Land Cover Trends Project

Abstracts of Published Papers

Get Abobe Reader from
Note: Some documents are in a PDF format, you will need Adobe Reader to view them.

Status and Trends


Acevedo, W. and Jellison, P.J., eds., 2007,Status and trends of Eastern United States land cover: U.S. Geological Survey Professional Paper, (in press).
This U.S. Geological Survey Professional Paper is the first volume of a four-volume series on the status and trends of the Nation’s land cover, providing an assessment of land cover and land cover change rates and causes in the eastern United States. Subsequent volumes will provide similar analyses for the Great Plains, Western U.S., and Midwest and Mississippi Gulf Coast. Land cover trends analyses are conducted on an ecoregion-by-ecoregion basis. Each ecoregion assessment is guided by a nationally-consistent study design, including statistical methods, field studies and analysis. Taken individually the assessments provide a picture of the types, rates, and sequence of change occurring in a given ecoregion; when taken together they provide a framework for understanding an extremely complex mosaic of change and its causes and consequences. Thus, each volume in this series provides a regional assessment of how (and how fast) land cover and land use are changing, and why. Taken together, they form the first comprehensive national picture of U.S. land change.  

Geographic Assessments

Raumann, C.G. and Soulard, C.E., 2007, Land cover trends in the Sierra Nevada ecoregion, 1973-2000: U.S. Geological Survey Scientific Investigations Report 2007-5011, 29 p.
The U.S. Geological Survey has developed and is implementing the Land Cover Trends project to estimate and describe the temporal and spatial distribution and variability of contemporary land-use and land-cover change in the United States. As part of the Land Cover Trends project, the purpose of this study was to assess land-use/land-cover change in the Sierra Nevada ecoregion for the period 1973 to 2000 using a probability sampling technique and satellite imagery. We randomly selected 36 100 km² sample blocks to derive thematic images of land-use/land-cover for five dates of Landsat imagery (1973, 1980, 1986, 1992, 2000). We visually interpreted as many as 11 land-use/land-cover classes using a 60-meter minimum mapping unit from the five dates of imagery yielding four periods for analysis. Change-detection results from post-classification comparison of our mapped data showed that landscape disturbance from fire was the dominant change from 1973-2000. The second most-common change was forest disturbance resulting from harvest of timber resources by way of clear-cutting. The rates of forest regeneration from temporary fire and harvest disturbances coincided with the rates of disturbance from the previous period. Relatively minor landscape changes were caused by new development and reservoir drawdown. Multiple linear regression analysis suggests that land ownership and the proportion of forest and developed cover types were significant determinants of the likelihood of direct human-induced change occurring in sampling units. Driving forces of change include land ownership, land management such as fire suppression policy, and demand for natural resources.  

Sleeter, B.M. and Raumann, C.G., 2006, Land-cover trends in the Mojave basin and range ecoregion: U.S. Geological Survey Scientific Investigations Report 2006-5098, 15 p.
The U.S. Geological Survey’s Land-Cover Trends Project aims to estimate the rates of contemporary land-cover change within the conterminous United States between 1972 and 2000. A random sampling approach was used to select a representative sample of 10-km by 10-km sample blocks and to estimate change within +/−1 percent at an 85-percent confidence interval (Stehman and others, 2003; Loveland and others, 2002). Landsat Multispectral Scanner, Thematic Mapper, and Enhanced Thematic Mapper Plus data were used, and each 60-m pixel was assigned to one of 11 distinct land-cover classes based upon a modified Anderson classification system. Upon completion of land-cover change mapping for five dates, land-cover change statistics were generated and analyzed. This paper presents estimates for the Mojave Basin and Range ecoregion located in the southwestern United States. Our research suggests land-cover change within the Mojave to be relatively rare and highly localized. The primary shift in land cover is unidirectional, with natural desert grass/shrubland being converted to development. We estimate that more than 1,300 km2 have been converted since 1973 and that the conversion is being largely driven by economic and recreational opportunities provided by the Mojave ecoregion. The time interval with the highest rate of change was 1986 to 1992, in which the rate was 0.21 percent (321.9 km2) per year total change.   

Soulard, C.E., 2006, Land-cover trends of the central basin and range ecoregion: U.S. Geological Survey Scientific Investigations Report 2006-5288, 20 p.
This report presents an assessment of land-use/land-cover (LU/LC) change in the Central Basin and Range ecoregion (fig. 1) for the period 1973-2000. The Central Basin and Range ecoregion is one of 84 Level-III ecoregions as defined by the Environmental Protection Agency (EPA, 1999, Omernik, 1987). Ecoregions have served as a spatial framework for environmental resource management and to denote areas that contain a geographically distinct assemblage of biotic and abiotic phenomena including geology, physiography, vegetation, climate, soils, land use, wildlife, and hydrology. The established Land Cover Trends methodology generates estimates of LU/LC change using a probability sampling approach and change-detection analysis of thematic land-cover images derived from Landsat satellite imagery.  

