William Hargrove | US Forest Service (original) (raw)
Papers by William Hargrove
Choice Reviews Online
... Framework 12 2.4 Map of Major Habitat Types of the United States and Canada 13 2.5 Component ... more ... Framework 12 2.4 Map of Major Habitat Types of the United States and Canada 13 2.5 Component Criteria and Weighting Used in the Biological Distinctiveness Index 15 2.6 Map of Bioregions of North America 18 2.7 Integration Matrix for Conservation Status and Biological ...
, IL, determined that a "Corridor Tool" developed by the Oak Ridge National Laboratory (ORNL) cou... more , IL, determined that a "Corridor Tool" developed by the Oak Ridge National Laboratory (ORNL) could serve to further focus ongoing "habitat" research on habitat fragmentation at the landscape scale. This work tested the ORNL Corridor Tool on data related to Red-cockaded Woodpecker habitat fragmentation in the southeastern United States using widely available data to run the Corridor Tool, and to develop a general corridor analysis. DISCLAIMER: The contents of this report are not to be used for advertising, publication, or promotional purposes. Citation of trade names does not constitute an official endorsement or approval of the use of such commercial products. All product names and trademarks cited are the property of their respective owners. The findings of this report are not to be construed as an official Department of the Army position unless so designated by other authorized documents. DESTROY THIS REPORT WHEN IT IS NO LONGER NEEDED. DO NOT RETURN IT TO THE ORIGINATOR.
Military installation training lands must be managed to support species at risk as well as to be ... more Military installation training lands must be managed to support species at risk as well as to be effective training environments for soldiers. Forecasts from various global climate change models suggest that the habitats associated with some military training installations will face pressures that induce biome-shifts, invasive species, loss of habitat, and changes in training opportunities. This study combined worldwide habitat forecast data with a current habitat map to identify major installations that appear to be most and least at-risk for habitat change. DISCLAIMER: The contents of this report are not to be used for advertising, publication, or promotional purposes. Citation of trade names does not constitute an official endorsement or approval of the use of such commercial products. All product names and trademarks cited are the property of their respective owners. The findings of this report are not to be construed as an official Department of the Army position unless so designated by other authorized documents.
Annals of the Entomological Society of America
An IBM PC or compatible microcomputer equipped with a Tekmatic Systems Video Van Gogh digitizer c... more An IBM PC or compatible microcomputer equipped with a Tekmatic Systems Video Van Gogh digitizer card can be interfaced with a standard video camera for rapidly measuring the percentage of leaf area lost (LAL) to herbivorous insects. This capability can be added to an existing color PC for about $600 in additional hardware. The system digitizes a video camera image of a reconstructed leaf and displays the amount of LAL in about 20 s. The video digitizer simultaneously measures area removed from, and area remaining in, the leaf without tracing of these areas by the operator. Minimized operator interaction greatly increases the number of leaves that can be measured. Before digitizing, a photocopied image of the leaf must have its outer perimeter completed with a dark pencil line if it has been interrupted by folivore feeding. The video signal is digitally encoded into a 256 x 256 element false-color picture of the reconstructed damaged leaf on the computer monitor. A keyboard-adjustable brightness threshold distinguishes between leaf and background. Digitized measurements are saved to a sequential ASCII data file compatible with most popular database, spreadsheet, and word processor programs. The video digitizer software allows two video camera work stations to be supported by a single computer. Software for the video digitizer is released into the public domain and can be obtained from the authors for the cost of a diskette and postage.
