1. Linking Remote Sensing with a State-of-the-Art Terrestrial Biosphere Model to Better Predict Ecosystem Dynamics
    Antonarakis, A. S.; Saatchi, S. S. B. B. M. P. R.  AGU  December 2010

    Until recently, estimates of forest structure and composition have been based on ground-based forest inventories that are limited in their spatial extent, and do not provide a comprehensive estimate of the current state of the above-ground ecosystem. Previous studies have shown how active remote sensing measurements can provide information on forest structural attributes, and how such remotely-sensed estimates of forest structural attributes can be used to help constrain the predictions of terrestrial biosphere models. However, to date, these remote sensing estimates of forest structure have focused on developing estimates for single forest metrics, such as lidar estimates of canopy height or radar estimates of above-ground biomass. In this study, we investigate how a combination of remote sensing data over Harvard Forest, MA, can be used to best estimate forest above-ground ecosystem state and subsequently constrain and test the dynamics of the ED2 state-of-the-art terrestrial biosphere model. Following on from previous work, this fusion of forest structural attributes is more than just canopy height and biomass, and aims at using remote sensing signals more completely. Recent results show that waveform lidar data from LVIS is successful (R2 = 0.8, RMSE = 0.7) in determining LAI over 40 plots at Harvard Forest from canopy gap probabilities. This was better than just considering NDVI-LAI relationships from optical remote sensing techniques such as EO-1 Hyperion (R2 = 0.45). We then use the remote-sensing derived estimates of forest structure to constrain the state of above-ground ecosystems within ED2, a state-of-the-art terrestrial biosphere model, and quantify the impacts of this reduction in uncertainty regarding the current ecosystem state on predictions of current and future biophysical and biogeochemical functioning at Harvard Forest.



  2. Large-area Ice Sheet and Sea Ice mapping from High-altitude Aircraft: Examples from the LVIS Sensor
    Blair, J. B.; Hofton, M. A. R. D. L.  AGU  December 2010

    High altitude airborne surveys of remote polar regions is a relatively recent addition to the remote sensing capabilities serving the Cryospheric science community. The NASA/GSFC-developed airborne sensor, LVIS (Land, Vegetation, and Ice Sensor) is a wide-swath, full-waveform laser altimeter system that produces large-area topographic maps with the highest levels of accuracy and precision. Recent data collections in support of NASA's Operation IceBridge over Antarctica and Greenland have demonstrated the extraordinary mapping capability of the LVIS sensor. Areal coverage is accumulated at a rate of > 1,000 sq. km/hr with repeatability of the surface elevation measurements at the decimeter level. With this new capability come new applications, new insights, the ability to fully capture the spatial extent and variability of changes occurring in highly dynamic areas, and enhanced input into ice sheet models. One example is over 7,000 sq. km collected over the Antarctic Peninsula in just 7 hours from 40,000 ft on the NASA DC-8 aircraft. The wide swath and dense coverage enabled by the LVIS sensor results in significant overlap with legacy ICESat data permitting statistically powerful comparisons and eliminate the need for interpolation or slope corrections. Several examples of ICESat comparisons and change detection between LVIS data takes and other topographic data sets will be presented . Further, a description of the LVIS waveform vector data product and examples of advanced data products and analysis techniques with be shown. A fully-autonomous version of LVIS is now under development (LVIS-GH) for use in the Global Hawk aircraft. Long duration flights over remote areas will be possible with this sensor. Testing on the Global Hawk UAV is scheduled for the Summer of 2011. The LVIS data are freely available from the NSIDC website (http://nsidc.org/data/icebridge/) and the LVIS website (http://lvis.gsfc.nasa.gov).