Brown, D.G., Johnson, K.M., Loveland, T.R., and Theobald, D.M., 2005, Rural land use change in the conterminous U.S., 1950-2000: Ecological Applications, Vol. 15, No. 6, pp. 1851–1863.
In order to understand the magnitude, direction, and geographic distribution of land-use changes, we evaluated land-use trends in U.S. counties during the latter half of the 20th century. Our paper synthesizes the dominant spatial and temporal trends in population, agriculture, and urbanized land uses, using a variety of data sources and an ecoregion classification as a frame of reference. A combination of increasing attractiveness of nonmetropolitan areas in the period 1970–2000, decreasing household size, and decreasing density of settlement has resulted in important trends in the patterns of developed land. By 2000, the area of low-density, exurban development beyond the urban fringe occupied nearly 15 times the area of higher density urbanized development. Efficiency gains, mechanization, and agglomeration of agricultural concerns has resulted in data that show cropland area to be stable throughout the Corn Belt and parts of the West between 1950 and 2000, but decreasing by about 22% east of the Mississippi River. We use a regional case study of the Mid-Atlantic and Southeastern regions to focus in more detail on the land-cover changes resulting from these dynamics. Dominating were land-cover changes associated with the timber practices in the forested plains ecoregions and urbanization in the piedmont ecoregions. Appalachian ecoregions show the slowest rates of land-cover change. The dominant trends of tremendous exurban growth, throughout the United States, and conversion and abandonment of agricultural lands, especially in the eastern United States, have important implications because they affect large areas of the country, the functioning of ecological systems, and the potential for restoration.  

Napton, D. and Loveland, T.R., 2004, Land cover and land use change: 1973-2000: in WorldMinds: Geographical Perspectives on 100 Problems, Janelle, D., Warf, B., and Hansen, K., eds., New York: Kluwer Academic Publishers, pp. 261-266.
Humans change land use in order to improve their quality of life, but these changes often have consequences that affect other land, water, air, people, or species.  Issues resulting from land alteration affect all Americans and include changes to climate, biodiversity, human health, water quality, and urban sprawl.  To manage and ameliorate the consequences of land use and land cover changes we must first determine the types and rates of changes that are occurring, their locations, and the driving forces of change.  The USGS is producing an atlas of recent U.S. land use and land cover change.  This research demonstrates how the integration of remote sensing, air photo interpretation, GIS, field work, regionalization techniques, and socioeconomic data allows geographers to evaluate land changes.  Geographers interpret satellite imagery from samples of ecoregions of the conterminous U.S., and the results are extrapolated to determine land change throughout the ecoregion.  Maps and other data analysis are combined with the imagery interpretation and with socioeconomic data and other ancillary information to provide an assessment of the rates, driving forces, and consequences of land changes.  This new method of large area land assessment should be applicable to continental and global land use change studies.   

Napton, D., Auch, R., Sohl, T.L., and Loveland, T.R., 2003, Land use and land cover change in the north central appalachians ecoregion: The Pennsylvania Geographer, Vol. 41, No. 2, pp. 46-66.
The North Central Appalachians ecoregion, spanning northern Pennsylvania and southern New York, has a long history of land use and land cover change. Turn-of-the-century logging dramatically altered the natural landscape of the ecoregion, but subsequent regeneration returned the ecoregion to a forest dominated condition. To understand contemporary land use and land cover changes, the U.S. Geological Survey with NASA and the U.S. Environmental Protection Agency used a random sample of satellite remotely sensed data for 1973, 1980, 1986, 1992, and 2000 to estimate the rates and assess the primary drivers of change in the North Central Appalachians. The overall change was 6.2%. The 1973-1980 period had the lowest rate of change (1.5%); the highest rate (2.9%) occurred during the 1992-2000 period.  The primary conversions were deforestation through harvesting and natural disturbance (i.e., tornados) followed by regeneration, and conversion of forests to mining and urban lands. The primary drivers of the change included changes in access, energy and forest prices, and attitudes toward the environment.  