Conservation Ecology, 2002
Ecol Lett, 2004
All global circulation models based on Intergovernmental Panel on Climate Change (IPCC) scenarios... more All global circulation models based on Intergovernmental Panel on Climate Change (IPCC) scenarios project profound changes, but there is no consensus on how to map their environmental consequences. Our multivariate representation of environmental space combines stable topographic and edaphic attributes with dynamic climatic attributes. We divide that environmental space into 500 unique domains and map their current locations and their projected locations in 2100 under contrasting emissions scenarios. The environmental domains found across half the study area today disappear under the higher emissions scenario, but persist somewhere in it under the lower emissions scenario. Locations affected least and those affected most under each scenario are mapped. This provides an explicit framework for designing conservation networks to include both areas at least risk (potential refugia) and areas at greatest risk, where novel communities may form and where sentinel ecosystems can be monitored for signs of stress.
Quart Rev Biol, 2001
ABSTRACT The National Early Warning System (EWS) provides an 8-day coast-to-coast snapshot of pot... more ABSTRACT The National Early Warning System (EWS) provides an 8-day coast-to-coast snapshot of potentially disturbed forests across the U.S.. A prototype system has produced national maps of potential forest disturbances every eight days since January 2010, identifying locations that may require further investigation. Through phenology, the system shows both early and delayed vegetation development and detects all types of unexpected forest disturbances, including insects, disease, wildfires, frost and ice damage, tornadoes, hurricanes, blowdowns, harvest, urbanization, landslides, drought, flood, and climate change. The USDA Forest Service Eastern Forest Environmental Threat Assessment Center is collaborating with NASA Stennis Space Center and the Western Wildland Environmental Threat Assessment Center to develop the tool. The EWS uses differences in phenological responses between an expectation based on historical data and a current view to strategically identify potential forest disturbances and direct attention to locations where forest behavior seems unusual. Disturbance maps are available via the Forest Change Assessment Viewer (FCAV) (http://ews.forestthreats.org/gis), which allows resource managers and other users to see the most current national disturbance maps as soon as they are available. Phenology-based detections show not only vegetation disturbances in the classical sense, but all departures from normal seasonal vegetation behavior. In 2010, the EWS detected a repeated late-frost event at high elevations in North Carolina, USA, that resulted in delayed seasonal development, contrasting with an early spring development at lower elevations, all within close geographic proximity. Throughout 2011, there was a high degree of correspondence between the National Climatic Data Center's North American Drought Monitor maps and EWS maps of phenological drought disturbance in forests. Urban forests showed earlier and more severe phenological drought disturbance than surrounding non-urban forests. An EWS news page (http://www.geobabbble.org/~hnw/EWSNews) highlights disturbances the system has detected during the 2011 season. Unsupervised statistical multivariate clustering of smoothed phenology data every 8 days over an 11-year period produces a detailed map of national vegetation types, including major disturbances. Examining the constancy of these phenological classifications at a particular location from year to year produces a national map showing the persistence of vegetation, regardless of vegetation type. Using spectral unmixing methods, national maps of evergreen decline can be produced which are a composite of insect, disease, and anthropogenic factors causing chronic decline in these forests, including hemlock wooly adelgid, mountain pine beetle, wildfire, tree harvest, and urbanization. Because phenology shows vegetation responses, all disturbance and recovery events detected by the EWS are viewed through the lens of the vegetation.
ABSTRACT Background/Question/Methods Up-scaling from sparse point based measurement to continuous... more ABSTRACT Background/Question/Methods Up-scaling from sparse point based measurement to continuous gridded product is a common problem in Earth System Science. We present a new general-purpose empirical imputation method based on associative clustering, which associates sparse measurements of dependent variables with particular multivariate clustered combinations of the independent variables, and then estimates values for unmeasured clusters, based on directional proximity in multidimensional data space, at both the cluster and map cell levels of resolution. A multivariate cluster analysis was applied to global output from a General Circulation Model (GCM) consisting of 17 variables downscaled to 4 km2 resolution. Present global growing conditions were divided into 30 thousand relatively homogeneous ecoregions describing climatic and topographic conditions. At every mapcell a multi-linear regression was applied in 17 dimensional hyperspace to derive the suitability of a tree species where not measured using the forest inventory data. The continuous species distribution maps obtained were compared and validated against existing tree range suitability maps. Associative Clustering is intended to be a general-purpose imputation tool, is model-free, and can be used to derive tree growth for future conditions that have no present-day analog. Results/Conclusions We demonstrate this new imputation tool on tree species range distribution maps, which describe the suitable extent and expected growth performance of a particular tree species over a wide area. Range maps having continuous estimates of tree growth performance are more useful than more classical tree range maps that simply show binary occurence suitability. The USDA Forest Service Forest Inventory Assessment (FIA) plots provide information about the occurence and growth performance for various tree species across the US, but such measurements are limited to FIA plots. Using Associative Clustering, we scale up the discontinuous FIA Inventory growth measurements into continuous maps that show the expected growth and suitabilty for individual tree species covering the Continental United States.