  3. Temporal variations in soil moisture content and its influence on biomass estimates, observed by UAVSAR, ALOS PALSAR, and in-situ field data
    Calderhead, A. I.; Simard, M. L. M.  AGU  December 2010

    Temporal changes of repeat-pass SAR backscatter over bare ground or forests results mostly from changes in the target's dielectric properties or moisture content; especially when the timescale is on the order of a few days or weeks. It is important to properly correct for moisture content when using SAR based estimates of tree height or biomass. The objective of this work is to quantify the error in biomass estimates associated with variations in moisture content in temperate and boreal forested areas. In addition, the accuracy of three polarimetric soil moisture surface inversion models (Dubois et al., 1995, Oh et al., 1992; Oh, 2004) are tested on UAVSAR and PALSAR data of bare soils in temperate and boreal forested areas. In addition to PALSAR data from 2007 to 2009, a JPL/UAVSAR campaign over parts of New England and Quebec was completed in August, 2009; L-band SAR images were acquired on August 5th, August 7th, and August 14th. In-situ soil moisture probes at three locations gathered hourly soil moisture content data. LVIS LIDAR is used for quantifying and classifying biomass ranges. Slope corrected backscatter values resampled to 1 hectare at HH, HV, and VV polarizations, and ratios thereof, are compared with soil moisture, precipitation, biomass, and incidence angle. It is seen that the backscatter for high biomass areas varies significantly due to moisture variations. An increase in 1% soil moisture content at the Laurentides field site leads to a change in HV backscatter of 1dB. Regions with high biomass do not vary uniformly with varying moisture content: this can be explained by saturation of the L-band at higher biomass levels. The three inversion algorithms produce varying results with the ‘Dubois et al’ inversion producing the best correlation at the Bartlett Forest site while the ‘Oh 2004’ inversion produces better results at the Laurentides site. Although the accuracy is often poor, the temporal variation of the moisture content for all three inversion algorithms is generally captured.



  4. A Bayesian functional data model for predicting forest variables using high-dimensional waveform LiDAR over large geographic domains
    Finley, A. O.  AGU  December 2010

    Recent advances in remote sensing, specifically waveform Light Detection and Ranging (LiDAR) sensors, provide the data needed to quantify forest variables at a fine spatial resolution over large domains. Of particular interest is LiDAR data from NASA's Laser Vegetation Imaging Sensor (LVIS), upcoming Deformation, Ecosystem Structure, and Dynamics of Ice (DESDynI) missions, and NSF's National Ecological Observatory Network planned Airborne Observation Platform. A central challenge to using these data is to couple field measurements of forest variables (e.g., species, indices of structural complexity, light competition, or drought stress) with the high-dimensional LiDAR signal through a model, which allows prediction of the tree-level variables at locations where only the remotely sensed data area are available. It is common to model the high-dimensional signal vector as a mixture of a relatively small number of Gaussian distributions. The parameters from these Gaussian distributions, or indices derived from the parameters, can then be used as regressors in a regression model. These approaches retain only a small amount of information contained in the signal. Further, it is not known a priori which features of the signal explain the most variability in the response variables. It is possible to fully exploit the information in the signal by treating it as an object, thus, we define a framework to couple a spatial latent factor model with forest variables using a fully Bayesian functional spatial data analysis. Our proposed modeling framework explicitly: 1) reduces the dimensionality of signals in an optimal way (i.e., preserves the information that describes the maximum variability in response variable); 2) propagates uncertainty in data and parameters through to prediction, and; 3) acknowledges and leverages spatial dependence among the regressors and model residuals to meet statistical assumptions and improve prediction. The proposed modeling framework is illustrated using DESDynI-like waveform LiDAR and spatially coinciding forest inventory data collected on the Penobscot Experimental Forest, Maine.



