Project Funding: 2012 - 2015

NRA: 2011 NASA: Carbon Monitoring System   

Funded by NASA

Abstract:
We propose to implement a novel approach for mapping forest biomass and associated errors using the fusion of airborne LiDAR, passive optical and thermal data and a Bayesian hierarchical model that accounts for spatial variances between ground observations and remotely sensed data. This method will be compared with the more traditional approach of using a variety of plot-scale LiDAR metrics in a generalized, multiple linear regression model for relatively large region of interest (e.g., county- or state- scale). Also, we will use fine-resolution LiDAR and passive optical data (<1 m) to delineate individual trees, identify species class, and derive additional tree-level attributes (e.g., crown dimensions, crown area weighted heights, stem density) to improve upon biomass estimates made with aggregated point cloud metrics and inventory data at the plot-level (the traditional approach). These three methods will be evaluated and compared at four study sites in the midAtlantic and New England regions of the eastern US: Howland Forest and Holt Research Forest, ME; Harvard Forest, MA; and the Smithsonian Environmental Research Center near Edgewater, MD. This study will leverage coincident and co-registered LiDAR, passive optical, and thermal data that were collected at these sites for NASA s local-scale biomass pilot project between 2011 and 2012. Remotely sensed data was collected with Goddard s LiDAR, Hyperspectral, and Thermal (G-LiHT) airborne imager, which PI Cook developed at NASA-GSFC for studying the complex relationship between terrestrial ecosystem form and function. Large-area stem maps (3 to 35 ha per site, in which all stems greater than 1 cm have been measured) exist at each of these study sites, and these data will be used to verify crown delineations and enable the creation of a fineresolution spectral library. Subsets of the stem map areas will be used to simulate inventory plots, which will then be used as inputs for the Bayesian spatial latent factor model. Each of the stem map areas contain a variety of over/understory tree species, variable topography and range of drainage conditions, which will allow us to validate each of the methods over a wide range of forest types between and within each of the four study sites. Benefits of the proposed Bayesian spatial latent factor prediction model are 1) variables are selected using an efficient, dimension reduction technique; 2) spatial dependencies are incorporated into the model to and improve inference; 3) data compression is used to reduce the computational burden; and 4) sources of uncertainty are acknowledged and propagated through to prediction. Benefits of using data fusion for biomass mapping is that LiDAR and passive optical data provide unique information on the 3- dimensional structure and species composition of the forest, respectively. This synergy has been the focus of recent research, and has spawned the development of multi-instrument airborne systems such at the Carnegie Airborne Observatory (CAO), NASA s G-LiHT, and National Ecological Observatory Network (NEON) system that will begin systematic data collections in 2012. New algorithms and model variables for mapping forest biomass, such as the Bayesian latent spatial factor model and individual tree attributes we propose in this study, are needed to take full advantage of the synergy offered by these new, complementary datasets.

Publications:

Finley, A. O., Banerjee, S., Zhou, Y., Cook, B. D., Babcock, C. 2017. Joint hierarchical models for sparsely sampled high-dimensional LiDAR and forest variables. Remote Sensing of Environment. 190, 149-161. DOI: 10.1016/j.rse.2016.12.004

Datta, A., Banerjee, S., Finley, A. O., Gelfand, A. E. 2016. On nearest-neighbor Gaussian process models for massive spatial data. WIREs Computational Statistics. 8(5), 162-171. DOI: 10.1002/wics.1383

Salazar, E., Hammerling, D., Wang, X., Sanso, B., Finley, A. O., Mearns, L. O. 2016. Observation-based blended projections from ensembles of regional climate models. Climatic Change. 138(1-2), 55-69. DOI: 10.1007/s10584-016-1722-1

Babcock, C., Finley, A. O., Cook, B. D., Weiskittel, A., Woodall, C. W. 2016. Modeling forest biomass and growth: Coupling long-term inventory and LiDAR data. Remote Sensing of Environment. 182, 1-12. DOI: 10.1016/j.rse.2016.04.014

Datta, A., Banerjee, S., Finley, A. O., Gelfand, A. E. 2016. Hierarchical Nearest-Neighbor Gaussian Process Models for Large Geostatistical Datasets. Journal of the American Statistical Association. 111(514), 800-812. DOI: 10.1080/01621459.2015.1044091

Duncanson, L. I., Dubayah, R. O., Cook, B. D., Rosette, J., Parker, G. 2015. The importance of spatial detail: Assessing the utility of individual crown information and scaling approaches for lidar-based biomass density estimation. Remote Sensing of Environment. 168, 102-112. DOI: 10.1016/j.rse.2015.06.021

Rosette, J., Cook, B., Nelson, R., Huang, C., Masek, J., Tucker, C., Sun, G., Huang, W., Montesano, P., Rubio-Gil, J., Ranson, J. 2015. Sensor Compatibility for Biomass Change Estimation Using Remote Sensing Data Sets: Part of NASA's Carbon Monitoring System Initiative. IEEE Geoscience and Remote Sensing Letters. 12(7), 1511-1515. DOI: 10.1109/LGRS.2015.2411262