Griffith, J.A., Stehman, S.V., Loveland, T.R., 2003, Landscape trends in mid-Atlantic and Southeastern United States ecoregions: Environmental Management, Vol. 32, No. 5, pp. 572-588.
Landscape pattern and composition metrics are potential indicators for broad-scale monitoring of change and for relating change to human and ecological processes. We used a probability sample of 20-km x 20-km sampling blocks to characterize landscape composition and pattern in five US ecoregions: the Middle Atlantic Coastal Plain, Southeastern Plains, Northern Piedmont, Piedmont, and Blue Ridge Mountains. Land use/and cover (LULC) data for five dates between 1972 and 2000 were obtained for each sample block. Analyses focused on quantifying trends in selected landscape pattern metrics by ecoregion and comparing trends in land cover proportions and pattern metrics among ecoregions. Repeated measures analysis of the landscape pattern documented a statistically significant trend in all five ecoregions towards a more fine-grained landscape from the early 1970s through 2000. The ecologically important forest cover class also became more fine-grained with time (i.e., more numerous and smaller forest patches). Trends in LULC, forest edge, and forest percent like adjacencies differed among ecoregions. These results suggest that ecoregions provide a geographically coherent way to regionalize the story of national land use and land cover change in the United States. This study provides new information on LULC change in the southeast United States. Previous studies of the region from the 1930s to the 1980s showed a decrease in landscape fragmentation and an increase in percent forest, while this study showed an increase in forest fragmentation and a loss of forest cover.  

Griffith, J.A. and Loveland, T.R., 2001, Changes in landscape pattern and land use/cover over a 20 year period in the middle Atlantic coastal plain region: in Proceedings, 3rd International Conference on Geospatial Information in Agriculture and Forestry, 8 p.
Changes in land use and land cover (LULC) have demonstrable effects on wildlife habitat, water quality, climate, and carbon cycling. Although scientists recognize the importance of LULC change, very little information exists on the geographic distribution and rates of LULC change throughout the U.S. Even less information exists on changes in landscape pattern. The U.S. Geological Survey has begun a 4-year research program to study LULC change across the Nation. The goal is to use 30 years of satellite data to document the types, geographic distribution, and rates of change, and to determine some of the key drivers and consequences of the changes. Results presented focus on land use/cover and landscape pattern changes in the Middle Atlantic Coastal Plain Ecoregion. Analysis of variance was used on six selected landscape pattern metrics in eleven 20 km x 20 km sample blocks of LULC. Of the set of metrics, which described landscape texture, patch shape, and patch size, four showed significant changes and the other two were consistent with a trend toward a more fine-grained landscape in the ecoregion over time.  

Applications of Trends Data


Sohl, T.L., K.L. Sayler, M. Drummond, and T.R. Loveland, 2007, The FORE-SCE model: a practical approach for projecting land cover change using scenario-based modeling: Journal of Land Use Science, Vol. 2, No. 2, pp. 1-24.
A wide variety of ecological applications require spatially explicit, historic, current, and projected land use and land cover data. The U.S. Land Cover Trends project is analyzing contemporary (1973–2000) land-cover change in the 15 conterminous United States. The newly developed FORE-SCE model used Land Cover Trends data and theoretical, statistical, and deterministic modeling techniques to project future land cover change through 2020 for multiple plausible scenarios. Projected proportions of future land use were initially developed, and then sited on the lands with the highest potential for supporting that land use and 20 land cover using a statistically based stochastic allocation procedure. Three scenarios of 2020 land cover were mapped for the western Great Plains in the US. The model provided realistic, high-resolution, scenario-based land-cover products suitable for multiple applications, including studies of climate and weather variability, carbon dynamics, and regional hydrology.  

Hale, R.C., K.P. Gallo, T.W. Owen, and T.R. Loveland. 2006. Land use/land cover change effects on temperature trends at U.S. Climate Normals stations: Geophysical Research Letters, Vol. 33, L11703.
Alterations in land use/land cover (LULC) in areas near meteorological observation stations can influence the measurement of climatological variables such as temperature. Urbanization near climate stations has been the focus of considerable research attention, however conversions between non-urban LULC classes may also have an impact. In this study, trends of minimum, maximum, and average temperature at 366 U.S. Climate Normal stations are analyzed based on changes in LULC defined by the U.S. Land Cover Trends Project. Results indicate relatively few significant temperature trends before periods of greatest LULC change, and these are generally evenly divided between warming and cooling trends. In contrast, after the period of greatest LULC change was observed, 95% of the stations that exhibited significant trends (minimum, maximum, or mean temperature) displayed warming trends.  

