Imputation of individual Longleaf Pine (Pinus palustris Mill.) Tree attributes from field and LiDAR data

C.A. Silva, A.T. Hudak, L.A. Vierling, E.L. Loudermilk, J.J. O'Brien, K. Hiers, S.B. Jack, C. Gonzalez-Benecke, c Lee, M.J. Falkowski, A. Khosravipour

Research output: Contribution to journalArticleAcademicpeer-review

24 Citations (Scopus)

Abstract

Light Detection and Ranging (LiDAR) has demonstrated potential for forest inventory at the individual-tree level. The aim in this study was to predict individual-tree height (Ht; m), basal area (BA; m2), and stem volume (V; m3) attributes, imputing Random Forest k-nearest neighbor (RF k-NN) and individual-tree-level-based metrics extracted from a LiDAR-derived canopy height model (CHM) in a longleaf pine (Pinus palustris Mill.) forest in southwestern Georgia, United States. We developed a new framework for modeling tree-level forest attributes that comprise 3 steps: (i) individual tree detection, crown delineation, and tree-level-based metrics computation from LiDAR-derived CHM; (ii) automatic matching of LiDAR-derived trees and field-based trees for a regression modeling step using a novel algorithm; and (iii) RF k-NN imputation modeling for estimating tree-level Ht, BA, and V and subsequent summarization of these metrics at the plot and stand levels. RMSDs for tree-level Ht, BA, and V were 2.96%, 58.62%, and 8.19%, respectively. Although BA estimation accuracy was poor because of the longleaf pine growth habitat, individual-tree locations, Ht, and V were estimated with high accuracy, especially in low-canopy-cover conditions. Future efforts based on the findings could help improve the estimation accuracy of individual-tree-level attributes such as BA.
Original languageEnglish
Pages (from-to)554-573
JournalCanadian journal of remote sensing
Volume42
Issue number5
DOIs
Publication statusPublished - 8 Aug 2016

Fingerprint

canopy
attribute
detection
modeling
forest inventory
basal area
stem
habitat

Keywords

  • METIS-317417
  • ITC-ISI-JOURNAL-ARTICLE

Cite this

Silva, C. A., Hudak, A. T., Vierling, L. A., Loudermilk, E. L., O'Brien, J. J., Hiers, K., ... Khosravipour, A. (2016). Imputation of individual Longleaf Pine (Pinus palustris Mill.) Tree attributes from field and LiDAR data. Canadian journal of remote sensing, 42(5), 554-573. https://doi.org/10.1080/07038992.2016.1196582
Silva, C.A. ; Hudak, A.T. ; Vierling, L.A. ; Loudermilk, E.L. ; O'Brien, J.J. ; Hiers, K. ; Jack, S.B. ; Gonzalez-Benecke, C. ; Lee, c ; Falkowski, M.J. ; Khosravipour, A. / Imputation of individual Longleaf Pine (Pinus palustris Mill.) Tree attributes from field and LiDAR data. In: Canadian journal of remote sensing. 2016 ; Vol. 42, No. 5. pp. 554-573.
@article{34f081a3928b44ef85bd38737cb1e027,
title = "Imputation of individual Longleaf Pine (Pinus palustris Mill.) Tree attributes from field and LiDAR data",
abstract = "Light Detection and Ranging (LiDAR) has demonstrated potential for forest inventory at the individual-tree level. The aim in this study was to predict individual-tree height (Ht; m), basal area (BA; m2), and stem volume (V; m3) attributes, imputing Random Forest k-nearest neighbor (RF k-NN) and individual-tree-level-based metrics extracted from a LiDAR-derived canopy height model (CHM) in a longleaf pine (Pinus palustris Mill.) forest in southwestern Georgia, United States. We developed a new framework for modeling tree-level forest attributes that comprise 3 steps: (i) individual tree detection, crown delineation, and tree-level-based metrics computation from LiDAR-derived CHM; (ii) automatic matching of LiDAR-derived trees and field-based trees for a regression modeling step using a novel algorithm; and (iii) RF k-NN imputation modeling for estimating tree-level Ht, BA, and V and subsequent summarization of these metrics at the plot and stand levels. RMSDs for tree-level Ht, BA, and V were 2.96{\%}, 58.62{\%}, and 8.19{\%}, respectively. Although BA estimation accuracy was poor because of the longleaf pine growth habitat, individual-tree locations, Ht, and V were estimated with high accuracy, especially in low-canopy-cover conditions. Future efforts based on the findings could help improve the estimation accuracy of individual-tree-level attributes such as BA.",
keywords = "METIS-317417, ITC-ISI-JOURNAL-ARTICLE",
author = "C.A. Silva and A.T. Hudak and L.A. Vierling and E.L. Loudermilk and J.J. O'Brien and K. Hiers and S.B. Jack and C. Gonzalez-Benecke and c Lee and M.J. Falkowski and A. Khosravipour",
year = "2016",
month = "8",
day = "8",
doi = "10.1080/07038992.2016.1196582",
language = "English",
volume = "42",
pages = "554--573",
journal = "Canadian journal of remote sensing",
issn = "0703-8992",
publisher = "Taylor & Francis",
number = "5",