ABSTRACT Background/Question/Methods Range shifts for 215 tree species were predicted globally fo... more ABSTRACT Background/Question/Methods Range shifts for 215 tree species were predicted globally following climate changes forecast by the Parallel Climate Model (PCM) and the Hadley climate model under IPCC scenarios A2 and B2 for the years 2050 and 2100. Results/Conclusions Most species' predicted suitable (or fundamental) ranges closely followed or were slightly more extensive than Little's actual (or realized) ranges under present conditions. Maps of Minimum Required Movement (MRM) to return to the closest geographic locations offering suitable conditions in the future show the likelihood of local extinction following climate change. Locations that are the nearest "lifeboats" for large surrounding areas may represent management and conservation targets.
J Environ Manage, 2008
Habitat valuation methods are most often developed and used to prioritize candidate lands for con... more Habitat valuation methods are most often developed and used to prioritize candidate lands for conservation. In this study the intent of habitat valuation was to inform the decision-making process for remediation of chemical contaminants on specific lands or surface water bodies. Methods were developed to summarize dimensions of habitat value for six representative aquatic and terrestrial contaminated sites at the East Tennessee Technology Park (ETTP) on the US Department of Energy Oak Ridge Reservation in Oak Ridge, TN, USA. Several general valuation metrics were developed for three broad categories: site use by groups of organisms, site rarity, and use value added from spatial context. Examples of use value metrics are taxa richness, a direct measure of number of species that inhabit an area, complexity of habitat structure, an indirect measure of potential number of species that may use the area, and land use designation, a measure of the length of time that the area will be available for use. Measures of rarity included presence of rare species or communities. Examples of metrics for habitat use value added from spatial context included similarity or complementarity of neighboring habitat patches and presence of habitat corridors. More specific metrics were developed for groups of organisms in contaminated streams, ponds, and terrestrial ecosystems. For each of these metrics, cutoff values for high, medium, and low habitat value were suggested, based on available information on distributions of organisms and landscape features, as well as habitat use information. A companion paper describes the implementation of these habitat valuation metrics and scoring criteria in the remedial investigation for ETTP.
Agu Fall Meeting Abstracts, Dec 1, 2004
ABSTRACT We propose and implement a cluster-based approach for identifying global phenological ob... more ABSTRACT We propose and implement a cluster-based approach for identifying global phenological observatories in which phenologically and climatologically self-similar pixel clusters are monitored. We developed clusters based on a wavelet-filtered subset of the 1982-1999 global Pathfinder Advanced Very High Resolution Radiometer (AVHRR) Land (PAL) Normalized Difference Vegetation Index (NDVI) dataset, a global 10-minute resolution climatology, and the clustering approach developed by the Oak Ridge National Laboratories (ORNL). In the ORNL approach: (1) n cluster centers are defined based on the multi-dimensional NDVI/climate space; (2) pixel distances from the centroids are calculated; (3) pixels are assigned to the minimum distance cluster. While any number of clusters may be specified, we found that a global 500-cluster approach provided a satisfactory global distribution. In traditional rectangular approaches a group of pixels could contain desert, grassland, and tropical forest. Here, longitudinally extensive but latitudinally limited regions such as the Sahel exist as distinct groups. Thus, our approach avoids problems affecting single-pixel approaches (misregistration, cloud contamination) and rectangular approaches (mixed phenological signals). Using the 1982-2003 GIMMS AVHRR dataset, we extracted phenological metrics such as the onset and offset of greenness for each cluster. We then ranked each cluster based on land cover homogeneity, evidence of human impacts, and political diversity. For each biome, we then identified the highest ranked clusters within four climate zones (hot/wet, hot/dry, cold/wet, cold/dry). This strategy provides: (1) selection of regions for which a strong annual is detectable, (2) a method of identifying regions least likely to be impacted by non-climatic factors, and (3) a strategy for ground validation.
