USGCRP
Home  Research Elements & Cross-cutting Activities
Land Use / Land Cover Change
Recent Accomplishments |
| Search |
Updated 1 December 2007
|
Land Use /
Land Cover Change |
|
Land Use / Land Cover ChangeRecent Accomplishments Archived News Postings (June 2000 - July 2005)
Additional Past Accomplishments: |
 |
The following are selected highlights of recent research supported by CCSP participating agencies (as reported in the fiscal year 2008 edition of the annual report, Our Changing Planet). Land-Use and Land-Cover Change Drivers Vary by Region and Geopolitical Events.2,3 As part of a national assessment of U.S. land-cover change, the USGS recently completed and web-published an analysis of land-cover and land-use change rates for 20 U.S. ecoregions that span 1,650,930 km2 of the eastern United States. Land-cover extent and changes through time were assessed for ten land-cover types through five time periods: 1973, 1980, 1986, 1992, and 2000. Major findings of this study showed that eastern U.S. land change was connected primarily to timber harvesting and urban growth, together with a decline in agricultural activity. The observed changes were complex and ecoregion specific. The southern United States is thought to be the largest sink for carbon in the conterminous United States. In a long-term assessment of change, researchers reconstructed an annual land cover of the southern United States from 1860 to 2003 with a spatial resolution of 8 km. The pattern of land-cover change in the southern United States was primarily driven by the change in cropland, including cropland expansion and forest regrowth on abandoned cropland. The net flux of carbon due to land-use change in this region was estimated to be zero from 1980 to 2003; but, from 1860 to 1980, land-use and land-cover change was estimated to have resulted in a net release of 9.4 GtC across the southern United States.
Changes in Eastern Europe due to Socioeconomic and Political Factors following the Breakdown of the Soviet Union.4 A comparison of land cover in the Polish, Slovak, and Ukrainian Carpathian Mountains in 2000 from Landsat images showed marked differences in forest cover, dominant forest species, and agricultural fragmentation. These differences can largely be explained by socialist forest management. Post-socialist land-cover change was greatest in Ukraine, where there was high agricultural fragmentation and widespread early-successional shrublands indicating extensive land abandonment. The abundance and pattern of arable land and grassland was attributed to land tenure in socialist times and economic transition since 1990. These results
suggest that broad-scale socioeconomic and political factors are of major significance for land-cover patterns in Eastern Europe.
Land-Cover Change Detection using MODIS Data for Non-Agricultural Areas of the U.S. East Coast.5Moderate Resolution Imaging Spectroradiometer (MODIS) 250 m data are used to form the Normalized Difference Vegetation Index (NDVI), a measure of the photosynthetic capacity of vegetation. Composite NDVI data sets were used to provide automated detection of vegetation change and alarm capability on a 1-year time step for the Albemarle-Pamlico Estuary System region of the U.S. east coast. Vegetation change detection accuracy was assessed for 2002 at 88% with a reasonable balance between change commission errors (22%) and change omission errors (28%). Annual change detection rates across the Albemarle-Pamlico Estuary System over the 2002 to 2005 study period were estimated at 0.7% per annum and varied from 0.4% in 2003 to 0.9% in 2004. Extended regional variations were also readily apparent ranging from 1.6 to 0.1% per annum for the tidal water and mountain ecological zones, respectively. This research included the application of an automated protocol to first filter the MODIS vegetation index data to remove unreliable data values and then estimate the missing data values using a statistical technique to provide high-quality uninterrupted data to support the change detection analysis. The methods and results detailed apply only to non-agricultural areas.Development and Verification of Improved Methods for Remote Assessment of Land-Use Variables Linked to Climate Forcings in Brazil.6 Brazil has become a major producer in world soybean markets, with about 10 million hectares currently planted. Cultivation of this crop is rapidly expanding into two Brazilian ecological zones: the savannah or cerrado, and the forested Amazonian region. The climatic consequences of converting forested Amazonian areas into cropland are significant due to their vastly different storage capacities for carbon. There is a need for accurate updated information on the newly expanded agricultural areas in Brazil and the current total production in order to predict contributions to climate forcings from these regions. A method for assessing crop area and retrieving crop condition parameters that can be used to assess crop yields was developed using data from the MODIS instrument onboard the Terra satellite.