}

Silva, CA, Hudak, AT, Vierling, LA, Loudermilk, EL, O'Brien, JJ, Hiers, K, Jack, SB, Gonzalez-Benecke, C, Lee, C, Falkowski, MJ & Khosravipour, A 2016, 'Imputation of individual Longleaf Pine (Pinus palustris Mill.) Tree attributes from field and LiDAR data' Canadian journal of remote sensing, vol. 42, no. 5, pp. 554-573. https://doi.org/10.1080/07038992.2016.1196582

Imputation of individual Longleaf Pine (Pinus palustris Mill.) Tree attributes from field and LiDAR data. / Silva, C.A.; Hudak, A.T.; Vierling, L.A.; Loudermilk, E.L.; O'Brien, J.J.; Hiers, K.; Jack, S.B.; Gonzalez-Benecke, C.; Lee, c; Falkowski, M.J.; Khosravipour, A.

In: Canadian journal of remote sensing, Vol. 42, No. 5, 08.08.2016, p. 554-573.

Research output: Contribution to journalArticleAcademicpeer-review

TY - JOUR

T1 - Imputation of individual Longleaf Pine (Pinus palustris Mill.) Tree attributes from field and LiDAR data

AU - Silva, C.A.

AU - Hudak, A.T.

AU - Vierling, L.A.

AU - Loudermilk, E.L.

AU - O'Brien, J.J.

AU - Hiers, K.

AU - Jack, S.B.

AU - Gonzalez-Benecke, C.

AU - Lee, c

AU - Falkowski, M.J.

AU - Khosravipour, A.

PY - 2016/8/8

Y1 - 2016/8/8

N2 - Light Detection and Ranging (LiDAR) has demonstrated potential for forest inventory at the individual-tree level. The aim in this study was to predict individual-tree height (Ht; m), basal area (BA; m2), and stem volume (V; m3) attributes, imputing Random Forest k-nearest neighbor (RF k-NN) and individual-tree-level-based metrics extracted from a LiDAR-derived canopy height model (CHM) in a longleaf pine (Pinus palustris Mill.) forest in southwestern Georgia, United States. We developed a new framework for modeling tree-level forest attributes that comprise 3 steps: (i) individual tree detection, crown delineation, and tree-level-based metrics computation from LiDAR-derived CHM; (ii) automatic matching of LiDAR-derived trees and field-based trees for a regression modeling step using a novel algorithm; and (iii) RF k-NN imputation modeling for estimating tree-level Ht, BA, and V and subsequent summarization of these metrics at the plot and stand levels. RMSDs for tree-level Ht, BA, and V were 2.96%, 58.62%, and 8.19%, respectively. Although BA estimation accuracy was poor because of the longleaf pine growth habitat, individual-tree locations, Ht, and V were estimated with high accuracy, especially in low-canopy-cover conditions. Future efforts based on the findings could help improve the estimation accuracy of individual-tree-level attributes such as BA.

AB - Light Detection and Ranging (LiDAR) has demonstrated potential for forest inventory at the individual-tree level. The aim in this study was to predict individual-tree height (Ht; m), basal area (BA; m2), and stem volume (V; m3) attributes, imputing Random Forest k-nearest neighbor (RF k-NN) and individual-tree-level-based metrics extracted from a LiDAR-derived canopy height model (CHM) in a longleaf pine (Pinus palustris Mill.) forest in southwestern Georgia, United States. We developed a new framework for modeling tree-level forest attributes that comprise 3 steps: (i) individual tree detection, crown delineation, and tree-level-based metrics computation from LiDAR-derived CHM; (ii) automatic matching of LiDAR-derived trees and field-based trees for a regression modeling step using a novel algorithm; and (iii) RF k-NN imputation modeling for estimating tree-level Ht, BA, and V and subsequent summarization of these metrics at the plot and stand levels. RMSDs for tree-level Ht, BA, and V were 2.96%, 58.62%, and 8.19%, respectively. Although BA estimation accuracy was poor because of the longleaf pine growth habitat, individual-tree locations, Ht, and V were estimated with high accuracy, especially in low-canopy-cover conditions. Future efforts based on the findings could help improve the estimation accuracy of individual-tree-level attributes such as BA.

KW - METIS-317417

KW - ITC-ISI-JOURNAL-ARTICLE

U2 - 10.1080/07038992.2016.1196582

DO - 10.1080/07038992.2016.1196582

M3 - Article

VL - 42

SP - 554

EP - 573

JO - Canadian journal of remote sensing

JF - Canadian journal of remote sensing

SN - 0703-8992

IS - 5

ER -