Liu, J., S. Liu and T. Loveland, 2006, Temporal evolution of carbon budgets of the Appalachian forests in the U.S. from 1972 to 2000: Forest Ecology and Management, Vol. 222, pp. 191–201.
Estimating dynamic terrestrial ecosystem carbon (C) sources and sinks over large areas is difficult. The scaling of C sources and sinks from the field level to the regional level has been challenging due to the variations of climate, soil, vegetation, and disturbances. As part of an effort to estimate the spatial, temporal, and sectional dimensions of the United States C sources and sinks (the U.S. Carbon Trends Project), this study estimated the forest ecosystem C sequestration of the Appalachian region (186,000 km2) for the period of 1972–2000 using the General Ensemble Biogeochemical Modeling System (GEMS) that has a strong capability of assimilating land use and land cover change (LUCC) data. On 82 sampling blocks in the Appalachian region, GEMS used sequential 60 m resolution land cover change maps to capture forest stand-replacing events and used forest inventory data to estimate non-stand-replacing changes. GEMS also used Monte Carlo approaches to deal with spatial scaling issues such as initialization of forest age and soil properties. Ensemble simulations were performed to incorporate the uncertainties of input data. Simulated results show that from 1972 to 2000 the net primary productivity (NPP), net ecosystem productivity (NEP), and net biome productivity (NBP) averaged 6.2 Mg C ha-1 y-1 (+or-1.1), 2.2 Mg C ha-1 y-1 (+or-0.6), and 1.8 Mg C ha-1 y-1 (+or-0.6), respectively. The inter-annual variability was driven mostly by climate. Detailed C budgets for the year 2000 were also calculated. Within a total 148,000 km2 forested area, average forest ecosystem C density was estimated to be 186 Mg C ha-1 (+or-20), of which 98 Mg C ha-1 (+or-12) was in biomass and 88 Mg C ha-1 (+or-13) was in litter and soil. The total simulated C stock of the Appalachian forests was estimated to be 2751 Tg C (+or-296), including 1454 Tg C (+or-178) in living biomass and 1297 Tg C (+or-192) in litter and soil. The total net C sequestration (i.e. NBP) of the forest ecosystem in 2000 was estimated to be 19.5 Tg C y-1 (+or-6.8).  

Price, S.J., Dorcas, M.E., Gallant, A.L., Klaver, R.W., and Willson, J.D., 2006, Three decades of urbanization: estimating the impacts of land-cover change on stream salamander populations: Biological Conservation, Vol. 133, pp. 436-441.
Urbanization has become the dominant form of landscape disturbance in parts of the United States. Small streams in the Piedmont region of the eastern United States support high densities of salamanders and are often the first habitats to be affected by landscape-altering factors such as urbanization. We used US Geological Survey land cover data from 1972 to 2000 and a relation between stream salamanders and land cover, established from recent research, to estimate the impact of contemporary land-cover change on the abundance of stream salamanders near Davidson, North Carolina, a Piedmont locale that has experienced rapid urbanization during this time. Our analysis indicates that southern two-lined salamander (Eurycea cirrigera) populations have decreased from 32% to 44% while northern dusky salamanders (Desmognathus fuscus) have decreased from 21% to 30% over the last three decades. Our results suggest that the widespread conversion of forest to urban land in small catchments has likely resulted in a substantial decline of populations of stream salamanders and could have serious effects on stream ecosystems.  

Tan, Z., S. Liu, C. A. Johnston, T. R. Loveland, L. L. Tieszen, J. Liu, and R. Kurtz, 2005. Soil organic carbon dynamics as related to land use history in the northwestern Great Plains: Global Biogeochemical Cycles, Vol. 19, GB3011.
Strategies for mitigating the global greenhouse effect must account for soil organic carbon (SOC) dynamics at both spatial and temporal scales, which is usually challenging owing to limitations in data and approach. This study was conducted to characterize the SOC dynamics associated with land use change history in the northwestern Great Plains ecoregion. A sampling framework (40 sample blocks of 10 x 10 km2 randomly located in the ecoregion) and the General Ensemble Biogeochemical Modeling System (GEMS) were used to quantify the spatial and temporal variability in the SOC stock from 1972 to 2001. Results indicate that C source and sink areas coexisted within the ecoregion, and the SOC stock in the upper 20-cm depth increased by 3.93 Mg ha-1 over the 29 years. About 17.5% of the area was evaluated as a C source at 122 kg C ha-1 yr-1. The spatial variability of SOC stock was attributed to the dynamics of both slow and passive fractions, while the temporal variation depended on the slow fraction only. The SOC change at the block scale was positively related to either grassland proportion or negatively related to cropland proportion. We concluded that the slow C pool determined whether soils behaved as sources or sinks of atmospheric CO2, but the strength depended on antecedent SOC contents, land cover type, and land use change history in the ecoregion.  

Liu, S., Loveland, T.R., and Kurtz, R.M., 2004, Contemporary carbon dynamics in terrestrial ecosystems in the southeastern plains of the United States: Environmental Management, Vol. 33, Sup. 1, pp. 442-456.
Quantifying carbon dynamics over large areas is frequently hindered by the lack of consistent, high-quality, spatially explicit land use and land cover change databases and appropriate modeling techniques. In this paper, we present a generic approach to address some of these challenges. Land cover change information in the Southeastern Plains ecoregion was derived from Landsat data acquired in 1973, 1980, 1986, 1992, and 2000 within 11 randomly located 20 km  by 20 km sample blocks. Carbon dynamics within each of the sample blocks was simulated using the General Ensemble Biogeochemical Modeling System (GEMS), capable of assimilating the variances and covariance of major input variables into simulations using an ensemble approach. Results indicate that urban and forest areas have been increasing, whereas agricultural land has been decreasing since 1973. Forest clear-cutting activity has intensified, more than doubling from 1973 to 2000. The Southeastern Plains has been acting as a carbon sink since 1973, with an average rate of 0.89 Mg C/ha/yr. Biomass, soil organic carbon (SOC), and harvested materials account for 56%, 34%, and 10% of the sink, respectively. However, the sink has declined continuously during the same period owing to forest aging in the northern part of the ecoregion and increased forest clear-cutting activities in the south. The relative contributions to the sink from SOC and harvested materials have increased, implying that these components deserve more study in the future. The methods developed here can be used to quantify the impacts of human management activities on the carbon cycle at landscape to global scales.  