Photogrammetric Engineering and Remote Sensing, 2009
We are using a statistical clustering method for delineating homogeneous ecoregions as a basis fo... more We are using a statistical clustering method for delineating homogeneous ecoregions as a basis for identifying disturbances in forests through time over large areas, up to national and global extents. Such changes can be shown relative to past conditions, or can be predicted relative to present conditions, as with forecasts of future climatic change. This quantitative ecoregion approach can be used to predict destinations for populations whose local environments are forecast to become unsuitable and are forced to migrate as their habitat shifts, and is also useful for predicting the susceptibility of new locations to invasive species like Sudden Oak Death. EFETAC and our sister western center WWETAC, along with our NASA and ORNL collaborators, are designing a new national-scale early warning system for forest threats, called FIRST. Envisioned as a change-detection system, FIRST will identify all land surface cover changes at the MODIS observational scale, and then try to discriminate normal, expected seasonal changes from locations having unusual activity that may represent potential forest threats. As a start, we have developed new national data sets every 16 days from 2002 through 2008, based on land surface phenology, or timing of leaf-out in the spring and brown-down in the fall. Changes in such phenological maps will be shown to contain important information about vegetation health status across the United States. The standard deviation of the duration of fall can be mapped, showing places where length of fall is relatively constant or is variable in length from year to year.
Choice Reviews Online
... Framework 12 2.4 Map of Major Habitat Types of the United States and Canada 13 2.5 Component ... more ... Framework 12 2.4 Map of Major Habitat Types of the United States and Canada 13 2.5 Component Criteria and Weighting Used in the Biological Distinctiveness Index 15 2.6 Map of Bioregions of North America 18 2.7 Integration Matrix for Conservation Status and Biological ...
, IL, determined that a "Corridor Tool" developed by the Oak Ridge National Laboratory (ORNL) cou... more , IL, determined that a "Corridor Tool" developed by the Oak Ridge National Laboratory (ORNL) could serve to further focus ongoing "habitat" research on habitat fragmentation at the landscape scale. This work tested the ORNL Corridor Tool on data related to Red-cockaded Woodpecker habitat fragmentation in the southeastern United States using widely available data to run the Corridor Tool, and to develop a general corridor analysis. DISCLAIMER: The contents of this report are not to be used for advertising, publication, or promotional purposes. Citation of trade names does not constitute an official endorsement or approval of the use of such commercial products. All product names and trademarks cited are the property of their respective owners. The findings of this report are not to be construed as an official Department of the Army position unless so designated by other authorized documents. DESTROY THIS REPORT WHEN IT IS NO LONGER NEEDED. DO NOT RETURN IT TO THE ORIGINATOR.
Military installation training lands must be managed to support species at risk as well as to be ... more Military installation training lands must be managed to support species at risk as well as to be effective training environments for soldiers. Forecasts from various global climate change models suggest that the habitats associated with some military training installations will face pressures that induce biome-shifts, invasive species, loss of habitat, and changes in training opportunities. This study combined worldwide habitat forecast data with a current habitat map to identify major installations that appear to be most and least at-risk for habitat change. DISCLAIMER: The contents of this report are not to be used for advertising, publication, or promotional purposes. Citation of trade names does not constitute an official endorsement or approval of the use of such commercial products. All product names and trademarks cited are the property of their respective owners. The findings of this report are not to be construed as an official Department of the Army position unless so designated by other authorized documents.
