National Land Cover Dataset Research Articles

Comparison of floodplain surface roughness parameters derived from land cover data and field measurements

Summary Parameterizing surface roughness is a key element in the application of tidal and storm surge inundation models. In this context, surface roughness refers to the ability of the terrain to act as a momentum sink to the overland water flow and also the prevailing winds that help drive this flow. These effects are typically parameterized using estimates of Manning’s n , surface canopy coverage and effective aerodynamic roughness length which vary spatially across the modeling domain as a function of the physical landscape. The current methodology for coastal inundation in the United States assigns these parameters based on published land use/land cover data such as the National Land Cover Dataset. This paper compares those assigned values to values computed based on field measurements at 24 sites in Florida that are representative of land use/land cover classes affected by storm surge. It is shown that while the land use/land cover method is capable of automatically parameterizing surface roughness over a large model domain, parameter prediction errors due to variability within land cover types, misclassification, and parameter value selection for specific land cover classes at the local level are significant which could result in inaccurate estimates of inundation extent and duration.

Evaluating the ability of regional models to predict local avian abundance

AbstractSpatial modeling over broad scales can potentially direct conservation efforts to areas with high species‐specific abundances. We examined the performance of regional models for predicting bird abundance at spatial scales typically addressed in conservation planning. Specifically, we used point count data on wood thrush (Hylocichla mustelina) and blue‐winged warbler (Vermivora cyanoptera) from 2 time periods (1995–1998 and 2006–2007) to evaluate the ability of regional models derived via Bayesian hierarchical techniques to predict bird abundance. We developed models for each species within Bird Conservation Region (BCR) 23 in the upper midwestern United States at 800‐ha, 8,000‐ha, and approximately 80,000‐ha scales. We obtained count data from the Breeding Bird Survey and land cover data from the National Land Cover Dataset (1992). We evaluated predictions from the best models, as defined by an information‐theoretic criterion, using point count data collected within an ecological subregion of BCR 23 at 131 count stations in the 1990s and again in 2006–2007. Competing (Deviance Information Criteria <5) blue‐winged warbler models accounted for 67% of the variability and suggested positive associations with forest edge and proportion of forest at the 8,000‐ha scale, and negative associations with forest patch area (800 ha) and wetness (800 ha and 80,000 ha). The regional model performed best for blue‐winged warbler predicted abundances from point counts conducted in Iowa during 1995–1996 (rs = 0.57; P = 0.14), the survey period that most closely aligned with the time period of data used for regional model construction. Wood thrush models exhibited positive correlations with point count data for all survey areas and years combined (rs = 0.58, P ≤ 0.001). In comparison, blue‐winged warbler models performed worse as time increased between the point count surveys and vintage of the model building data (rs = 0.03, P = 0.92 for Iowa and rs = 0.13, P = 0.51 for all areas, 2006–2007), likely related to the ephemeral nature of their preferred early successional habitat. Species abundance and sensitivity to changing habitat conditions seems to be an important factor in determining the predictive ability of regional models. Hierarchical models can be a useful tool for concentrating efforts at the scale of management units and should be one of many tools used by land managers, but we caution that the utility of such models may decrease over time for species preferring relatively ephemeral habitats if model inputs are not updated accordingly. © 2012 The Wildlife Society.

Impervious Surface Area Change in Arkansas from 2001 to 2006

Impervious Surface Area (ISA) is a measurement used to determine stream quality as well as urban sprawl. ISA was calculated as part of the National Land Cover Dataset (NLCD) using Landsat imagery by the Multi-Resolution Land Characteristics Consortium (MRLC) in both 2001 and 2006. ISA for each of the 75 counties in Arkansas was taken from the NLCD for both 2001 and 2006. Using the ISA data, percent imperviousness was determined for each county in each time period as well as the difference between the two periods. These data were also compared to census projections for the two time periods as well as the difference between them. The differences between percent ISA change and census change were compared to determine consistency.