Goetz, S. J., Hansen, M., Houghton, R. A., Walker, W., Laporte, N., Busch, J. 2015. Measurement and monitoring needs, capabilities and potential for addressing reduced emissions from deforestation and forest degradation under REDD+. Environmental Research Letters. 10(12), 123001. DOI: 10.1088/1748-9326/10/12/123001

Finley, A. O., Banerjee, S., Cook, B. D. 2014. Bayesian hierarchical models for spatially misaligned data in R. Methods in Ecology and Evolution. 5(6), 514-523. DOI: 10.1111/2041-210X.12189

White, J. C., Wulder, M. A., Varhola, A., Vastaranta, M., Coops, N. C., Cook, B. D., Pitt, D., Woods, M. 2013. A best practices guide for generating forest inventory attributes from airborne laser scanning data using an area-based approach. The Forestry Chronicle. 89(06), 722-723. DOI: 10.5558/tfc2013-132

White, J.C., M. A. Wulder, A. Varhola, M. Vastaranta, N. C. Coops, B. D. Cook, D. Pitt, and M. Woods. 2013. A best practices guide for generating forest inventory attributes from airborne laser scanning data using the area-based approach. Information Report FI-X-10. Natural Resources Canada, Canadian Forest Service, Canadian Wood Fibre Centre, Pacific Forestry Centre, Victoria, BC. 50 p. http://cfs.nrcan.gc.ca/pubwarehouse/pdfs/34887.pdf

Duncanson, L. I., Cook, B. D., Hurtt, G. C., Dubayah, R. O. 2014. An efficient, multi-layered crown delineation algorithm for mapping individual tree structure across multiple ecosystems. Remote Sensing of Environment. 154, 378-386. DOI: 10.1016/j.rse.2013.07.044

Cook, B., Corp, L., Nelson, R., Middleton, E., Morton, D., McCorkel, J., Masek, J., Ranson, K., Ly, V., Montesano, P. 2013. NASA Goddard's LiDAR, Hyperspectral and Thermal (G-LiHT) Airborne Imager. Remote Sensing. 5(8), 4045-4066. DOI: 10.3390/rs5084045

Huang, W., Sun, G., Dubayah, R., Cook, B., Montesano, P., Ni, W., Zhang, Z. 2013. Mapping biomass change after forest disturbance: Applying LiDAR footprint-derived models at key map scales. Remote Sensing of Environment. 134, 319-332. DOI: 10.1016/j.rse.2013.03.017

Babcock, C., Matney, J., Finley, A. O., Weiskittel, A., Cook, B. D. 2013. Multivariate Spatial Regression Models for Predicting Individual Tree Structure Variables Using LiDAR Data. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing. 6(1), 6-14. DOI: 10.1109/JSTARS.2012.2215582

Finley, A. O., Banerjee, S., Cook, B. D., Bradford, J. B. 2013. Hierarchical Bayesian spatial models for predicting multiple forest variables using waveform LiDAR, hyperspectral imagery, and large inventory datasets. International Journal of Applied Earth Observation and Geoinformation. 22, 147-160. DOI: 10.1016/j.jag.2012.04.007

Montesano, P. M., Cook, B. D., Sun, G., Simard, M., Nelson, R. F., Ranson, K. J., Zhang, Z., Luthcke, S. 2013. Achieving accuracy requirements for forest biomass mapping: A spaceborne data fusion method for estimating forest biomass and LiDAR sampling error. Remote Sensing of Environment. 130, 153-170. DOI: 10.1016/j.rse.2012.11.016

2015 NASA Carbon Cycle & Ecosystems Joint Science Workshop Poster(s)

• G-LiHT: Multi-Sensor Airborne Image Data from Denali to the Yucatan   --   (Bruce Cook, Lawrence A Corp, Douglas Morton, Joel McCorkel)   [abstract]   [poster]

2013 NASA Terrestrial Ecology Science Team Meeting Poster(s)

• Lidar derived canopy height models of Harvard Forest   --   (Ian Paynter, Edward Saenz, Xiaoyuan Yang, Yan Liu, Zhuosen Wang, Crystal Schaaf, Zhan Li, Alan Strahler, Bruce Cook, Keith Krause, Nathan Leisso, Courtney Meier, Darius Culvenor, Glenn Newnham, David Jupp, Jenny Lovell, Ewan Douglas, Jason Martel, Supriya Chakrabarti, Timothy Cook, Glenn Howe, Kuravi Hewawasam, Jeffrey Thomas, Jihyun Kim, Shabnam Rouhani, Yun Yang, Nima Pahlevan, Qingsong Sun, Francesco Peri, Angela Erb)   [abstract]
• G-LiHT: Goddard’s LiDAR, Hyperspectral and Thermal Airborne Imager   --   (Bruce Cook, Lawrence Corp, Ross Nelson, Douglas Morton, Kenneth J Ranson, Jeffery Masek, Elizabeth Middleton)   [abstract]

More details may be found in the following project profile(s):

• NASA Terrestrial Ecology (TE) Project Profile
• North American Carbon Program (NACP) Project Profile
• NASA Carbon Monitoring System (CMS) Project Profile