Marc Simard | Jet Propulsion Laboratory, California Institute of Technology (original) (raw)
Papers by Marc Simard
Canopy height is one of the strongest predictors of biomass and carbon in forested ecosystems. Ad... more Canopy height is one of the strongest predictors of biomass and carbon in forested ecosystems. Additionally, mangrove ecosystems represent one of the most concentrated carbon reservoirs that are rapidly degrading as a result of deforestation, development, and hydrologic manipulation. Therefore, the accuracy of Canopy Height Models (CHM) over mangrove forest can provide crucial information for monitoring and verification protocols. We compared four CHMs derived from independent remotely sensed imagery and identified potential errors and bias between measurement types. CHMs were derived from three spaceborne datasets; Very-High Resolution (VHR) stereophotogrammetry, TerraSAR-X add-on for Digital Elevation Measurement, and Shuttle Radar Topography Mission (TanDEM-X), and lidar data which was acquired from an airborne platform. Each dataset exhibited different error characteristics that were related to spatial resolution, sensitivities of the sensors, and reference frames. Canopies over 10 m were accurately predicted by all CHMs while the distributions of canopy height were best predicted by the VHR CHM. Depending on the guidelines and strategies needed for monitoring and verification activities, coarse resolution CHMs could be used to track canopy height at regional and global scales with finer resolution imagery used to validate and monitor critical areas undergoing rapid changes.
Canadian Journal of Remote Sensing, 1998
... La méthodologie est décrite en détail. Les approximations à basse-resolution, produites par l... more ... La méthodologie est décrite en détail. Les approximations à basse-resolution, produites par l'algorithme de transformée par ondelettes sont utilisées pour la construction de la mosaïque RSO ERS-1 centre africaine. Revue / Journal Title. ... Algorithme. ; ... Transformation ondelette. ; ...
This paper discusses the innovative concept and technology development of a Ka-band (35 GHz) rada... more This paper discusses the innovative concept and technology development of a Ka-band (35 GHz) radar for mapping the surface topography of glaciers and ice sheets. Dubbed the "Glacier and Land Ice Surface Topography Interferometer" (GLISTIN) the system is a single-pass, single platform interferometric synthetic aperture radar (InSAR) with an 8mm wavelength, which minimizes snow penetration yet remains relatively impervious to
We characterized global vegetation 3D structure using ICEsat-I/Geoscience Laser Altimeter (GLAS) ... more We characterized global vegetation 3D structure using ICEsat-I/Geoscience Laser Altimeter (GLAS) and improved spatial resolution using ALOS/Phased Array L-band Synthetic Aperture radar (PALSAR) data over 3 sites in the United States. GLAS is a 70m footprint lidar altimeter sampling the ground along-track every 170m with a track separation near the equator around 30km. Forest type classes were initially defined according
The combination of Lidar and radar has the potential to produce regional canopy height maps. Prev... more The combination of Lidar and radar has the potential to produce regional canopy height maps. Previous work has demonstrated these maps can be generated with interferometric radar and calibrated using Lidar points. This calibration procedure can be improved by incorporating pre-existing knowledge on spatial patterns of canopy height. In this case, the relationship between radar and Lidar can be computed
We present the first analysis and results of the August 2009 UAVSAR and field campaign in tempera... more We present the first analysis and results of the August 2009 UAVSAR and field campaign in temperate and boreal forests. UAVSAR is an airborne fully polarimetric L-band radar capable of repeat-pass interferometry (inSAR). We focus on 3 sites located in New Hampshire, Maine and Québec. The UAVSAR sites covered Bartlett, Hubbard Brook, Penobscot, Howland and Mont-Morency experimental forests in addition to covering part of White Mountain National Forest, Laurentides Wildlife Reserve, Jacques-Cartier and Grand-Jardins National Parks. We collected field data on forest structure in a total of 50 plots with measurements including tree height, DBH, species as well as crown sizes. We compiled the field data to derive canopy structure metrics such as canopy height and tree size class distribution. UAVSAR collected data over the 3 sites on 4 different days spread throughout the 11-day campaign. This timing strategy was to enable us to investigate the impact of temporal change on inSAR coherence. However, in this presentation, we focus on the analysis of single coherence measurements and polarimetric radar backscatter. We investigate their dependence on vegetation type, slope as well as weather. This is achieved by direct comparison with field, forestry and lidar data, and with a 3D vegetation model. We use a "Regression Tree" model with explicit relationships between field- and lidar-estimated canopy heights, environmental variables, radar backscatter and coherence, and MODIS percent tree cover. We assess the potential of this approach and the usefulness of each input datasets. Finally, we report on the specific accuracy of L-band radar polarimetry to estimate biomass throughout the UAVSAR sites. UAVSAR first image near Quebec city.