Development and Verification of Improved Methods for Remote Assessment of Land-Use Variables Linked to Climate Forcings in the Mid-West United States.7Crop type, yield, and land management affect the balance of greenhouse gas fluxes from land cover in the mid-western United States. The magnitude of surface changes such as tillage intensity affects residue cover and thus the moisture and radiation energy balances at the land surface through changes in evaporation and reflectance. However, these distinctions are difficult to assess across landscapes. Agricultural Research Service scientists working in Iowa have developed a method using Landsat Thematic Mapper and EO-1 Hyperion imaging spectrometer data to classify tillage intensity in cropland.Development and Verification of Improved Methods for Remote Assessment of Land-Use Variables Linked to Climate Forcings in Central America.8
Regional Climate Change due to Agricultural and Urban Development in California.9In the western United States, large changes in land cover and land use have occurred over the past century with rapidly expanding urbanization along the Pacific coast, and extensive agricultural development inland. A regional climate model was used by researchers to quantify the differences in surface energy fluxes and atmospheric circulation associated with land-cover changes between approximately 1990 and the present. They showed that irrigated agriculture in California lowered mean and maximum surface air temperatures, while conversion of natural vegetation to urban areas increased ground temperatures. This land-use change pattern resulted in changes in the spatial patterns of air pressure and energy balance causing reduced westerly breezes and increased inland breezes. Overall, conversion of natural vegetation to irrigated agriculture has had a larger effect on California’s climate than urban growth up until now, but future projections of increased conversion of irrigated land to urban/suburban development could alter this balance.Western Wildfires, Land-Cover Disturbance and Response to Climate Warming.10,11
National Land Cover Database Available for Use in Climate Models and Assessments.12The USGS, on behalf of the interagency Multi-Resolution Land Characteristics Consortium (MRLC), has made available the National Land Cover Database (NLCD 2001) products for the conterminous United States. These products are available for download from the MRLC web site at <www.mrlc.gov>. NLCD 2001 products include 21 classes of land cover, percent tree canopy, and percent urban imperviousness at 30-m resolution derived from Landsat imagery. NLCD 2001 will support a wide variety of users, institutional sectors, and local- to national-scale applications with this updated land-cover data. This baseline data set is essential in determining the effects of land-cover change on climate as well as the effects of climate change on land cover.Deforestation Dynamics Assessed in Brazilian Amazon.13,14 There has been considerable international attention on the Brazilian Amazon Basin because of the alarming rate of destruction of this high-biodiversity biome and because it is a major terrestrial buffer against rising atmospheric CO2, a primary driver of global warming. Land-cover change from pristine forests to other uses accounts for most of Brazil’s carbon emissions. Investigators combined deforestation maps, field surveys, and satellite-based information on vegetation phenology to characterize the fate of large (>25 hectare) clearings as cropland, cattle pasture, or regrowing forest in the years after initial clearing in Mato Grosso, the state with the highest deforestation rate and soybean production since 2001. Cropland deforestation averaged twice the size of pasture clearings, and more than 90% of the conversion to cropland occurred the first year after deforestation. Economics appears to be a major driver of this change as the area deforested for cropland per year was directly related to the mean annual soybean price. Large-scale logging is relatively new in this region, and impacts of selective logging alone have been estimated to increase the annual carbon release from deforestation in the Amazon by 4 to 7%, thus adding to the CO2 concentration of the atmosphere. Remote-sensing techniques were applied to identify and map areas of selective logging in the tropical upland or terra firme forests of the Brazilian Amazon in 1992, 1996, and 1999. The research results indicate that selective logging is rapidly increasing in both intensity and area. By 1992, at least 6,000 km2 of forest had been logged, and the area affected expanded by an additional 10,000 km2 from 1992 to 1996 and 26,000 km2 from 1996 to 1999. Selective logging within protected areas increased more than two-fold between 1992 and 1996, and more than five-fold between 1996 and 1999 in that region (see Figure 7).