Large-Scale Hydrologic Modeling of the Michigan and Wisconsin Agricultural Regions to Study Impacts of Land Use Changes

Hydrologic variables in the Great Lakes region have been altered relative to pre-settlement conditions in response to major land use changes during the past 150 years. One of the goals of the present work is to develop a baseline scenario relative to which the impacts of land use changes on hydrological and environmental processes can be evaluated. In addition, this study can help in quantifying the potential impacts of future projected changes in land use in order to mitigate the negative impacts of these changes, especially in regard to a shift toward urbanization and second-generation bioenergy crop production derived from lignocellulosic crops. The present study explores the relationship between land use changes and hydrologic indicators within the agricultural regions of Michigan and Wisconsin. Two sets of land use data, the circa 1800 county base and the 2001 National Land Cover Dataset, were used to set up the Soil and Water Assessment Tool (SWAT) model. First, sensitivity analyses were performed based on both pre-settlement and current land use scenarios. Results showed that parameter sensitivity analysis may not always explain how the variation in model output can be attributed to different sources of variation in the model input. Therefore, effort should be made to determine the true importance of sensitive parameters by considering their placement in model algorithms. The model was then calibrated against measured daily streamflow data obtained from eight U.S. Geological Survey gauging stations. The impacts of land use changes were studied at three scales: subwatershed, watershed, and basin. At the subwatershed scale, most of the hydrologic behavior can be described by percent change in land cover. However, the trend was more apparent for land use conversion from mixed forest to urban and agriculture lands than for other land use conversions. At the watershed scale, significant differences were observed based on the long-term average hydrologic variables under the current and pre-settlement scenarios. In addition, an increase in evapotranspiration (up to 16.5%) and surface runoff (up to 93.9%) contribution to streamflow, a decrease in recharge to aquifers (up to -51.5%) and baseflow (up to -50.1%), and mixed impacts on water yield (-21.5% to 24.6%) were detected. However, at the basin scale, more than 70% of the study area experienced decreased lateral subsurface flow and recharge to aquifers, while 65% of the area experienced increased overland flow and minor changes in evapotranspiration and water yield.

Modeling broad-scale patterns of avian species richness across the Midwestern United States with measures of satellite image texture

Avian biodiversity is threatened, and in order to prioritize limited conservation resources and conduct effective conservation planning a better understanding of avian species richness patterns is needed. The use of image texture measures, as a proxy for the spatial structure of land cover and vegetation, has proven useful in explaining patterns of avian abundance and species richness. However, prior studies that modeled habitat with texture measures were conducted over small geographical extents and typically focused on a single habitat type. Our goal was to evaluate the performance of texture measures over broad spatial extents and across multiple habitat types with varying levels of vertical habitat structure. We calculated a suite of texture measures from 114 Landsat images over a study area of 1,498,000km2 in the Midwestern United States, which included habitats ranging from grassland to forest. Avian species richness was modeled for several functional guilds as a function of image texture. We subsequently compared the explanatory power of texture-only models with models fitted using landscape composition metrics derived from the National Land Cover Dataset, as well as models fitted using both texture and composition metrics. Measures of image texture were effective in modeling spatial patterns of avian species richness in multiple habitat types, explaining up to 51% of the variability in species richness of permanent resident birds. In comparison, landscape composition metrics explained up to 56% of the variability in permanent resident species richness. In the most heavily forested ecoregion, texture-measures outperformed landscape metrics, and the two types of measurements were complementary in multivariate models. However, in two out of three ecoregions examined, landscape composition metrics consistently performed slightly better than texture measures, and the variance explained by the two types of measures overlapped considerably. These results show that image texture measures derived from satellite imagery can be an important tool for modeling patterns of avian species richness at broad spatial extents, and thus assist in conservation planning. However, texture measures were slightly inferior to landscape composition metrics in about three-fourths of our models. Therefore texture measures are best considered in conjunction with landscape metrics (if available) and are best used when they show explanatory ability that is complementarity to landscape metrics.