Liu S., Liu, J., and Loveland, T.R., 2004, Spatial-temporal carbon sequestration under land use and land cover change: in Proceedings, Geoiformatics 2004, University of Gavle, Sweden, pp. 525-532.
In this research carbon (C) sequestration of the Blue Ridge ecoregion of North America was investigated using the General Ensemble biogeochemical Modeling System (GEMS). GEMS assimilated historical land use and land cover change (LUCC) data within ten 20- km by 20-km sampling blocks in the ecoregion and performed biogeochemical C simulations for the period of 1973 – 2000. The LUCC data were derived from both low spatial resolution census and survey data (forest structure and agricultural cropping practices), and from high spatial resolution sequential land cover maps. The land cover maps were derived from Landsat remote sensing data at 60-meter resolution. GEMS used Monte Carlo approaches to deal with some spatial and temporal LUCC scaling issues such as initialization of forest age and crop species. It also prescribed the land use activities such as forest selective cuttings that were not reflected in the land cover change maps. Results showed that this ecoregion was a C sink during the simulation period. The sink averaged 100 – 120 g C m-2 yr-1 with a major portion (50-80%) attributed to living biomass and smaller portions attributed to soil and harvested C. Net primary productivity (NPP) in Blue Ridge ecoregion was about 600 to 800 g C m-2 yr-1. Based on the 10 sample blocks, estimation error of C sequestration at 95% confidence level is about 15 to 45 g C m-2 yr-1, varying by year. Model simulations also indicated that LUCC played a significant role in determining the magnitude of carbon sink strength in the region. Without considering the dynamics of LUCC, the C sink strength would be underestimated by 30 to50 percent.  

Scott, J.M., Loveland, T.R., Gergely, K., Strittholt, J., and Staus, N., 2004, National wildlife refuge system: ecological context and integrity: Natural Resources Journal, Vol. 44, No. 4, pp. 1041-1066<
The Refuge Improvement Act of 1997 established a statutory mission and management standards for the National Wildlife Refuge system. The U.S. Fish and Wildlife Service subsequently issued a policy for ensuring the biological integrity, diversity, and environmental health of the system. This policy requires understanding the management objectives of each refuge in a local, regional, and national context. An assessment of the refuge system in a national and regional context reveals that refuges are typically smaller than many conservation holdings and are unevenly distributed across the conterminous U.S. Western rangelands, coastal wetlands, and northern grasslands; wetlands are the best-represented ecosystems while temperate forests have the poorest representation. In contrast to other agency holdings or management designations in the national protected areas network (e.g., national parks, national forests, wilderness areas), refuges tend to occupy sites at lower elevations and that have higher productivity and soil quality. This difference points to the important contribution of the refuges in providing much needed ecological balance within the national protected areas network. However, the ecological integrity of the refuge system is challenged by the proximity of individual refuges to development. Overall, the refuges are becoming islands in a landscape matrix of urban and agricultural development. This creates future challenges for meeting management objectives to ensure the biological integrity, diversity, and environmental health of the system. If the policy to ensure biological integrity, diversity, and environmental health of the refuge system is to be successful, it may be more important to address issues about what happens on adjacent lands than uses within refuges.  

Overall Project Design

Gallant, A.L., Loveland, T.R., Sohl, T.L., and Napton, D., 2004, Using an ecoregion framework to analyze land cover and land use dynamics: Environmental Management, Vol. 34, pp. 89-110.
The United States has a highly varied landscape because of wide-ranging differences in combinations of climatic, geologic, edaphic, hydrologic, vegetative, and human management (land use) factors. Land uses are dynamic, with the types and rates of change dependent on a host of variables, including land accessibility, economic considerations, and the internal increase and movement of the human population. There is a convergence of evidence that ecoregions are very useful for organizing, interpreting, and reporting information about land-use dynamics. Ecoregion boundaries correspond well with patterns of land cover, urban settlement, agricultural variables, and resource-based industries. We implemented an ecoregion framework to document trends in contemporary land-cover and land-use dynamics over the conterminous United States from 1973 to 2000. Examples of results from six eastern ecoregions show that the relative abundance, grain of pattern, and human alteration of landcover types organize well by ecoregion and that these characteristics of change, themselves, change through time.  

