Wetland, Non-mechanically dis-
turbed, Ice and snow). Sleeter
et al
. (2013) described each
land-cover class in detail. Geographers used the resulting
land-cover data for each sample block to determine change
across four temporal periods and to generate change estimates
for the entire ecoregion. The statistics on rates of change were
combined with field observations and corollary socioeconom-
ic variables to determine the overall trends, driving forces,
and key consequences of change for each ecoregion.
The update process started by creating a prioritized list
of
EPA
Level III ecoregions that underwent specific types and
amounts of land change between 1973 and 2000. Ecoregions
Kristi L. Sayler is with the U.S. Geological Survey, Earth Re-
sources Observation and Science (EROS) Center, 47914 252
nd
Street, Sioux Falls, SD 57198 (
).
William Acevedo is with the U.S. Geological Survey, Western
Geographic Science Center, 345 Middlefield Road MS 531,
Menlo Park, CA 94025.
Janis L. Taylor is with SGT Inc., Contractor to U.S. Geological
Survey, Earth Resources Observation and Science (EROS)
Center, 47914 252
nd
Street, Sioux Falls, SD 57198.
Photogrammetric Engineering & Remote Sensing
Vol. 82, No. 9, September 2016, pp. 687–697.
0099-1112/16/687–697
© 2016 American Society for Photogrammetry
and Remote Sensing
doi: 10.14358/PERS.82.9.687