Estimating Wetland Losses and Gains in Coastal North Carolina: 1994–2001

Wetland resources have continued to decline in coastal regions of the United States despite passage of the 1972 Clean Water Act, creation of federal and state wetland permitting programs, and declaration of the goal of “no net loss” of wetland resources. Wetland loss rates were assessed across 20 counties in coastal North Carolina from 1994 to 2001 using 1994 wetland inventory data from the North Carolina Division of Coastal Management (NCDCM) and the 2001 National Land Cover Dataset. Accuracy assessment results were used to generate adjusted estimates of wetland loss, and the results were compared to restoration data from the NCDCM. Total wetland loss from 1994 to 2001 was 25,303 ha, or 1.95% of the 1994 wetland area. Of the total, 17,858 ha converted to upland and 7,445 ha converted to open water. These results are somewhat higher than published loss rates for Atlantic coastal watersheds and for the United States as a whole. By contrast, agency records suggest that a maximum of 4,591 ha of wetland restoration and 68 ha of wetland creation were begun in the 20 North Carolina coastal counties from 1994 to 2001.

Capabilities and limitations of Landsat and land cover data for aboveground woody biomass estimation of Uganda

Aboveground woody biomass for circa-2000 is mapped at national scale in Uganda at 30-m spatial resolution on the basis of Landsat ETM + images, a National land cover dataset and field data using an object-oriented approach. A regression tree-based model (Random Forest) produces good results (cross-validated R² 0.81, RMSE 13 T/ha) when trained with a sufficient number of field plots representative of the vegetation variability at national scale. The Random Forest model captures non-linear relationships between satellite data and biomass density, and is able to use categorical data (land cover) in the regression to improve the results. Biomass estimates were strongly correlated (r = 0.90 and r = 0.83) with independent LiDAR measurements. In this study, we demonstrate that in certain contexts Landsat data provide the capability to spatialize field biomass measurements and produce accurate and detailed estimates of biomass distribution at national scale. We also investigate limitations of this approach, which tend to provide conservative biomass estimates. Specific limitations are mainly related to saturation of the optical signal at high biomass density and cloud cover, which hinders the compilation of a radiometrically consistent multi-temporal dataset. As a result, a Landsat mosaic created for Uganda with images acquired in the dry season during 1999–2003 does not contain phenological information useful for discriminating some vegetation types, such as deciduous formations. The addition of land cover data increases the model performance because it provides information on vegetation phenology. We note that Landsat data present higher spatial and thematic resolution compared to land cover and allow detailed and spatially continuous biomass estimates to be mapped. Fusion of satellite and ancillary data may improve biomass predictions but, to avoid error propagation, accurate, detailed and up-to-date land cover or other ancillary data are necessary.

Mapping biomass with remote sensing: a comparison of methods for the case study of Uganda

BackgroundAssessing biomass is gaining increasing interest mainly for bioenergy, climate change research and mitigation activities, such as reducing emissions from deforestation and forest degradation and the role of conservation, sustainable management of forests and enhancement of forest carbon stocks in developing countries (REDD+). In response to these needs, a number of biomass/carbon maps have been recently produced using different approaches but the lack of comparable reference data limits their proper validation. The objectives of this study are to compare the available maps for Uganda and to understand the sources of variability in the estimation. Uganda was chosen as a case-study because it presents a reliable national biomass reference dataset.ResultsThe comparison of the biomass/carbon maps show strong disagreement between the products, with estimates of total aboveground biomass of Uganda ranging from 343 to 2201 Tg and different spatial distribution patterns. Compared to the reference map based on country-specific field data and a national Land Cover (LC) dataset (estimating 468 Tg), maps based on biome-average biomass values, such as the Intergovernmental Panel on Climate Change (IPCC) default values, and global LC datasets tend to strongly overestimate biomass availability of Uganda (ranging from 578 to 2201 Tg), while maps based on satellite data and regression models provide conservative estimates (ranging from 343 to 443 Tg). The comparison of the maps predictions with field data, upscaled to map resolution using LC data, is in accordance with the above findings. This study also demonstrates that the biomass estimates are primarily driven by the biomass reference data while the type of spatial maps used for their stratification has a smaller, but not negligible, impact. The differences in format, resolution and biomass definition used by the maps, as well as the fact that some datasets are not independent from the reference data to which they are compared, are considered in the interpretation of the results.ConclusionsThe strong disagreement between existing products and the large impact of biomass reference data on the estimates indicate that the first, critical step to improve the accuracy of the biomass maps consists of the collection of accurate biomass field data for all relevant vegetation types. However, detailed and accurate spatial datasets are crucial to obtain accurate estimates at specific locations.