Annals of the Entomological Society of America
An IBM PC or compatible microcomputer equipped with a Tekmatic Systems Video Van Gogh digitizer c... more An IBM PC or compatible microcomputer equipped with a Tekmatic Systems Video Van Gogh digitizer card can be interfaced with a standard video camera for rapidly measuring the percentage of leaf area lost (LAL) to herbivorous insects. This capability can be added to an existing color PC for about $600 in additional hardware. The system digitizes a video camera image of a reconstructed leaf and displays the amount of LAL in about 20 s. The video digitizer simultaneously measures area removed from, and area remaining in, the leaf without tracing of these areas by the operator. Minimized operator interaction greatly increases the number of leaves that can be measured. Before digitizing, a photocopied image of the leaf must have its outer perimeter completed with a dark pencil line if it has been interrupted by folivore feeding. The video signal is digitally encoded into a 256 x 256 element false-color picture of the reconstructed damaged leaf on the computer monitor. A keyboard-adjustable brightness threshold distinguishes between leaf and background. Digitized measurements are saved to a sequential ASCII data file compatible with most popular database, spreadsheet, and word processor programs. The video digitizer software allows two video camera work stations to be supported by a single computer. Software for the video digitizer is released into the public domain and can be obtained from the authors for the cost of a diskette and postage.
Conservation Ecology, 2002
Ecol Lett, 2004
All global circulation models based on Intergovernmental Panel on Climate Change (IPCC) scenarios... more All global circulation models based on Intergovernmental Panel on Climate Change (IPCC) scenarios project profound changes, but there is no consensus on how to map their environmental consequences. Our multivariate representation of environmental space combines stable topographic and edaphic attributes with dynamic climatic attributes. We divide that environmental space into 500 unique domains and map their current locations and their projected locations in 2100 under contrasting emissions scenarios. The environmental domains found across half the study area today disappear under the higher emissions scenario, but persist somewhere in it under the lower emissions scenario. Locations affected least and those affected most under each scenario are mapped. This provides an explicit framework for designing conservation networks to include both areas at least risk (potential refugia) and areas at greatest risk, where novel communities may form and where sentinel ecosystems can be monitored for signs of stress.
Quart Rev Biol, 2001
ABSTRACT The National Early Warning System (EWS) provides an 8-day coast-to-coast snapshot of pot... more ABSTRACT The National Early Warning System (EWS) provides an 8-day coast-to-coast snapshot of potentially disturbed forests across the U.S.. A prototype system has produced national maps of potential forest disturbances every eight days since January 2010, identifying locations that may require further investigation. Through phenology, the system shows both early and delayed vegetation development and detects all types of unexpected forest disturbances, including insects, disease, wildfires, frost and ice damage, tornadoes, hurricanes, blowdowns, harvest, urbanization, landslides, drought, flood, and climate change. The USDA Forest Service Eastern Forest Environmental Threat Assessment Center is collaborating with NASA Stennis Space Center and the Western Wildland Environmental Threat Assessment Center to develop the tool. The EWS uses differences in phenological responses between an expectation based on historical data and a current view to strategically identify potential forest disturbances and direct attention to locations where forest behavior seems unusual. Disturbance maps are available via the Forest Change Assessment Viewer (FCAV) (http://ews.forestthreats.org/gis), which allows resource managers and other users to see the most current national disturbance maps as soon as they are available. Phenology-based detections show not only vegetation disturbances in the classical sense, but all departures from normal seasonal vegetation behavior. In 2010, the EWS detected a repeated late-frost event at high elevations in North Carolina, USA, that resulted in delayed seasonal development, contrasting with an early spring development at lower elevations, all within close geographic proximity. Throughout 2011, there was a high degree of correspondence between the National Climatic Data Center's North American Drought Monitor maps and EWS maps of phenological drought disturbance in forests. Urban forests showed earlier and more severe phenological drought disturbance than surrounding non-urban forests. An EWS news page (http://www.geobabbble.org/~hnw/EWSNews) highlights disturbances the system has detected during the 2011 season. Unsupervised statistical multivariate clustering of smoothed phenology data every 8 days over an 11-year period produces a detailed map of national vegetation types, including major disturbances. Examining the constancy of these phenological classifications at a particular location from year to year produces a national map showing the persistence of vegetation, regardless of vegetation type. Using spectral unmixing methods, national maps of evergreen decline can be produced which are a composite of insect, disease, and anthropogenic factors causing chronic decline in these forests, including hemlock wooly adelgid, mountain pine beetle, wildfire, tree harvest, and urbanization. Because phenology shows vegetation responses, all disturbance and recovery events detected by the EWS are viewed through the lens of the vegetation.
