skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: An Examination of Diameter Density Prediction with k-NN and Airborne Lidar

Abstract

While lidar-based forest inventory methods have been widely demonstrated, performances of methods to predict tree diameters with airborne lidar (lidar) are not well understood. One cause for this is that the performance metrics typically used in studies for prediction of diameters can be difficult to interpret, and may not support comparative inferences between sampling designs and study areas. To help with this problem we propose two indices and use them to evaluate a variety of lidar and k nearest neighbor (k-NN) strategies for prediction of tree diameter distributions. The indices are based on the coefficient of determination ( R 2), and root mean square deviation (RMSD). Both of the indices are highly interpretable, and the RMSD-based index facilitates comparisons with alternative (non-lidar) inventory strategies, and with projects in other regions. K-NN diameter distribution prediction strategies were examined using auxiliary lidar for 190 training plots distribute across the 800 km 2 Savannah River Site in South Carolina, USA. In conclusion, we evaluate the performance of k-NN with respect to distance metrics, number of neighbors, predictor sets, and response sets. K-NN and lidar explained 80% of variability in diameters, and Mahalanobis distance with k = 3 neighbors performed best according to amore » number of criteria.« less

Authors:
 [1];  [2];  [3];  [4];  [1];  [4]
  1. Univ. of Washington, Seattle, WA (United States)
  2. Washington State Dept. of Natural Resources, Olympia, WA (United States)
  3. Univ. of Eastern Finland, Joensuu (Finland)
  4. Oregon State Univ., Corvallis, OR (United States)
Publication Date:
Research Org.:
USDA Forest Service-Savannah River, New Ellenton, SC (United States)
Sponsoring Org.:
USDOE Office of Environmental Management (EM), Acquisition and Project Management (EM-50)
OSTI Identifier:
1415439
Report Number(s):
17-05-P
Journal ID: ISSN 1999-4907; PII: f8110444
Grant/Contract Number:
AI09-00SR22188
Resource Type:
Journal Article: Accepted Manuscript
Journal Name:
Forests
Additional Journal Information:
Journal Volume: 8; Journal Issue: 11; Journal ID: ISSN 1999-4907
Publisher:
MDPI
Country of Publication:
United States
Language:
English
Subject:
60 APPLIED LIFE SCIENCES; forest inventory; dbh; diameter distribution; performance criteria

Citation Formats

Strunk, Jacob L., Gould, Peter J., Packalen, Petteri, Poudel, Krishna P., Andersen, Hans -Erik, and Temesgen, Hailemariam. An Examination of Diameter Density Prediction with k-NN and Airborne Lidar. United States: N. p., 2017. Web. doi:10.3390/f8110444.
Strunk, Jacob L., Gould, Peter J., Packalen, Petteri, Poudel, Krishna P., Andersen, Hans -Erik, & Temesgen, Hailemariam. An Examination of Diameter Density Prediction with k-NN and Airborne Lidar. United States. doi:10.3390/f8110444.
Strunk, Jacob L., Gould, Peter J., Packalen, Petteri, Poudel, Krishna P., Andersen, Hans -Erik, and Temesgen, Hailemariam. 2017. "An Examination of Diameter Density Prediction with k-NN and Airborne Lidar". United States. doi:10.3390/f8110444. https://www.osti.gov/servlets/purl/1415439.
@article{osti_1415439,
title = {An Examination of Diameter Density Prediction with k-NN and Airborne Lidar},
author = {Strunk, Jacob L. and Gould, Peter J. and Packalen, Petteri and Poudel, Krishna P. and Andersen, Hans -Erik and Temesgen, Hailemariam},
abstractNote = {While lidar-based forest inventory methods have been widely demonstrated, performances of methods to predict tree diameters with airborne lidar (lidar) are not well understood. One cause for this is that the performance metrics typically used in studies for prediction of diameters can be difficult to interpret, and may not support comparative inferences between sampling designs and study areas. To help with this problem we propose two indices and use them to evaluate a variety of lidar and k nearest neighbor (k-NN) strategies for prediction of tree diameter distributions. The indices are based on the coefficient of determination (R2), and root mean square deviation (RMSD). Both of the indices are highly interpretable, and the RMSD-based index facilitates comparisons with alternative (non-lidar) inventory strategies, and with projects in other regions. K-NN diameter distribution prediction strategies were examined using auxiliary lidar for 190 training plots distribute across the 800 km2 Savannah River Site in South Carolina, USA. In conclusion, we evaluate the performance of k-NN with respect to distance metrics, number of neighbors, predictor sets, and response sets. K-NN and lidar explained 80% of variability in diameters, and Mahalanobis distance with k = 3 neighbors performed best according to a number of criteria.},
doi = {10.3390/f8110444},
journal = {Forests},
number = 11,
volume = 8,
place = {United States},
year = 2017,
month =
}

