Current Research Foci
Study of the High Park Fire using lidar and hyperspectral imagery
2012-2013 RAPID Response to the High Park Fire, Larimer County, CO. NSF [PI]
2013-2016 Spatial Dynamics of Burn Severity and Post-Fire Recovery in the High Park Fire Burn Area. NSF [PI]
Forest Structure Mapping and Application using of Historic Aerial Photography
Past Research Foci
Structure and Biomass from Waveform Lidar
My dissertation focused on demonstrating the methods and applications of the SLICER sensor, one of the earliest waveform sampling lidar sensors. In that work, I developed and validated algorithms for estimating height, aboveground biomass and the vertical distribution of plant area index from waveforms, and provided a basic exploration of the constraints on canopy geometry that allow consistent relationships between canopy height metrics and aboveground biomass. After moving to the USFS Pacific Northwest Research Station, I was funded by NASA to determine the consistency of height-biomass relationships plots selected across a gradient of stand composition and productivity at five locations (True Fir at Mt Rainier, Ponderosa Pine at the Metolius Forest in Oregon, Douglas-fir / Western Hemlock at HJ Andrews Experimental Forest and in the Coast Range of Oregon, and Spruce-Fir at Cascade Head, OR). We found consistent results for height-biomass relationships across this gradient – the first result of this kind. We also demonstrated similar results for a dataset that combined Douglas-fir/Western Hemlock stands at H.J. Andrews, Boreal forest stands at the BOREAS sites, and eastern deciduous forest stands at SERC, Maryland. In related work, I presented a method for three-dimensional analysis of forests.
While in the PNW, I had access to a unique dataset collected due to the forethought of David Harding, a long term collaborator. While SLICER collected waveforms over footprints 10 and 25 m in diameter, one collection was made at higher altitude using 50 , footprints. This dataset was the best approximation to the waveforms to be collected by the ICESat/GLAS mission which were expected to be 65 m in width. Using that data, I determined that it was possible to estimate forest height using these large footprints using a new algorithm Although that work was never published, we used it to secure funding from NASA to demonstrate this approach using early waveforms from ICESat/GLAS. The earlier algorithm used both waveform characteristics and ancillary information from digital elevation models, a revised algorithm eliminated the dependence on the DEMs, a necessary step for GLAS waveforms to be used globally (where DEMs were not available). As part of this work, I wrote a successful proposed to join the ICESat Science Team.
Over the next 8 years, I refined the algorithm using field measurements coincident with GLAS waveforms in the Amazon and United States and from airborne discrete-return data from a variety of forested biomes. Estimates of canopy were developed using this algorithm and then used in multiple studies of data fusion between GLAS, spaceborne SAR instruments and optical sensors. I was also involved in a detailed study of the use of model-based inference using GLAS waveforms as a data source.
1992-1995 NASA Earth System Science Fellowship
1998-2001 Integration of Landsat, Lidar, Ground, and Environmental Data to Improve Characterizations of Forest Structure and Composition. NASA, Terrestrial Ecology Program NASA [Co-PI]
2004-2006 Using the Geoscience Laser Altimeter System to Predict Forest Height and Biomass. NASA [PI]
2005-2006 Integration of GLAS and Airborne LIDAR Data with Field Data to Characterize the Vertical Structure and Loading of Forest Fuels . NASA [Advisee Fellowship]
2005-2008 A Global Forest Canopy Height and Vertical Structure Product from the Geoscience Laser Altimeter System NASA [PI]
2006-2008 Spatially-Explicit Estimates of Forest Biomass in the Amazon Basin using MODIS and the Geoscience Laser Altimeter System NASA [PI]
2009-2012 Estimates of Above-Ground Biomass from Lidar and L-Band Radar in the Amazon Basin. NASA [Co-PI / Senior Personnel]
2011-2012 GLAS Reprocessing and Biomass Equations for the Carbon Management System NASA [PI]
2012-2013 Forest Inventory and Analysis Support. NASA JPL [Co-PI / Senior Personnel]
2013-2014 The Global Forest Biomass Survey. USFS
Spaceborne Single Photon Lidar for Forest Structure and Biomass
Based on my work with ICESat, I wrote successful proposals to join the ICESat-2 Ad-hoc Science Definition Team and subsequently to join the formal Science Definition Team. In the course of working through the system design process, the importance of the vegetation mission was subsumed to those of the cryosphere mission. Ultimately, system design decisions significantly reduced the utility of the planned observations for terrestrial ecology. My PhD student David Gwenzi was able to develop an algorithm to estimate canopy heights for the California oak savanna vegetation type. However, we concluded that this was probably close to an optimal vegetation type for use with ICESat-2. The single photon ATLAS sensor on ICESat-2 will emit and receive a low number of signal photons and will record a large number of background noise photons. As a consequence, the system will struggle to observe the tops of needleleaf trees in vegetation types such as boreal forest, while in broad leaf forests with high cover it will struggle to make observations of terrain. In the savannas used in this work, there are widely spaced canopies that have high cover. High cover canopies allow observation of canopy height and terrain cam be observed in the wide spaces between them. We concluded that ICESat-2/ATLAS estimates of height are likely to have uncertainty similar to that of ICESat/GLAS.
(2009-2011 Terrestrial Ecology Science Definition and Planning for the ICESat-II Satellite Mission NASA [Co-PI / Senior Personnel]
2012-2015 A Proposal to Participate in the ICESat-II Science Definition Team. NASA [PI]
2012-2015 Canopy Structure from Micropulse LIDAR NASA [Co-PI / Senior Personnel]
Carbon Management using Airborne Lidar
[TBD]
2005-2007 Linking Landscape-Scale Carbon Monitoring with Forest Management USFS [Co-PI / Senior Personnel]
2005-2007 Airborne Laser Terrain Mapping. TNC [Co-PI / Senior Personnel]
Development of flash lidar technology
[TBD]
2008-2010 Ecosystem Science Applications Of The Electronically Steerable Flash Ladar (Esfl). NASA [Co-PI / Senior Personnel]
2011-2013 An Advanced Imaging LIDAR for Forest Carbon Studies. NASA [Co-PI / Senior Personnel]
Mapping with Landsat in Tropical Environments
[TBD]
2004-2006 Characterizing forested landscapes in the Caribbean Islands using Landsat ImageryUSFS [Co-PI / Senior Personnel]
2011-2017 Remote Sensing Of Tropical Forest Disturbance: The Challenges Of Clouds & Phenology USFS [Co-PI / Senior Personnel]
Michael Lefsky 