Urban growth and aerosol effects on convection over Houston. Part II: Dependence of aerosol effects on instability

Our studies of the urban growth and aerosol effects on convection over Houston were performed in two phases. The first phase (Part I) used a sea-breeze-induced convective case (August 24 2000) as a benchmark for a series of sensitivity experiments varying the city size and the intensity of aerosol sources linked to sub-grid urban area fractions. The use of National Land Cover Dataset (NLCD) for three different years allowed an objective way to design the land-use sensitivity experiments as well as to consider the spatial distribution and differential intensities of the urban aerosol sources. This paper extends Part I to explore how the relative intensification of the cells downwind of the city (due to urban aerosol sources) can change for environments with more or less instability. This second series of numerical experiments consists of a fairly large number of multi-grid simulations (almost 100) that varied not only the intensity of the urban sources but also the value of CAPE. Our results were in agreement with previous studies; higher cloud condensation nuclei (CCN) concentrations reduced the sizes of cloud droplets and increased their probability of reaching supercooled levels and thus reducing coalescence. Therefore, downwind convective cells were intensified by additional release of latent heat of freezing. However, our results show a non-monotonic behavior on surface precipitation. This interesting response was linked to the riming efficiency reduction of ice particles when aerosol concentrations are greatly enhanced. Therefore, a greater fraction of the ice-phase condensed water mass is transported out of the storm as pristine ice crystals instead of being transferred to precipitating water species. Moreover, the precipitation efficiency of cells downwind of the city exhibited a similar behavior, increasing when CCN concentrations are initially increased and then decreasing when aerosol concentrations are further enhanced. The CCN concentration needed to reach maximum precipitation efficiency is higher for more unstable environments.

Evaluations of land cover risk factors for canine leptospirosis: 94 cases (2002–2009)

Associations of land cover/land use variables and the presence of dogs in urban vs. rural address locations were evaluated retrospectively as potential risk factors for canine leptospirosis in Kansas and Nebraska using Geographic Information Systems (GIS). The sample included 94 dogs positive for leptospirosis predominantly based on a positive polymerase chain reaction test for leptospires in urine, isolation of leptospires on urine culture, a single reciprocal serum titer of 12,800 or greater, or a four-fold rise in reciprocal serum titers over a 2–4 weeks period; and 185 dogs negative for leptospirosis based on a negative polymerase chain reaction test and reciprocal serum titers less than 400. Land cover features from 2001 National Land Cover Dataset and 2001 Kansas Gap Analysis Program datasets around geocoded addresses of case/control locations were extracted using 2500m buffers, and the presence of dogs’ address locations within urban vs. rural areas were estimated in GIS. Multivariate logistic models were used to determine the risk of different land cover variables and address locations to dogs. Medium intensity urban areas (OR=1.805, 95% C.I.=1.396, 2.334), urban areas in general (OR=2.021, 95% C.I.=1.360, 3.003), and having urban address locations (OR=3.732, 95% C.I.=1.935, 7.196 entire study region), were significant risk factors for canine leptospirosis. Dogs regardless of age, sex and breed that live in urban areas are at higher risk of leptospirosis and vaccination should be considered.