Loveland, T.R. and DeFries, R., 2004, Observing and monitoring land use and land cover change: in Ecosystems and Land Use Change, Geophysical Monograph Series, Vol. 153, DeFries, R., Asner, G., and Houghton, R., eds., American Geophysical Union, Washington, DC, pp. 231-246.
Understanding the consequences of land use change requires robust documentation on the characteristics of change. Land use change observation and monitoring programs now rely on remotely sensed data coupled with field observations and corroborating information describing the social, economic, and physical dimensions of land use and land cover. Remote sensing approaches for observing and monitoring change vary depending on the geographic scope, ecological complexity, and the information required to understand ecosystem interactions. Strategies based on identifying spectral variability are useful for targeting areas of rapid change. Measuring changes in land cover biophysical properties requires a more complex approach, where different dates of remotely sensed data are transformed to such variables as surface imperviousness, canopy structure, and phenology, and then compared. Mapping the conversion of land use and land cover from one category to another (e.g., forest to urban) requires maps of the land use and land cover for two or more periods. These approaches have been used successfully at local, regional, and global scales using a range of remote sensing data (e.g., aerial photography, Landsat Thematic Mapper, Terra MODIS, Space Imaging’s IKONOS), field measurements, and other supplemental sources. Challenges remain, however, and scientific advances in change detection methods, accuracy assessment procedures, and improved strategies for using land cover to more specifically infer land use are needed so that continued improvements in the types and quality of change measures used to study land use and ecosystem interactions can be realized.  

Sohl, T.L., Gallant, A.L., and Loveland, T.R., 2004, The characteristics and interpretability of land surface change and implications for project design: Photogrammetric Engineering and Remote Sensing, Vol. 70, No. 4, pp. 439-448.
The need for comprehensive, accurate information on land cover change has never been greater. While remotely sensed imagery affords the opportunity to provide information on land-cover change over large geographic expanses at a relatively low cost, the characteristics of land-surface change bring into question the suitability of many commonly used methodologies. Algorithm-based methodologies to detect change generally cannot provide the same level of accuracy as the analyses done by human interpreters. Results from the Land Cover Trends project, a cooperative venture that includes the U.S. Geological Survey, Environmental Protection Agency, and National Aeronautics and Space Administration, have shown that land-cover conversion is a relatively rare event, occurs locally in small patches, varies geographically and temporally, and is spectrally ambiguous. Based on these characteristics of change and the type of information required, manual interpretation was selected as the primary means of detecting change in the Land Cover Trends project. Mixtures of algorithm-based detection and manual interpretation may often prove to be the most feasible and appropriate design for change-detection applications. Serious examination of the expected characteristics and measurability of change must be considered during the design and implementation phase of any change analysis project.  

Loveland, T.R., Sohl, T.L., Stehman, S.V., Gallant, A.L., Sayler, K.L., and Napton, D.E., 2002, A strategy for estimating the rates of recent United States land cover changes: Photogrammetric Engineering and Remote Sensing, Vol. 68, No. 10, pp. 1091-1099.
Information on the rates of land-use and land-cover change is important in addressing issues ranging from the health of aquatic resources to climate change. Unfortunately, there is a paucity of information on land-use and land-cover change except at very local levels. We describe a strategy for estimating land-cover change across the conterminous United States over the past 30 years. Change rates are estimated for 84 ecoregions using a sampling procedure and five dates of Landsat imagery. We have applied this methodology to six eastern U.S. ecoregions. Results show very high rates of change in the Plains ecoregions, high to moderate rates in the Piedmont ecoregions, and moderate to low rates in the Appalachian ecoregions. This indicates that ecoregions are appropriate strata for capturing unique patterns of land-cover change. The results of the study are being applied as we undertake the mapping of the rest of the conterminous United States.  

Loveland, T.R., Sohl, T.L., Sayler, K., Gallant, A., Dwyer, J., Vogelmann, J.E., and Zylstra,G.J., 1999, Land cover trends: rates, causes, and consequences of late-twentieth century U.S. land cover change: EPA Report, 52 p.
Information on the rates, driving forces, and consequences of land use and land cover change is important in studies addressing issues ranging from the health of aquatic resources to climate change. Land use and land cover changes occur at all scales, and changes at local scales can have dramatic, cumulative impacts at broader scales. Consequently, land use and land cover changes are not just of concern at local and regional levels (i.e., because of impacts on land management practices, economic health and sustainability, and social processes), but globally as well. Unfortunately, there is a paucity of information on land use and land cover change except at very local levels. This four-year research project between the U.S. Geological Survey and the U.S. Environmental Protection Agency has a goal to document the types, geographic distributions, and rates of land cover change on a region-by-region basis over the past 30 years for the conterminous U.S., and to determine some of the key drivers and consequences of the changes. The objectives of the study are to:
  1. Develop a comprehensive methodology for using sampling and change analysis techniques and Landsat MSS and TM data for measuring regional land cover change across the U.S.
  2. Develop a comprehensive methodology for using sampling and change analysis techniques and Landsat MSS and TM data for measuring regional land cover change across the U.S.
  3. Characterize the types, rates, and temporal variability of change for a 30-year period.
  4. Document regional driving forces and consequences of change.
  5. Prepare a national synthesis of land cover change.