Remote Sensing of Environment
2006 IEEE International Symposium on Geoscience and Remote Sensing, 2006
ABSTRACT Remote sensing technologies have provided valuable data for landscape modeling, vegetati... more ABSTRACT Remote sensing technologies have provided valuable data for landscape modeling, vegetation mapping and comprehensive studies of ecosystems. Airborne laser mapping or LIDAR (Light Detection and Ranging) can directly measure the three dimensional structure of plant canopies, as well as, provide accurate digital terrain models (DTM). While temperate and boreal pine forests have been studied using these methods, very limited work has been done in subtropical pine forests. In this study, airborne LIDAR was used to characterize the three dimensional structure of the forest on Long Pine Key at Everglades National Park. Analysis of the vertical distribution of airborne LIDAR data points has shown that distinctive patterns can be described which are characteristic of the vegetation communities and transition zones for pine forests , hammocks and marshes. This information is a valuable resource for forest managers by providing landscape structural data over large areas.
Vegetation 3D structure measurements from active remote sensing (i.e. lidar and radar) are usuall... more Vegetation 3D structure measurements from active remote sensing (i.e. lidar and radar) are usually averaged and reported at the regional level. However, environmental gradients and disturbance can structure vegetation patterns at multiple scales. Thus, a critical challenge in designing global observation strategies is to obtain confidence intervals on vegetation parameters as a function of biome, sensor, and resolution of observation.
This paper will present quantitative results of vegetation parameter extraction using inteferomet... more This paper will present quantitative results of vegetation parameter extraction using inteferometric data collected using the TOPSAR and GeoSAR mapping instruments. These radar operate interferometrically at an range of frequencies from X-band to P-band. Radar data derived vegetation parameters are compared to LIDAR data and in situ measurements for a variety of canopy and terrain types. Comparison of how the different frequencies interact with the vegetation as a function of tree height, and incidence angle and canopy parameters will be presented.
We review our recent advances in understanding the role of temporal decorrelation in SAR interfer... more We review our recent advances in understanding the role of temporal decorrelation in SAR interferometry and polarimetric SAR interferometry. We developed a physical model of temporal decorrelation based on Gaussian-statistic motion that varies along the verti-cal direction in forest canopies. Temporal decorre-lation depends on structural parameters such as for-est height, is sensitive to polarization and affects co-herence amplitude and phase. A model of temporal-volume decorrelation valid for arbitrary spatial base-line is discussed. We tested the inversion of this model to estimate forest height from model simulations sup-ported by JPL/UAVSAR data and lidar LVIS data. We found a general good agreement between forest height estimated from radar data and forest height es-timated from lidar data.
Canopy height is one of the strongest predictors of biomass and carbon in forested ecosystems. Ad... more Canopy height is one of the strongest predictors of biomass and carbon in forested ecosystems. Additionally, mangrove ecosystems represent one of the most concentrated carbon reservoirs that are rapidly degrading as a result of deforestation, development, and hydrologic manipulation. Therefore, the accuracy of Canopy Height Models (CHM) over mangrove forest can provide crucial information for monitoring and verification protocols. We compared four CHMs derived from independent remotely sensed imagery and identified potential errors and bias between measurement types. CHMs were derived from three spaceborne datasets; Very-High Resolution (VHR) stereophotogrammetry, TerraSAR-X add-on for Digital Elevation Measurement, and Shuttle Radar Topography Mission (TanDEM-X), and lidar data which was acquired from an airborne platform. Each dataset exhibited different error characteristics that were related to spatial resolution, sensitivities of the sensors, and reference frames. Canopies over 10 m were accurately predicted by all CHMs while the distributions of canopy height were best predicted by the VHR CHM. Depending on the guidelines and strategies needed for monitoring and verification activities, coarse resolution CHMs could be used to track canopy height at regional and global scales with finer resolution imagery used to validate and monitor critical areas undergoing rapid changes.
Canadian Journal of Remote Sensing, 1998
... La méthodologie est décrite en détail. Les approximations à basse-resolution, produites par l... more ... La méthodologie est décrite en détail. Les approximations à basse-resolution, produites par l'algorithme de transformée par ondelettes sont utilisées pour la construction de la mosaïque RSO ERS-1 centre africaine. Revue / Journal Title. ... Algorithme. ; ... Transformation ondelette. ; ...