Weather and Land Cover Influences on Mosquito Populations in Sioux Falls, South Dakota

This study compared the spatial and temporal patterns of Culex tarsalis Coquillett and Aedes vexans Meigen populations and examined their relationships with land cover types and climatic variability in Sioux Falls, SD. Between 24 and 30 CDC CO2-baited light traps were set annually in Sioux Falls from May to September 2005-2008. Land cover data were acquired from the 2001 National Land Cover Dataset and the percentages of selected land cover types were calculated within a 600-m buffer zone around each trap. Meteorological information was summarized from local weather stations. Cx. tarsalis exhibited stronger spatial autocorrelation than Ae. vexans. Land cover analysis indicated that Cx. tarsalis was positively correlated with grass/hay, and Ae. vexans was positively correlated with wetlands. No associations were identified between irrigation and the host-seeking population of each species. Higher temperature in the current week and 2 wk prior and higher precipitation 3-4 wk before collection of host-seeking adult mosquitoes had positive influences on Cx. tarsalis abundance. Temperature in the current week and rainfall 2-3 wk before sampling had positive influences on Ae. vexans abundance. This study revealed the different influences of weather and land cover on important mosquito species in the Northern Great Plains region, which can be used to improve local vector control strategies and West Nile virus prevention efforts.

Land use and climate influences on waterbirds in the Prairie Potholes

Aim We examined the influences of regional climate and land-use variables on mallard (Anas platyrhynchos), blue-winged teal (Anas discors), ruddy duck (Oxyura jamaicensis) and pied-billed grebe (Podilymbus podiceps) abundances to inform conservation planning in the Prairie Pothole Region of the United States. Location The US portion of Bird Conservation Region 11 (US-BCR11, the Prairie Potholes), which encompasses six states within the United States: Montana, North Dakota, South Dakota, Nebraska, Minnesota and Iowa. Methods We used data from the North American Breeding Bird Survey (NABBS), the National Land Cover Data Set, and the National Climatic Data Center to model the effects of environmental variables on waterbird abundance. We evaluated land-use covariates at three logarithmically related spatial scales (1000, 10,000 and 100,000 ha), and constructed hierarchical spatial count models a priori using information from published habitat associations. Model fitting was performed using a hierarchical modelling approach within a Bayesian framework. Results Models with the same variables expressed at different scales were often in the best model subset, indicating that the influence of spatial scale was small. Both land-use and climate variables contributed strongly to predicting waterbird abundance in US-BCR11. The strongest positive influences on waterbird abundance were the percentage of wetland area across all three spatial scales, herbaceous vegetation and precipitation variables. Other variables that we included in our models did not appear to influence waterbirds in this study. Main conclusions Understanding the relationships of waterbird abundance to climate and land use may allow us to make predictions of future distribution and abundance as environmental factors change. Additionally, results from this study can suggest locations where conservation and management efforts should be focused.

Effect of land cover change on runoff curve number estimation in Iowa, 1832–2001

AbstractWithin the first few decades of European‐descended settlers arriving in Iowa, much of the land cover across the state was transformed from prairie and forest to farmland, patches of forest, and urbanized areas. Land cover change over the subsequent 126 years was minor in comparison. Between 1832 and 1859, the General Land Office conducted a survey of the State of Iowa to aid in the disbursement of land. In 1875, an illustrated atlas of the State of Iowa was published, and in 2001, the US Geological Survey National Land Cover Dataset was compiled. Using these three data resources for classifying land cover, the hydrologic impact of the land cover change at three points in time over a period of 132+ years is presented in terms of the effect on the area‐weighted average curve number, a term commonly used to predict peak runoff from rainstorms. In the four watersheds studied, the area‐weighted average curve number associated with the first 30 years of settlement increased from 61·4 to 77·8. State‐wide mapped forest area over this same period decreased 19%. Over the next 126 years, the area‐weighted average curve number decreased to 76·7, despite an additional forest area reduction of 60%. This suggests that degradation of aquatic resources (plants, fish, invertebrates, and habitat) arising from hydrologic alteration was likely to have been much higher during the 30 years of initial settlement than in the subsequent period of 126 years in which land cover changes resulted primarily from deforestation and urbanization. Published 2010. This article is a US Government work and is in the public domain in the USA.