Statistical Design


Stehman, S.V. Sohl, T.L., and Loveland, T.R., 2005, An evaluation of sampling strategies to improve precision of estimates of gross change in land use and land cover: International Journal of Remote Sensing, Vol. 26, No. 22, pp. 4941-4957.
Statistical sampling offers a cost-effective, practical alternative to complete coverage mapping for the objective of estimating gross change in land cover over large areas. Because land cover change is typically rare, the sampling strategy must take advantage of design and analysis tools that enhance precision. Using two populations of land cover change in the eastern United States, we demonstrate that the choice of sampling unit size and use of a survey sampling regression estimator can significantly improve precision with only a minor increase in cost.  

Stehman, S.V., 2005, Comparing estimators of gross change derived from complete coverage mapping versus statistical sampling of remotely sensed data: Remote Sensing of Environment, Vol. 96, pp. 466-474.
Area of gross change in land cover can be derived from a complete coverage land cover change map of a region of interest or estimated from a statistical sample of the region. Sampling may produce significant cost savings and more timely results because change is determined over a smaller total area than required by complete coverage mapping. Mean square error (MSE) defined in the context of a survey sampling measurement model is used to compare gross change estimators obtained from the two approaches. Measurement error bias attributable to error in classifying land cover change may occur with either the sampling or complete coverage mapping approach. An additional contribution to MSE attributable to sampling variability exists for the sampling-based estimator, but not the complete coverage estimator. If this sampling variability is small, the classification error bias of the sampling approach need not be reduced very far relative to the classification error bias of complete coverage to achieve similar MSE. Data from several published change accuracy error matrices are used to provide MSE comparisons for specific applications.  

Griffith, J.A., Stehman, S.V., Sohl, T.L., and Loveland, T.R., 2003, Detecting trends in landscape pattern metrics over a 20-year period using a sampling-based monitoring programme: International Journal of Remote Sensing, Vol. 24, No. 1, pp. 175-181.
Temporal trends in landscape pattern metrics describing texture, patch shape and patch size were evaluated in the US Middle Atlantic Coastal Plain Ecoregion. The landscape pattern metrics were calculated for a sample of land use/cover data obtained for four points in time from 1973-1992. The multiple sampling dates permit evaluation of trend, whereas availability of only two sampling dates allows only evaluation of change. Observed statistically significant trends in the landscape pattern metrics demonstrated that the sampling-based monitoring protocol was able to detect a trend toward a more fine-grained landscape in this ecoregion. This sampling and analysis protocol is being extended spatially to the remaining 83 ecoregions in the US and temporally to the year 2000 to provide a national and regional synthesis of the temporal and spatial dynamics of landscape pattern covering the period 1973-2000.  

Stehman, S.V., Sohl, T.L., and Loveland, T.R., 2003, Statistical sampling to characterize recent United States land-cover change: Remote Sensing of Environment, Vol. 86, Issue 4, pp. 517-529.
The U.S. Geological Survey, in conjunction with the U.S. Environmental Protection Agency, is conducting a study focused on developing methods for estimating changes in land-cover and landscape pattern for the conterminous United States from 1973 to 2000. Eleven land-cover and land-use classes are interpreted from Landsat imagery for five sampling dates. Because of the high cost and potential effect of classification error associated with developing change estimates from wall-to-wall land-cover maps, a probability sampling approach is employed. The basic sampling unit is a 20×20 km area, and land cover is obtained for each 60×60 m pixel within the sampling unit. The sampling design is stratified based on ecoregions, and land-cover change estimates are constructed for each stratum. The sampling design and analyses are documented, and estimates of change accompanied by standard errors are presented to demonstrate the methodology. Analyses of the completed strata suggest that the sampling unit should be reduced to a 10×10 km block, and poststratified estimation and regression estimation are viable options to improve precision of estimated change.  