  5. Characterizing Ice Sheet Surface Topography and Structure Using High-Altitude Waveform Airborne Laser Altimetry
    Hofton, M. A.; Blair, B. L. S. B. R. D. M. C. B. M.  AGU  December 2010

    Surface topographic information of ice surfaces is important for a wide range of applications including mass balance investigations and dynamical modeling. Airborne LIght Detection And Ranging (lidar) uses laser ranging to map surface topography with high precision and accuracy. In October-November 2010, NASA's Land, Vegetation and Ice Sensor (LVIS) system imaged areas of Antarctica as part of NASA's Operation IceBridge (OIB) on board the DC-8 aircraft. The LVIS is an airborne, medium-footprint (~25 m diameter), wide swath (~2 km) full waveform-recording, scanning lidar system that has been used extensively for mapping surface structure for various investigations. The system digitally records the shape of the returning laser echo, or waveform, after its interaction with the various reflecting surfaces of the earth, providing a true 3-dimensional record of the surface structure within each footprint in the data swath. During the 2009 Antarctica deployment, LVIS Lidar data were collected over the Antarctic Peninsula, Pine Island Glacier and along a ~1100 km-long transect around 86S from flight altitudes of ~35,000' to ~39,000', and complemented and enhanced the low-altitude radar and laser datasets also collected during the mission. We examine the LVIS lidar imaging data sets, in particular that collected in the Antarctic Peninsula where an 11 hour flight enabled the complete mapping of a ~250 km by ~30 km area centered on the Crane Glacier. A precision and accuracy assessment of the high-resolution digital topographic data products is presented, as well as comparisons of the data to existing elevation maps from ASTER and SPOT5 to quantify surface changes within several mapped glacier systems. This mapping demonstrates the unprecedented spatial coverage and accuracy of the high-altitude lidar data collected during OIB.



  6. Forest Biomass Mapping Using Lidar-derived Canopy Height Metrics at Maine in USA
    Huang, W.; Sun, G.  AGU  December 2010

    Forest biomass from regional to global level is important for underlying and monitoring the ecosystem responses to natural and human activities. Lidar provides the ability to directly measure canopy height index for aboveground biomass estimation. Our study site is located in Howland, Maine, United States. Data source consists of airborne medium footprint lidar data in 2009 and ground data from DESDynI field campaign in August 2009 and 2010. Canopy vertical structures are captured by the Laser Vegetation Imaging Sensor (LVIS) with entire return signal (i.e. in ~30 cm vertical bins). We first calculated height metrics (i.e. h10 to h100, totally 15 indices) by waveform decomposition using either Gaussian or numeric filter. Then, metrics were compared with RH indices at different levels: footprint of 20m diameter circle, squared plot of 25 x 25m, 50 x 50 m, 50 x 100 m and 50 x 200 m, respectively. At last, the biomass map was created. Height metrics from h50 to h80 show high correlation with biomass. Among them, h65 and h70 are the best, which is consistent with previous perspective that RH50 (or HOME, height of median energy) and RH75 have the best linear relationship with aboveground biomass. Comparison between h metrics and RH indices shows the latter one is better. In addition, both single and multi-variable linear regression model significant improvement with the increasing of field plot size.



  7. Monitoring Forest Carbon Dynamics for REDD: A Landsat-Lidar Fusion Approach
    Huang, C.; Dubayah, R. H. G. C. G. S. N. M. J. G. Z. Z.  AGU  December 2010

    Reducing Emissions from Deforestation and Forest Degradation (REDD) is an effort to create a financial value for the carbon stored in forests and to offer incentives for developing countries to reduce emissions from forested lands. Implementing this effort requires methods for quantifying forest carbon and change. Such methods should be accurate enough to allow reliable reporting and efficient enough to enable timely verification and monitoring. Here we present a Landsat-lidar fusion approach for monitoring the dynamics of forest carbon. In this approach, time series Landsat observations are used to detect and date forest disturbance and to track the spectral trajectory of post-disturbance recovery using a vegetation change tracker (VCT) algorithm. Biomass estimates derived from LVIS lidar samples will then be used to establish relationships between standing biomass and age since disturbance and the recovery trajectory. Such relationships can be used to estimate forest biomass not only during the model year, but also for the years after the model year. This is because each disturbance has a time stamp, which can be used to calculate the age since disturbance and the post-disturbance recovery trajectory for any year after the disturbance year using available Landsat images. Therefore, it can not only be used to establish baseline estimates, but also to monitor changes due to both disturbances and recovery. Furthermore, the fine spatial resolutions of the Landsat and LVIS data allow the biomass and biomass change estimates to be derived at hectare or sub-hectare levels. Such fine grain sizes will allow reliable reporting at patch or individual land owner level, which is required for fine scale carbon management and carbon trade at individual land owner level. Critical environmental variables controlling biomass recovery rates may also be revealed by analyzing the variability of age/biomass relationships among patches. The effectiveness of the described approach has been demonstrated in a statewide study in Mississippi, and will be further demonstrated through case studies conducted in the southeastern, Mid-Atlantic, and northeastern regions of the U.S.