Pollution Sources and Mortality Rates Across Rural-Urban Areas in the United States

To conduct an assessment of rural environmental pollution sources and associated population mortality rates. The design is a secondary analysis of county-level data from the Environmental Protection Agency (EPA), Department of Agriculture, National Land Cover Dataset, Energy Information Administration, Centers for Disease Control and Prevention, the US Census, and others. We described the types of pollution sources present in metropolitan and nonmetropolitan counties and examined the associations between these sources and rates of all-cause, cardiovascular, respiratory, and cancer mortality while controlling for age, race, and other covariates. Rural counties had 65,055 EPA-monitored pollution discharge sites. As expected, rural counties had significantly greater exposure to potential agriculture-related pollution. Regression models specific to rural counties indicated that greater density of water pollution sources was significantly associated with greater total and cancer mortality. Rural air pollution sources were associated with greater cancer mortality rates. Rural coal mining areas had higher total, cancer, and respiratory disease mortality rates. Agricultural production was generally associated with lower mortality rates. Greater levels of human development were significantly related to higher adjusted total and cancer mortality. The association between pollution sources and mortality risk is not a phenomenon limited to metropolitan areas. Results carry policy implications regarding the need for effective environmental standards and monitoring. Further research is needed to better understand the types and distributions of pollution in rural areas, and the health consequences that result.

Methodologic Issues in Using Land Cover Data to Characterize Living Environments of Geocoded Addresses

Methodologic Issues in Using Land Cover Data to Characterize Living Environments of Geocoded Addresses

Development of an object-based framework for classifying and inventorying human-dominated forest ecosystems

This paper presents the development of a framework for classifying and inventorying Eastern US forestland based on the level of anthropogenic disturbance and fragmentation using high spatial resolution remote sensing data and a multiscale object-based classification system. We implemented the framework using a suburban area in Baltimore County, Maryland, USA as a case study. We developed a three-level hierarchical scheme of image objects. The object-based, multiscale classification and inventory framework provides an effective and flexible way of showing different mixes of human development and forest cover in a hierarchical fashion for human-dominated forest ecosystems. At the finest scale (level 1), the classification nomenclature describes basic land cover feature types, which are divided up into trees and individual features that fragment forests. The overall accuracy of the classification was 91.25%. At level 2, forest patches were delineated and classified into different categories based on the degree of human disturbance. At level 3, major roads were used to segment the study area into larger objects, which were classified on the basis of relative composition and spatial arrangement of forests and fragmenting features. This study provides decision makers, planners and the public with a new methodological framework that can be used to more precisely classify and inventory forest cover. The comparisons of the estimates of forest cover from our analyses with those from the 2001 National Land Cover Dataset (NLCD) show that aggregated figures of forest cover are misleading and that much of what is mapped as forest is highly degraded and is more suburban than natural in its land use.

From the redwood forest to the Gulf Stream waters: human signature nearly ubiquitous in representative US landscapes

What landscapes best represent the land uses and land covers (LU/LC) of the continental United States? Would the set include a cornfield? A forest? A backyard? Combining principles of landscape ecology and computer science, we identified a small set of “exemplar landscapes”, representing distinct LU/LC pattern types of the conterminous US. We first partitioned the 1992 US National Land Cover Dataset into 193 705 landscapes, and quantified patterns with standard measures of LU/LC composition and configuration. Using the values to estimate similarity of LU/LC patterns between landscapes, we applied an algorithm developed to find representatives in large sets. In the resulting 17‐member set of exemplar landscapes, patterns created and managed by human activity are by far the most evident features. This set of representatives summarizes the nation's LU/LC, demonstrating the degree to which human‐influenced patterns dominate: aggregations of rectangular fields, farmlands within cleared forests, shrublands/pasture, and suburbs. The algorithm's selection of an exemplar for each group may have other ecological applications when an objectively determined subset of representative items is needed.