Papers in Press or in Review


Auch, R.F., Northern Piedmont ecoregion report: U.S. Geological Survey, Scientific Investigations Report 2007-xxxx, (in review).
In this report we present an assessment of land-use and land-cover change in the Northern Piedmont ecoregion for the period 1973 to 2000. The Northern Piedmont is one of 84 Level-III ecoregions defined by the Environmental Protection Agency (U.S. EPA, 1999, Omernik, 1987). Ecoregions have been designed to serve as a spatial framework for environmental resource management and denote areas that contain a geographically distinct assemblage of biotic and abiotic phenomena including geology, physiography, vegetation, climate, soils, land use, wildlife, and hydrology. We used the established Land Cover Trends methodology (Loveland and others, 2002) of a sampling approach where randomly selected sample blocks were used to estimate land cover change within the ecoregion (Fig.1). Historical Landsat multi-spectral scanner (MSS), thematic mapper (TM ), and enhanced thematic mapper (ETM+) satellite imagery, along with historical aerial photographs, were used to derive land cover maps for five separate dates (nominally 1973, 1980, 1986, 1992, and 2000). The sample block land cover data were used to analyze the spatial, temporal, and thematic dimensions of change.  The statistical estimates of land cover classes and land cover change come from this standard project methodology.  

Drummond, M.A., Regional dynamics of grassland change in the Western Great Plains: Great Plains Research, (in press).
This paper examines the contemporary land-cover changes in two western Great Plains ecoregions between 1973 and 2000.  Agriculture and other land uses can have a substantial effect on grassland cover that varies regionally depending on the primary driving forces of change.  To understand change, the rates, types and causes of land conversion were examined for 1973, 1980, 1986, 1992 and 2000 using Landsat satellite data and a statistical sampling strategy.  The overall rate of land-cover change between 1973 and 2000 was 7.4% in the Northwestern Great Plains and 11.5% in the Western High Plains.  Trends in both ecoregions have similarities, although the dynamics of change differ temporally depending on driving forces.  Between 1973 and 1986, grassland cover declined when economic opportunity drove an expansion of agriculture.  Between 1986 and 2000, grassland expanded as public policy and a combination of socioeconomic factors drove a conversion from agriculture to grassland.   

Napton, D.E., Auch, R.F., Kurtz, R., and Taylor, J.L., Land changes and their driving forces in the Southeastern United States: (in USGS review).
The forested ecoregions of the Middle Atlantic Coastal Plain, Southeastern Plains, Piedmont, and Blue Ridge comprise eight percent of the conterminous United States and provide a continuum of land cover from the Atlantic Ocean to the highest peaks in the East. Each ecoregion has a unique mosaic of land covers. The Blue Ridge exhibited a stable, forested land cover and provided mountain amenity lands that were used for recreation and second homes. Forestsin the Piedmont declined while the ecoregion’s developed area increased in response to large high population increases. The Southeastern Plains became, arguably, the nation’s most significant commercial forest area as most land cover change in the ecoregion occurred as a result of the forest-harvest-replant cycle of commercial forestry. Additionally, agricultural lands were converted to forest except where there were highly productive soils or innovative entrepreneurs. Forests in theMiddle Atlantic Coastal Plains slowly declined, and development related to recreation and retirement increased along the coast. Across all four ecoregions, the most important driving forces of land conversion were associated with commercial forestry, competition between forest and agriculture, and economic and population growth. These and other drivers of land change were modified by each ecoregion’s unique suitability and land use legacies with the result that the same drivers often produced different land changes in different ecoregions.  

Sohl, T.L., 2007, Atlantic coastal pine barrens: ecoregion report: U.S. Geological Survey Scientific Investigations Report 2007-xxxx, (in review).
In this report we present an assessment of land-use and land-cover change in the Atlantic Coastal Pine Barrens ecoregion for the period 1973-2000.  This work was completed as part of the USGS Land Cover Trends project, a joint effort between the U.S. Geological Survey, the U.S. Environmental Protection Agency, and NASA to study contemporary land cover change in the conterminous United States.   The Atlantic Coastal Pine Barrens is one of 84 Level-III ecoregions defined by the Environmental Protection Agency (Omernik, 1987). Ecoregions have been designed to serve as a spatial framework for environmental resource management and denote areas that contain a geographically distinct assemblage of biotic and abiotic phenomena including geology, physiography, vegetation, climate, soils, land use, wildlife, and hydrology.   A sampling approach using randomly selected sample blocks is used to estimate 60m resolution land cover change in each ecoregion.  Historical Landsat multispectral scanner (MSS) and thematic mapper (TM) satellite images, along with historical aerial photographs, are used to derive land cover maps for five separate dates (1973, 1980, 1986, 1992, and 2000).  The sample block data are used to analyze the spatial, temporal, and sectoral dimensions of change.  A full discussion of project methodology can be found in Loveland et al. (2002).  


Accessibility FOIA Privacy Policies and Notices

Take Pride in America logo logo U.S. Department of the Interior | U.S. Geological Survey
Page Contact Information: WRG Web Team
Page Last Modified: Thursday, 07-Jul-2016 11:19:52 EDT (mg)