  8. Interoperable Data Systems for Satellite, Airborne, and Terrestrial LiDAR Data
    Meertens, C. M.; Baru, C. B. B. C. C. J. H. T. M. H. D. J. H. M. A. K. S. S. M. J.  AGU  December 2010

    LiDAR (Light Detection and Ranging) technology is being widely applied to scientific problems on global to local scales using a range of laser technologies mounted on satellite, low- and high-altitude airborne and terrestrial platforms. Modern laser ranging instruments are increasingly capable of providing full waveform data, multiple detectors, higher sample rates and longer ranges. Accompanying these improvements, however, are rapidly growing data volumes and ever more complex data formats and processing algorithms. This presents significant challenges for existing Earth science data systems serving these data and creates barriers to the efficient use of these data by a growing and diverse community of scientific and other users who are studying deformation of the solid Earth, the cryosphere, vegetation structure, and land form evolution. To address these challenges, a group of data centers is collaborating under a project funded by the NASA ROSES ACCESS Program to develop interoperable LiDAR data access systems to provide integrated access to data and derived products in common data formats via simple-to-navigate web interfaces. The web service-based systems created by this project, called NLAS, will enhance access to existing laser data sources hosted at the National Snow and Ice Data Center DAAC, Goddard Space Flight Center LVIS Data Center, UNAVCO, and the OpenTopography Facility at the San Diego Supercomputer Center (SDSC). Through the OpenTopography portal, NLAS systems will provide access to satellite laser altimetry data from ICESat and high altitude airborne laser scanning data from LVIS, as well as low altitude airborne LiDAR and terrestrial laser scanning data



  9. Bruce Plateau, Antarctic Peninsula: Ice-Core Site Characterization
    Pettit, E. C.; Scambos, T. A. B. R. J. M.-T. E. S. T. M. B. B.  AGU  December 2010

    The Bruce Plateau is a broad, gently-undulating ice plateau spanning the divide of the Antarctic Peninsula near 66°S. The western side is the catchment area for the glaciers of numerous inlet fjords such as Andvord Bay, Beascochea Bay, and Barilari Bay. The eastern side is the catchment area for the glaciers of the southern Larsen B Ice Shelf and northern Larsen C Ice Shelf. Because it is the catchment for the Larsen B, the Bruce Plateau was chosen as a site to drill an ice core for paleoclimate studies. We present the results of a site characterization study of a 10 km × 10 km area near the ridge crest. We mapped surface topography using the LVIS (Laser Vegetation Imaging Sensor) instrument collected as part of NASA’s Ice Bridge program and extended these data using ground-based kinematic GPS profiles. We mapped bedrock topography through a 5 MHz Radio Echo Sounding (RES) survey. A weather station augmented with firn temperature sensors and C/A code GPS was installed at the ice-core site in February 2010 and was active for 5 months. We measured the spatial accumulation rate pattern and internal structure of the ice to 400 m with a 25 MHz RES survey near the ice-core site. The RES-mapped relative accumulation pattern was tied to measurements at the weather station, a temporary stake network, and data from the ice core. The surface topography data show that the crest of the ice divide ranges from 1980 to 2020 m above the ellipsoid, with surfaces sloping at 0.047 to the northeast (Leppard Glacier catchment) and approximately 0.04 to the southwest (Atlee Glacier). The bedrock topography consists of two hills of a few hundred meters relief, with deeper troughs under the regions leading to both Leppard and Atlee Glaciers. Ice thickness ranges from 200 m above the hills to more than 700 m deep in the troughs, as determined by a 5 MHz RES system. To the west of the hills, bedrock slopes gently downward, while surface elevation increases to the crest. This suggests that the ice divide may be up to 2 km to the west of the bedrock divide. We selected the ice-core site at 1976 m above the ellipsoid, just to the east of the divide crest, with an ice thickness of 450 m and a surface flow speed of 10±4 m/yr. Accumulation at the ice-core site averaged 1.1 meters of snow per month between February 17 and July 20, 2010. The data indicate a seasonal variation in snowfall, lower in the summer and higher in fall and early winter. This is supported by summertime stake measurements which averaged less than 0.4 m/mo over two months. During the five months of weather station data, air temperatures varied from -2°C to less than -25°C (communication typically ceased below -25°C). Firn temperature as measured by thermistors at several depths (up to 120 m) indicate a recent mean annual temperature of -15.1°C. This suite of data suggests a strong spatial gradient in accumulation, possible strong seasonal accumulation variation, and indicates of recent (within the top 150 m of internal layering) migration of the crest of the ridge.