Using NHDPlus as the Land Base for the Noah‐distributed Model

AbstractThe National Elevation, Hydrography and Land Cover datasets of the United States have been synthesized into a geospatial dataset called NHDPlus which is referenced to a spheroidal Earth, provides geospatial data layers for topography on 30 m rasters, and has vector coverages for catchments and river reaches. In this article, we examine the integration of NHDPlus with the Noah‐distributed model. In order to retain compatibility with atmospheric models, Noah‐distributed utilizes surface domain fields referenced to a spherical rather than spheroidal Earth in its computation of vertical land surface/atmosphere water and energy budgets (at coarse resolution) as well as horizontal cell‐to‐cell water routing across the land surface and through the shallow subsurface (at fine resolution). Two data‐centric issues affecting the linkage between Noah‐distributed and NHDPlus are examined: (1) the shape of the Earth; and (2) the linking of gridded landscape with a vector representation of the stream and river network. At mid‐latitudes the errors due to projections between spherical and spheroidal representations of the Earth are significant. A catchment‐based “pour point” technique is developed to link the raster and vector data to provide lateral inflow from the landscape to a one‐dimensional river model. We conclude that, when Noah‐distributed is run uncoupled to an atmospheric model, it is advantageous to implement Noah‐distributed at the native spatial scale of the digital elevation data and the spheroidal Earth of the NHDPlus dataset rather than transforming the NHDPlus dataset to fit the coarser resolution and spherical Earth shape of the Noah‐distributed model.

Automated Urban Delineation from Landsat Imagery Based on Spatial Information Processing

Methodologies are developed and assessed for the delineation of urban areas on moderate resolution satellite images based on line feature extraction and processing. It is argued that a simple two component urban/rural road network model and a preliminary estimate of the extent of urban cover in a satellite scene can be used to quantify processing settings, such as line thresholds. As a result, urban delineation can be fully automated. Tests have been conducted with Landsat Thematic Mapper data of a candidate set of cities that exemplify the spectrum of North American urban landscapes. Producer accuracies in the range 60 to 80 percent are consistently for a variety of urban landscapes. While these performance levels are lower than those previously reported for small area, city-specific studies, they are comparable to those previously achieved in large-area, approaches (e.g., the National Land Cover Dataset initiative) that rely primarily on spectral attributes. It is concluded that better integrated use of spectral-spatial data processing have the potential to lead to improved operational urban mapping accuracies.

Quantifying scaling effects on satellite‐derived forest area estimates for the conterminous USA

We quantified the scaling effects on forest area estimates for the conterminous USA using regression analysis and the National Land Cover Dataset 30 m satellite‐derived maps in 2001 and 1992. The original data were aggregated to: (1) broad cover types (forest vs. non‐forest); and (2) coarser resolutions (1 km and 10 km). Standard errors of the model estimates were 2.3% and 4.9% at 1 km and 10 km resolutions, respectively. Our model improved the accuracies for 1 km by 0.6% (12 556 km2) in 2001 and 1.9% (43 198 km2) in 1992, compared to the forest estimates before the adjustments. Forest area observed from Moderate Resolution Imaging Spectroradiometer (MODIS) 2001 1 km land‐cover map for the conterminous USA might differ by 80 811 km2 from what would be observed if MODIS was available at 30 m. Of this difference, 58% (46 870 km2) could be a relatively small net improvement, equivalent to 1444 Tg (or 1.5%) of total non‐soil forest CO2 stocks. With increasing attention to accurate monitoring and evaluation of forest area changes for different regions of the globe, our results could facilitate the removal of bias from large‐scale estimates based on remote sensors with coarse resolutions.