ABSTRACT Background/Question/Methods Up-scaling from sparse point based measurement to continuous... more ABSTRACT Background/Question/Methods Up-scaling from sparse point based measurement to continuous gridded product is a common problem in Earth System Science. We present a new general-purpose empirical imputation method based on associative clustering, which associates sparse measurements of dependent variables with particular multivariate clustered combinations of the independent variables, and then estimates values for unmeasured clusters, based on directional proximity in multidimensional data space, at both the cluster and map cell levels of resolution. A multivariate cluster analysis was applied to global output from a General Circulation Model (GCM) consisting of 17 variables downscaled to 4 km2 resolution. Present global growing conditions were divided into 30 thousand relatively homogeneous ecoregions describing climatic and topographic conditions. At every mapcell a multi-linear regression was applied in 17 dimensional hyperspace to derive the suitability of a tree species where not measured using the forest inventory data. The continuous species distribution maps obtained were compared and validated against existing tree range suitability maps. Associative Clustering is intended to be a general-purpose imputation tool, is model-free, and can be used to derive tree growth for future conditions that have no present-day analog. Results/Conclusions We demonstrate this new imputation tool on tree species range distribution maps, which describe the suitable extent and expected growth performance of a particular tree species over a wide area. Range maps having continuous estimates of tree growth performance are more useful than more classical tree range maps that simply show binary occurence suitability. The USDA Forest Service Forest Inventory Assessment (FIA) plots provide information about the occurence and growth performance for various tree species across the US, but such measurements are limited to FIA plots. Using Associative Clustering, we scale up the discontinuous FIA Inventory growth measurements into continuous maps that show the expected growth and suitabilty for individual tree species covering the Continental United States.
ABSTRACT Background/Question/Methods Range shifts for 215 tree species were predicted globally fo... more ABSTRACT Background/Question/Methods Range shifts for 215 tree species were predicted globally following climate changes forecast by the Parallel Climate Model (PCM) and the Hadley climate model under IPCC scenarios A2 and B2 for the years 2050 and 2100. Results/Conclusions Most species' predicted suitable (or fundamental) ranges closely followed or were slightly more extensive than Little's actual (or realized) ranges under present conditions. Maps of Minimum Required Movement (MRM) to return to the closest geographic locations offering suitable conditions in the future show the likelihood of local extinction following climate change. Locations that are the nearest "lifeboats" for large surrounding areas may represent management and conservation targets.