  10. Linking tree size distribution to active remote sensing parameters: consequences for observation strategies and impacts on biomass retrieval (Invited)
    Pinto, N.; Simard, M. B. K. D. K. T. H.  AGU  December 2010

    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. We present strategies to gain knowledge on forest spatial heterogeneity that can be translated into confidence intervals for above ground biomass and canopy height measurements. We use data from two airborne systems: the Laser Vegetation Imaging Sensor (LVIS) and the Uninhabited Aerial Vehicle Synthetic Aperture Radar (UAVSAR) acquired over sites in the US (NH and ME), Canada (Quebec) and Costa Rica. We first describe two parameters (alpha and beta) that summarize tree size distribution for individual patches, thereby capturing forest successional stage. In this scenario, the uncertainty in predicting above ground biomass stems from: (1) the ability to estimate alpha and beta with the lidar/radar signals, and (2) the error in deriving above ground biomass from tree size distribution statistics. The processes of competition and self-thinning create skewed tree size distributions where smaller individuals are common and large individuals are rare. Using a global dataset of spaceborne lidar points from the sensor ICESat (Ice, Cloud, and land Elevation Satellite), we show the importance of sampling extreme values when using spatially sparse data. This raises the need to obtain expectations for the second-order properties of forest stands. To this end, we employed wavelet transforms to quantify variation in lidar-derived canopy height metrics across >20 Km transects and asked whether environmental gradients such as elevation can constrain the spatial autocorrelation among large trees.



  11. Modelling Sensor and Target effects on LiDAR Waveforms
    Rosette, J.; North, P. R. R. J. C. B. D. S. J.  AGU  December 2010

