Mapping U.S. forest biomass using nationwide forest inventory data and moderate resolution information

D. Salajanu f, D.L Weyennann g, K.C. Winterberger h, T.J. Brandeis f, R.L. Czaplewski i, R.E. McRoberts d, P.L. Patterson a, R.P. Tymcio a a Rocky Mtn. Research Station, 507 25th Street, Ogden, UT 84401, United States b Remote Sensing Applications Center, 2200 W 2300 S, Salt Lake City, UT 84119, United States C International Institute o/Tropical Forestry, Jardin Botimico Sur, 1201 Calle Ceiba, Rio Piedras, 00926, Puerto Rico d North Central Research Station, 1992 Folwell Ave, St. Paul, MN 55108, United States e Northeastern Research Station, 11 Campus Blvd, Newtown Square, PA 19073, United States f Southern Research Station, 4700 Old Kingston Pike, Knoxville, TN 37919, United States g Pacific Northwest Research Station, 1221 SW Yamhill, Portland, OR 97205, United States h Pacific Northwest Research Station, 3301 CSt, Anchorage, AK 99503, United States i Rocky Mtn Research Station, 240 W Prospect Rd, Fort Collins, CO 80526, United States Received 20 December 2005; received in revised form 24 August 2007; accepted 24 August 2007 A spatially explicit dataset of aboveground live forest biomass was made from ground measured inventory plots for the contenninous U.S., Alaska and Puerto Rico. The plot data are from the USDA Forest Service Forest Inventory and Analysis (PIA) program. To scale these plot data to maps, we developed models relating field-measured response variables to plot attributes serving as the predictor variables. The plot attributes came from intersecting plot coordinates with geospatial datasets. Consequently, these models serve as mapping models. The geospatial predictor variables included Moderate Resolution Imaging Spectrometer (MODIS)-derived image composites and percent tree cover; land cover proportions and other data from the National Land Cover Dataset (NLCD); topographic variables; monthly and annual climate parameters; and other ancillary variables. We segmented the mapping models for the U.S. into 65 ecologically similar mapping zones, plus Alaska and Puerto Rico. First, we developed a forest mask by modeling the forest vs. nonforest assignment of field plots as functions of the predictor layers using classification trees in See5©. Secondly, forest biomass models were built within the predicted forest areas using tree-based algorithms in Cubist©. To validate the models, we compared field-measured with modelpredicted forestlnonforest classification and biomass from an independent test set, randomly selected from available plot data for each mapping zone. The estimated proportion of correctly classified pixels for the forest mask ranged from 0.79 in Puerto Rico to 0.94 in Alaska. For biomass, model correlation coefficients ranged from a high of 0.73 in the Pacific Northwest, to a low of 0.31 in the Southern region. There was a tendency in all regions for these models to over-predict areas of small biomass and under-predict areas oflarge biomass, not capturing the full range in variability. Map-based estimates of forest area and forest biomass compared well with traditional plot-based estimates for individual states and for four scales of spatial aggregation. Variable importance analyses revealed that MODIS-derived infonnation could contribute more predictive power than other classes of infonnation when used in isolation. However, the true contribution of each variable is confounded by high correlations. Consequently, excluding anyone class of variables resulted in only small effects on overall map accuracy. An estimate oftotal C pools in live forest biomass of U.S. forests, derived from the nationwide biomass map, also compared well with previously published estimates. © 2007 Elsevier Inc. All rights reserved.