Additional Past Accomplishments:References1) USAID, 2007: The Congo Basin Forests: State of the Forests 2006. Central African Regional Program on the Environment, U.S. AID, Washington, D.C., USA, 256 pp. 2) See eros.usgs.gov/LT/LCCEUS.html. 3) Acevedo, W., J.L. Taylor, D.J. Hester, C.S. Mladinich, and S. Glavac (eds.), 2006: Rates, Trends, Causes, and Consequences of Urban Land-Use Change in the United States. USGS Professional Paper 1726, US Geological Survey, 200 pp. Available at pubs.usgs.gov/pp/pp1726/pp1726.pdf. 4) Tobias, K., V.C. Radeloff, K. Perzanowski, P. Hostert, and K. Perzanowski, 2006: Cross-border comparison of land cover and landscape pattern in Eastern Europe using a hybrid classification technique. Remote Sensing of Environment, 103(4), 449-464. 5) Lunetta, R.S., J.F. Knight, J. Ediriwickrem, J.G. Lyon, and L.D. Worthy, 2006: Land-cover change detection using multi-temporal MODIS NDVI data. Remote Sensing of Environment, 105(2), 142-154. 6) Doraiswamy, P.C., B. Akhmedov, L. Beard, A. Stern, and R. Mueller, 2007: Operational prediction of crop yields using MODIS data and products. International Archives of Photogrammetry, Remote Sensing and Spatial Information Sciences Special Publications (in press). Available at <www.ars.usda.gov/SP2UserFiles/person/ 1430/ISPRS_AGRIFISH_Final.pdf>. 7) Daughtry, C.S.T., P.C. Doraiswamy, E.R. Hunt, A.J. Stern, J.E. McMurtrey, and J.H. Prueger, 2006: Remote sensing of crop residue cover and soil tillage intensity. Soil and Tillage Research, 91, 101-108. 8) Hayes, D.J. and W.B. Cohen, 2007: Spatial, spectral, and temporal patterns of tropical forest cover change as observed with multiple scales of optical satellite data. Remote Sensing of Environment, 106(1), 1-16. 9) Kueppers, L.M., M.A. Snyder, and L.C. Sloan, 2007: Irrigation cooling effect. Geophysical Research Letters, 34(3), L03703, doi:10.1029/2006GL028679. 10) Westerling, A.L, H.G. Hidalgo, D.R. Cayan, and T.W. Swetnam, 2006: Warming and earlier spring increases western U.S. forest wildfire activity. Science, 313, 940-943. 11) Dore, S., M.C. Montes-Helu, B. Sullivan, J.P. Kaye, S.C. Hart, G. Koch, and B. Hungate, 2007: The effect of intense wildfires on ecosystem gas exchange of ponderosa pine forests in northern Arizona. North American Carbon Program Investigators meeting, 22-26 January 2007, Colorado Springs, Colorado. Abstract E.2 p. 45. Available at <www.nacarbon.org/cgi-nacp/2007_meetings/ mtg2007_agenda.pl?meeting_id=1>. 12) See <epa.gov/mrlc/nlcd.html>. 13) Morton, D.C., R.S. DeFries, Y.E. Shimabukuro, L.O. Anderson, E. Arai, F. del Bon Espirito-Santo, R. Freitas, and J. Morisette, 2006: Cropland expansion changes deforestation dynamics in Southern Brazilian Amazon. Proceedings of the National Academy of Sciences, 103(39), 14637-14641. 14) Matricardi, E.A.T., D.L. Skole, M.A. Cochrane, M. Pedlowski, and W.H. Chomentowski, 2007: Multi-temporal assessment of selective logging in the Brazilian Amazon using Landsat data. International Journal of Remote Sensing, 28(1-2), 63-82
|
|
A methodology has been developed for observing changes in tropical forest cover for large areas using data with high temporal frequency from coarse-resolution satellite imagery. Proportional forest cover change is estimated from multi-spectral, multi-temporal MODIS data that are transformed to optimize the spectral detection of vegetation changes. This methodology has been applied using MODIS data in Central America. Landsat data are also used to record higher detail changes of forest cover in Central America. This work describes the distinct patterns of change from year to year due to land-cover changes resulting from forest clearing, regeneration, and changes in climate. It was found that the ability to detect forest cover change patterns using this methodology was relatively independent of the spatial resolution of the data. Associated model simulations indicated the best metrics for detecting tropical forest clearing and regeneration are the shortwave infrared information from the MODIS data at 500-m resolution. Errors were found to range from 7 to 11% across the time periods of analysis.
Wildfire intensity and duration can have a large effect on forest recovery. Scientists from Scripps Institute of Oceanography, the University of California Merced, the USGS, and the University of Arizona compiled a comprehensive database of large wildfires in the western United States since 1970 and compared it with land surface data and hydroclimatic data to show that marked and sudden increases in large wildfire activity in the mid-1980s—including higher frequencies, longer seasons, and longer duration of the fires—were strongly associated with increased spring and summer temperatures and earlier snowmelt. Recent studies in ponderosa pine-dominated forests of northern Arizona have found that intense stand-replacing wildfire converts forests to grasslands and shrublands and converts the land from a CO2 sink to a source, even 10 years after burning, creating a positive feedback to increasing temperatures. Work emerging from CCSP is contributing to the Joint Fire Science Program, which is attempting to improve understanding of climate controls of wildfire (see Figure 6). 