    The aim of this research is to explore the influence of sensor characteristics and interactions with vegetation and terrain properties on the estimation of vegetation parameters from LiDAR waveforms. This is carried out using waveform simulations produced by the FLIGHT radiative transfer model which is based on Monte Carlo simulation of photon transport (North, 1996; North et al., 2010). The opportunities for vegetation analysis that are offered by LiDAR modelling are also demonstrated by other authors e.g. Sun and Ranson, 2000; Ni-Meister et al., 2001. Simulations from the FLIGHT model were driven using reflectance and transmittance properties collected from the Howland Research Forest, Maine, USA in 2003 together with a tree list for a 200m x 150m area. This was generated using field measurements of location, species and diameter at breast height. Tree height and crown dimensions of individual trees were calculated using relationships established with a competition index determined for this site. Waveforms obtained by the Laser Vegetation Imaging Sensor (LVIS) were used as validation of simulations. This provided a base from which factors such as slope, laser incidence angle and pulse width could be varied. This has enabled the effect of instrument design and laser interactions with different surface characteristics to be tested. As such, waveform simulation is relevant for the development of future satellite LiDAR sensors, such as NASA’s forthcoming DESDynI mission (NASA, 2010), which aim to improve capabilities of vegetation parameter estimation. ACKNOWLEDGMENTS We would like to thank scientists at the Biospheric Sciences Branch of NASA Goddard Space Flight Center, in particular to Jon Ranson and Bryan Blair. This work forms part of research funded by the NASA DESDynI project and the UK Natural Environment Research Council (NE/F021437/1). REFERENCES NASA, 2010, DESDynI: Deformation, Ecosystem Structure and Dynamics of Ice. http://desdyni.jpl.nasa.gov/ (accessed May 2010). NI-MEISTER, W., JUPP, D. L. B. and DUBAYAH, R., 2001, Modeling Lidar Waveforms in Heterogeneous and Discrete Canopies. IEEE Transactions on Geoscience and Remote Sensing, 39 (9): 1943-1958. NORTH, P. R. J., 1996, Three-Dimensional Forest Light Interaction Model Using a Monte Carlo Method. IEEE Transactions on Geoscience and Remote Sensing, 34 (4): 946-956. NORTH, P. R. J., ROSETTE, J. A. B., SUÁREZ, J. C. and LOS, S. O., 2010, A Monte Carlo radiative transfer model of satellite waveform lidar. International Journal of Remote Sensing, 31 (5): 1343-1358. SUN, G. and RANSON, K. J., 2000, Modeling lidar returns from forest canopies. IEEE Transactions on Geoscience and Remote Sensing, 38 (6): 2617-2626.



  12. Applying the Moment Distance Framework to LiDAR Waveforms
    Salas, E. L.; Aguilar-Amuchastegui, N. H. G. M.  AGU  December 2010

    In the past decade or so, there have only been limited approaches formulated for the analysis of waveform LiDAR data. We illustrate how the Moment Distance (MD) framework can characterize the shape of the LiDAR waveforms using simple, computationally fast, geometric operations. We assess the relationship of the MD metrics to some key waveform landmarks - such as locations of peaks, power of returns, and pseudo-heights - using LVIS datasets acquired over a tropical forest in La Selva, Costa Rica in 1998 and 2005. We also apply the MD framework to 2003 LVIS data from Howland Forest, Maine. We also explore the effects of noise on the MD Index (MDI). Our results reveal that the MDI can capture important dynamics in canopy structure. Movement in the location of the peaks is detected by shifts in the MDI. Because this new approach responds to waveform shape, it is more sensitive to changes of location of peak returns than to the power of the return. Results also suggest a positive relationship between the MDI and the canopy pseudo-height.



  13. Producing Science-Ready radar datasets for the retrieval of forest 3D structure: Correcting for terrain topography and temporal changes
    Simard, M.; Lavalle, M. R. B. V. P. N. D. R. H. S. C. A. I.  AGU  December 2010