J Environ Manage, 2008
Habitat valuation methods are most often developed and used to prioritize candidate lands for con... more Habitat valuation methods are most often developed and used to prioritize candidate lands for conservation. In this study the intent of habitat valuation was to inform the decision-making process for remediation of chemical contaminants on specific lands or surface water bodies. Methods were developed to summarize dimensions of habitat value for six representative aquatic and terrestrial contaminated sites at the East Tennessee Technology Park (ETTP) on the US Department of Energy Oak Ridge Reservation in Oak Ridge, TN, USA. Several general valuation metrics were developed for three broad categories: site use by groups of organisms, site rarity, and use value added from spatial context. Examples of use value metrics are taxa richness, a direct measure of number of species that inhabit an area, complexity of habitat structure, an indirect measure of potential number of species that may use the area, and land use designation, a measure of the length of time that the area will be available for use. Measures of rarity included presence of rare species or communities. Examples of metrics for habitat use value added from spatial context included similarity or complementarity of neighboring habitat patches and presence of habitat corridors. More specific metrics were developed for groups of organisms in contaminated streams, ponds, and terrestrial ecosystems. For each of these metrics, cutoff values for high, medium, and low habitat value were suggested, based on available information on distributions of organisms and landscape features, as well as habitat use information. A companion paper describes the implementation of these habitat valuation metrics and scoring criteria in the remedial investigation for ETTP.
Agu Fall Meeting Abstracts, Dec 1, 2004
ABSTRACT We propose and implement a cluster-based approach for identifying global phenological ob... more ABSTRACT We propose and implement a cluster-based approach for identifying global phenological observatories in which phenologically and climatologically self-similar pixel clusters are monitored. We developed clusters based on a wavelet-filtered subset of the 1982-1999 global Pathfinder Advanced Very High Resolution Radiometer (AVHRR) Land (PAL) Normalized Difference Vegetation Index (NDVI) dataset, a global 10-minute resolution climatology, and the clustering approach developed by the Oak Ridge National Laboratories (ORNL). In the ORNL approach: (1) n cluster centers are defined based on the multi-dimensional NDVI/climate space; (2) pixel distances from the centroids are calculated; (3) pixels are assigned to the minimum distance cluster. While any number of clusters may be specified, we found that a global 500-cluster approach provided a satisfactory global distribution. In traditional rectangular approaches a group of pixels could contain desert, grassland, and tropical forest. Here, longitudinally extensive but latitudinally limited regions such as the Sahel exist as distinct groups. Thus, our approach avoids problems affecting single-pixel approaches (misregistration, cloud contamination) and rectangular approaches (mixed phenological signals). Using the 1982-2003 GIMMS AVHRR dataset, we extracted phenological metrics such as the onset and offset of greenness for each cluster. We then ranked each cluster based on land cover homogeneity, evidence of human impacts, and political diversity. For each biome, we then identified the highest ranked clusters within four climate zones (hot/wet, hot/dry, cold/wet, cold/dry). This strategy provides: (1) selection of regions for which a strong annual is detectable, (2) a method of identifying regions least likely to be impacted by non-climatic factors, and (3) a strategy for ground validation.
Photogrammetric Engineering and Remote Sensing, 2009
We are using a statistical clustering method for delineating homogeneous ecoregions as a basis fo... more We are using a statistical clustering method for delineating homogeneous ecoregions as a basis for identifying disturbances in forests through time over large areas, up to national and global extents. Such changes can be shown relative to past conditions, or can be predicted relative to present conditions, as with forecasts of future climatic change. This quantitative ecoregion approach can be used to predict destinations for populations whose local environments are forecast to become unsuitable and are forced to migrate as their habitat shifts, and is also useful for predicting the susceptibility of new locations to invasive species like Sudden Oak Death. EFETAC and our sister western center WWETAC, along with our NASA and ORNL collaborators, are designing a new national-scale early warning system for forest threats, called FIRST. Envisioned as a change-detection system, FIRST will identify all land surface cover changes at the MODIS observational scale, and then try to discriminate normal, expected seasonal changes from locations having unusual activity that may represent potential forest threats. As a start, we have developed new national data sets every 16 days from 2002 through 2008, based on land surface phenology, or timing of leaf-out in the spring and brown-down in the fall. Changes in such phenological maps will be shown to contain important information about vegetation health status across the United States. The standard deviation of the duration of fall can be mapped, showing places where length of fall is relatively constant or is variable in length from year to year.