Journal Article:
Free Publicly Available Full Text
Publisher's Version of Record

Save / Share:
  • This article presents the results of airborne lidar measurements of aerosol and polar stratospheric clouds (PSC) above Kiruna. Polarization measurements allow the distinction between volcanic aerosols, and PSC. They observed PSC formations near Kiruna on December 11, 1991, extending over 100's of km west and east.
  • During the period 1988-1990, studies of atmospheric aerosol have been made over the Atlantic. These include measurement programs out of Ascension Island (8{degrees}S, 14{degrees}W), the Azores (38{degrees}N, 25{degrees}W), Iceland (63{degrees}N, 23{degrees}W), and from the United Kingdom over the Northeast Atlantic. For these studies the equipment deployed included an airborne backscatter lidar (operating at 10.6 {mu}m), airborne particle-sounding probes, ground-based lidars (operating at 10.6, 0.53, and 0.35 {mu}m), balloon radiosondes, and a Sun-tracking photometer. In addition, standard meteorological information has been incorporated along with, when appropriate, data from the SAGE II limb-sounding satellite. The present paper thus introduces an overview ofmore » the program together with an outline of the technology, measurement characteristics, and performance of the airborne equipment that determined the strategic planning of much of the work. The measurements themselves, made in the relatively clean period for the atmosphere before the Mount Pinatubo eruption, will be presented in a subsequent series of papers. 26 refs., 5 figs., 3 tabs.« less
  • The authors discuss aircraft measurements in the northern polar vortex which show that in the winter of 1994/95 that the ozone levels showed a 50 percent decline in the February to March period. This observation was correlated with simultaneous observations of stratospheric aerosols which support the argument that this density depletion was not the result of a purely dynamic process.
  • The California Research at the Nexus of Air Quality and Climate Change (CalNex) and Carbonaceous Aerosol and Radiative Effects Study (CARES) field campaigns during May and June 2010 provided a data set appropriate for studying characteristics of the planetary boundary layer (PBL). The NASA Langley Research Center (LaRC) airborne High Spectral Resolution Lidar (HSRL) was deployed to California onboard the NASA LaRC B-200 aircraft to aid incharacterizing aerosol properties during these two field campaigns. Measurements of aerosol extinction (532 nm), backscatter (532 and 1064 nm), and depolarization (532 and 1064 nm) profiles during 31 flights, many in coordination with othermore » research aircraft and ground sites, constitute a diverse data set for use in characterizing the spatial and temporal distribution of aerosols, as well as the depth and variability of the daytime mixed layer (ML), which is a subset within the PBL. This work illustrates the temporal and spatial variability of the ML in the vicinity of Los Angeles and Sacramento, CA. ML heights derived from HSRL measurements are compared to PBL heights derived from radiosonde profiles, ML heights measured from ceilometers, and simulated PBL heights from the Weather Research and Forecasting Chemistry (WRF-Chem) community model. Comparisons between the HSRL ML heights and the radiosonde profiles in Sacramento result in a correlation coefficient value (R) of 0.93 (root7 mean-square (RMS) difference of 157 m and bias difference (HSRL radiosonde) of 5 m). HSRL ML heights compare well with those from the ceilometer in the LA Basin with an R of 0.89 (RMS difference of 108 m and bias difference (HSRL Ceilometer) of -9.7 m) for distances of up to 30 km between the B-200 flight track and the ceilometer site. Simulated PBL heights from WRF-Chem were compared with those obtained from all flights for each campaign, producing an R of 0.58 (RMS difference of 604 m and a bias difference (WRF-Chem HSRL) of -157 m) for CalNex and 0.59 (RMS difference of 689 m and a bias difference (WRF-Chem HSRL) of 220 m) for CARES. Aerosol backscatter simulations are also available from WRF15 Chem and are compared to those from HSRL to examine differences among the methods used to derive ML heights.« less
  • The authors describe a data analysis technique for determining gravity wave intrinsic parameters including wave propagation direction. The technique involves measuring the altitude variations of the wave-induced density perturbations of the atmospheric Na layer. This technique can be used with airborne lidars, multiple ground-based lidars, and steerable lidars. In this paper the technique is applied to airborne Na lidar data obtained during a roundtrip flight from Denver, Colorado, to the Pacific Coast in November 1986. During the flight, strong wave perturbations were observed in the Na layer near the Pacific Caost over a horizontal distance of nearly 700 km. Themore » intrinsic horizontal wavelength of this wave was estimated to be about 85 km, and the vertical wavelength was 4.1 km. The intrinsic period was about 102 min, and the propagation direction was almost due south.« less