    We present the results of the 2009-2010 airborne L-band radar and lidar campaigns in boreal, temperate and tropical forests. The main objective is to improve canopy height and biomass retrieval from radar data both radiometrically and interferometrically. To achieve this, we assessed and designed models to compensate for the impact of terrain topography and temporal decorrelation on the radar data. The UAVSAR is an L-band radar capable of repeat-pass interferometry producing fully polarimetric images with a spatial resolution of 5m. The LVIS system is a laser altimeter providing a spatially dense sampling of full waveforms. The lidar data is used to determine radar scattering model parameters as well as validate model predictions. During the campaigns, we also collected weather as well as forest structure data in a total of 95 plots. First, we present science-ready UAVSAR datasets that are radiometrically corrected for terrain topography and vegetation reflectivity pattern. This is a critical step before accurate estimation of forest parameters. We implemented a generic and homomorphic transform that can also handle UAVSAR’s antenna steering capabilities which otherwise introduce significant distortions of the image radiometry. We show results obtained from the radiometric calibration. The improvements on the biomass retrieval are significant. Another method to estimate forest 3D structure is polarimetric interferometry (polinSAR). However, since UAVSAR is a repeat-pass interferometric system, changes in forest canopy between radar acquisitions tend to decorrelate successive images. To quantify temporal decorrelation, we collected four radar datasets within a period of 11 days. The data enabled quantification of the temporal decorrelation and its relationship to weather patterns. To compensate for temporal decorrelation, we developed a polinSAR inversion model that account for the target changes. The canopy height inversion is demonstrated through a forward model based on radar observations. The impact of terrain topography on the radar radiometry can be corrected and is required for any retrieval of forest biomass. The use of a generic radiometric correction for the canopy’s reflectivity pattern is sufficient, but prior knowledge of the forest type is important. We have made great progress in the systematic implementation of temporal decorrelation into polinSAR canopy height model. However, the knowledge of environmental variables is found to have the significant impact on our ability to retrieve canopy structure.



  14. Retrieval of Vegetation Structural Parameters and 3-D Reconstruction of Forest Canopies Using Ground-Based Echidna® Lidar
    Strahler, A. H.; Yao, T. Z. F. Y. X. S. C. W. C. E. J. D. L. C. D. N. G. L. J.  AGU  December 2010

    A ground-based, scanning, near-infrared lidar, the Echidna® validation instrument (EVI), built by CSIRO Australia, retrieves structural parameters of forest stands rapidly and accurately, and by merging multiple scans into a single point cloud, the lidar also provides 3-D stand reconstructions. Echidna lidar technology scans with pulses of light at 1064 nm wavelength and digitizes the full return waveform sufficiently finely to recover and distinguish the differing shapes of return pulses as they are scattered by leaves, trunks, and branches. Deployments in New England in 2007 and the southern Sierra Nevada of California in 2008 tested the ability of the instrument to retrieve mean tree diameter, stem count density (stems/ha), basal area, and above-ground woody biomass from single scans at points beneath the forest canopy. Parameters retrieved from five scans located within six 1-ha stand sites matched manually-measured parameters with values of R2 = 0.94-0.99 in New England and 0.92-0.95 in the Sierra Nevada. Retrieved leaf area index (LAI) values were similar to those of LAI-2000 and hemispherical photography. In New England, an analysis of variance showed that EVI-retrieved values were not significantly different from other methods (power = 0.84 or higher). In the Sierra, R2 = 0.96 and 0.81 for hemispherical photos and LAI-2000, respectively. Foliage profiles, which measure leaf area with canopy height, showed distinctly different shapes for the stands, depending on species composition and age structure. New England stand heights, obtained from foliage profiles, were not significantly different (power = 0.91) from RH100 values observed by LVIS in 2003. Three-dimensional stand reconstruction identifies one or more “hits” along the pulse path coupled with the peak return of each hit expressed as apparent reflectance. Returns are classified as trunk, leaf, or ground returns based on the shape of the return pulse and its location. These data provide a point cloud of hit locations, intensities, and object classes within a three-axis coordinate system centered at the instrument. Merging point clouds from overlapping scans produces the 3-D reconstruction, which can be used to measure individual DBH (R2 = 0.97, 0.99, n = 20, 15 trees, two Sierra Nevada sites) and tree height (R2 = 0.98, 0.98, n = 18, 16 trees, compared to LVIS RH100 values). The point clouds should also allow more realistic measurements of green and woody biomass as well as crown size and shape. They point the way toward measurements of directional gap probability for applications in radiative transfer modeling. A second-generation instrument, the Dual-Wavelength Echidna Lidar (DWEL), is currently under development by the Echidna Lidar Team at Boston University with NSF support. Supported by NASA grants NNG06GI92G and NNX08AE94A and NSF grant DBI-0923389.