This paper discusses the innovative concept and technology development of a Ka-band (35 GHz) rada... more This paper discusses the innovative concept and technology development of a Ka-band (35 GHz) radar for mapping the surface topography of glaciers and ice sheets. Dubbed the "Glacier and Land Ice Surface Topography Interferometer" (GLISTIN) the system is a single-pass, single platform interferometric synthetic aperture radar (InSAR) with an 8mm wavelength, which minimizes snow penetration yet remains relatively impervious to
We characterized global vegetation 3D structure using ICEsat-I/Geoscience Laser Altimeter (GLAS) ... more We characterized global vegetation 3D structure using ICEsat-I/Geoscience Laser Altimeter (GLAS) and improved spatial resolution using ALOS/Phased Array L-band Synthetic Aperture radar (PALSAR) data over 3 sites in the United States. GLAS is a 70m footprint lidar altimeter sampling the ground along-track every 170m with a track separation near the equator around 30km. Forest type classes were initially defined according
The combination of Lidar and radar has the potential to produce regional canopy height maps. Prev... more The combination of Lidar and radar has the potential to produce regional canopy height maps. Previous work has demonstrated these maps can be generated with interferometric radar and calibrated using Lidar points. This calibration procedure can be improved by incorporating pre-existing knowledge on spatial patterns of canopy height. In this case, the relationship between radar and Lidar can be computed
We present the first analysis and results of the August 2009 UAVSAR and field campaign in tempera... more We present the first analysis and results of the August 2009 UAVSAR and field campaign in temperate and boreal forests. UAVSAR is an airborne fully polarimetric L-band radar capable of repeat-pass interferometry (inSAR). We focus on 3 sites located in New Hampshire, Maine and Québec. The UAVSAR sites covered Bartlett, Hubbard Brook, Penobscot, Howland and Mont-Morency experimental forests in addition to covering part of White Mountain National Forest, Laurentides Wildlife Reserve, Jacques-Cartier and Grand-Jardins National Parks. We collected field data on forest structure in a total of 50 plots with measurements including tree height, DBH, species as well as crown sizes. We compiled the field data to derive canopy structure metrics such as canopy height and tree size class distribution. UAVSAR collected data over the 3 sites on 4 different days spread throughout the 11-day campaign. This timing strategy was to enable us to investigate the impact of temporal change on inSAR coherence. However, in this presentation, we focus on the analysis of single coherence measurements and polarimetric radar backscatter. We investigate their dependence on vegetation type, slope as well as weather. This is achieved by direct comparison with field, forestry and lidar data, and with a 3D vegetation model. We use a "Regression Tree" model with explicit relationships between field- and lidar-estimated canopy heights, environmental variables, radar backscatter and coherence, and MODIS percent tree cover. We assess the potential of this approach and the usefulness of each input datasets. Finally, we report on the specific accuracy of L-band radar polarimetry to estimate biomass throughout the UAVSAR sites. UAVSAR first image near Quebec city.
Remote Sensing of Environment
2006 IEEE International Symposium on Geoscience and Remote Sensing, 2006
ABSTRACT Remote sensing technologies have provided valuable data for landscape modeling, vegetati... more ABSTRACT Remote sensing technologies have provided valuable data for landscape modeling, vegetation mapping and comprehensive studies of ecosystems. Airborne laser mapping or LIDAR (Light Detection and Ranging) can directly measure the three dimensional structure of plant canopies, as well as, provide accurate digital terrain models (DTM). While temperate and boreal pine forests have been studied using these methods, very limited work has been done in subtropical pine forests. In this study, airborne LIDAR was used to characterize the three dimensional structure of the forest on Long Pine Key at Everglades National Park. Analysis of the vertical distribution of airborne LIDAR data points has shown that distinctive patterns can be described which are characteristic of the vegetation communities and transition zones for pine forests , hammocks and marshes. This information is a valuable resource for forest managers by providing landscape structural data over large areas.
Vegetation 3D structure measurements from active remote sensing (i.e. lidar and radar) are usuall... more Vegetation 3D structure measurements from active remote sensing (i.e. lidar and radar) are usually averaged and reported at the regional level. However, environmental gradients and disturbance can structure vegetation patterns at multiple scales. Thus, a critical challenge in designing global observation strategies is to obtain confidence intervals on vegetation parameters as a function of biome, sensor, and resolution of observation.
This paper will present quantitative results of vegetation parameter extraction using inteferomet... more This paper will present quantitative results of vegetation parameter extraction using inteferometric data collected using the TOPSAR and GeoSAR mapping instruments. These radar operate interferometrically at an range of frequencies from X-band to P-band. Radar data derived vegetation parameters are compared to LIDAR data and in situ measurements for a variety of canopy and terrain types. Comparison of how the different frequencies interact with the vegetation as a function of tree height, and incidence angle and canopy parameters will be presented.
We review our recent advances in understanding the role of temporal decorrelation in SAR interfer... more We review our recent advances in understanding the role of temporal decorrelation in SAR interferometry and polarimetric SAR interferometry. We developed a physical model of temporal decorrelation based on Gaussian-statistic motion that varies along the verti-cal direction in forest canopies. Temporal decorre-lation depends on structural parameters such as for-est height, is sensitive to polarization and affects co-herence amplitude and phase. A model of temporal-volume decorrelation valid for arbitrary spatial base-line is discussed. We tested the inversion of this model to estimate forest height from model simulations sup-ported by JPL/UAVSAR data and lidar LVIS data. We found a general good agreement between forest height estimated from radar data and forest height es-timated from lidar data.