  15. NASA's Operation IceBridge: using instrumented aircraft to bridge the observational gap between ICESat and ICESat-2 laser altimeter measurements (Invited)
    Studinger, M.; Koenig, L. M. S. S. J. G.  AGU  December 2010

    In 2009, the NASA satellite laser altimeter mission ICESat (Ice, Cloud and Land Elevation Satellite), which was launched in 2003, ceased to operate. To bridge the gap in polar laser observations between ICESat and its replacement ICESat-2, which is not scheduled for launch until 2015, Operation IceBridge, a six-year NASA airborne mission, was initiated in 2009. From a series of yearly polar flights, Operation IceBridge uses airborne instruments to map rapidly changing areas in the Arctic and Antarctic, building on two decades of repeat airborne and satellite measurements. Combined with previous aircraft observations, as well as ICESat, CryoSat-2 and the forthcoming ICESat-2 observations, Operation IceBridge will produce a cross-calibrated 17-year time series of ice sheet and sea-ice elevation data over Antarctica, as well as a 27-year time series over Greenland. These time series will be a critical resource for predictive models of sea ice and ice sheet behavior. In addition to laser altimetry, Operation IceBridge is using a comprehensive suite of instruments to produce a three-dimensional view of the Arctic and Antarctic ice sheets, ice shelves and the sea ice. The suite includes two NASA laser altimeters, the Airborne Topographic Mapper (ATM) and the Land, Vegetation and Ice Sensor (LVIS); four radar systems from the University of Kansas’ Center for Remote Sensing of Ice Sheets (CReSIS), a Ku-band radar altimeter, accumulation radar, snow radar and the Multichannel Coherent Radar Depth Sounder (MCoRDS); a Sander Geophysics airborne gravimeter (AIRGrav) and a high resolution stereographic camera (DMS). The first Operation IceBridge flights were conducted between March and May 2009 over the Arctic and between October and November 2009 over Antarctica. Since its start in 2009, Operation IceBridge has flown 69 science missions, 580 flight hours and collected more than 350,000 km of data. All Operation IceBridge data are available at NSDIC: http://nsidc.org/data/icebridge. Further information on Operation IceBridge is available at http://www.nasa.gov/icebridge or http://twitter.com/IceBridge.



  16. Mapping Canopy Height and Biomass Dynamics in the Sierra Nevada using Waveform Lidar
    Swatantran, A.; Dubayah, R. H. M. A. B. B.  AGU  December 2010

    In this study, we explored the use of multi-date waveform lidar in quantifying and mapping canopy dynamics over the Sierra National Forest in California. Structural metrics from the Laser Vegetation Imaging Sensor (LVIS) from 1999 and 2008 were compared with field measurements to detect changes in canopy height and biomass. Nearly co-incident footprints from both datasets were calibrated for geolocation shifts and canopy height transitions were analyzed. Changes were also directly calculated from LVIS metrics at plot (0.07ha) and hectare level (1 ha). Net positive changes in canopy height and biomass were observed in LVIS metrics at all scales as well as in field data. Height distributions in 2008 showed a greater shift towards steady state dynamics as compared to 1999 also supporting re-growth, however, both years were consistent with not being in steady state. Around 15.9 % of the total area showed significant biomass changes greater that 2Mg/ha/yr out of which 70% changes were positive. Similarly, 26% of the study area showed significant canopy height changes with 66% being positive. These results suggest that the landscape is regenerating from past disturbances with more growth than losses and could be a net carbon sink. Although changes from LVIS could not be validated with field observations, spatial patterns of large losses and gains were consistent with those on optical imagery. Areas identified with high canopy stress from hyperspectral data showed greater height losses and vice versa. Around 80% of the statistically significant height changes were visually identified as canopy losses on ground. This study provides further evidence that lidar data can be used to quantify large changes and analyze canopy transitions while emphasizing the need for rigorous sensor calibration and field validation. Results from this study can provide useful inputs for space borne lidar missions such as the Deformation Ecosystem Structure and Dynamics of Ice (DESDynI)



    17. Return to the top of the page