Estimation of airborne Lidar-derived tropical forest canopy height using landsat time series in Cambodia

Tetsuji Ota, Oumer S. Ahmed, Steven E. Franklin, Michael A. Wulder, Tsuyoshi Kajisa, Nobuya Mizoue, Shigejiro Yoshida, Gen Takao, Yasumasa Hirata, Naoyuki Furuya, Takio Sano, Sokh Heng, Ma Vuthy

Research output: Contribution to journalArticle

15 Citations (Scopus)

Abstract

In this study, we test and demonstrate the utility of disturbance and recovery information derived from annual Landsat time series to predict current forest vertical structure (as compared to the more common approaches, that consider a sample of airborne Lidar and single-date Landsat derived variables). Mean Canopy Height (MCH) was estimated separately using single date, time series, and the combination of single date and time series variables in multiple regression and random forest (RF) models. The combination of single date and time series variables, which integrate disturbance history over the entire time series, overall provided better MCH prediction than using either of the two sets of variables separately. In general, the RF models resulted in improved performance in all estimates over those using multiple regression. The lowest validation error was obtained using Landsat time series variables in a RF model (R2 = 0.75 and RMSE = 2.81 m). Combining single date and time series data was more effective when the RF model was used (opposed to multiple regression). The RMSE for RF mean canopy height prediction was reduced by 13.5% when combining the two sets of variables as compared to the 3.6% RMSE decline presented by multiple regression. This study demonstrates the value of airborne Lidar and long term Landsat observations to generate estimates of forest canopy height using the random forest algorithm.

Original languageEnglish
Pages (from-to)10750-10772
Number of pages23
JournalRemote Sensing
Volume6
Issue number11
DOIs
Publication statusPublished - Jan 1 2014

Fingerprint

forest canopy
lidar
tropical forest
Landsat
time series
multiple regression
canopy
date (time)
disturbance
prediction
history

All Science Journal Classification (ASJC) codes

  • Earth and Planetary Sciences(all)

Cite this

Estimation of airborne Lidar-derived tropical forest canopy height using landsat time series in Cambodia. / Ota, Tetsuji; Ahmed, Oumer S.; Franklin, Steven E.; Wulder, Michael A.; Kajisa, Tsuyoshi; Mizoue, Nobuya; Yoshida, Shigejiro; Takao, Gen; Hirata, Yasumasa; Furuya, Naoyuki; Sano, Takio; Heng, Sokh; Vuthy, Ma.

In: Remote Sensing, Vol. 6, No. 11, 01.01.2014, p. 10750-10772.

Research output: Contribution to journalArticle

Ota, T, Ahmed, OS, Franklin, SE, Wulder, MA, Kajisa, T, Mizoue, N, Yoshida, S, Takao, G, Hirata, Y, Furuya, N, Sano, T, Heng, S & Vuthy, M 2014, 'Estimation of airborne Lidar-derived tropical forest canopy height using landsat time series in Cambodia', Remote Sensing, vol. 6, no. 11, pp. 10750-10772. https://doi.org/10.3390/rs61110750
Ota, Tetsuji ; Ahmed, Oumer S. ; Franklin, Steven E. ; Wulder, Michael A. ; Kajisa, Tsuyoshi ; Mizoue, Nobuya ; Yoshida, Shigejiro ; Takao, Gen ; Hirata, Yasumasa ; Furuya, Naoyuki ; Sano, Takio ; Heng, Sokh ; Vuthy, Ma. / Estimation of airborne Lidar-derived tropical forest canopy height using landsat time series in Cambodia. In: Remote Sensing. 2014 ; Vol. 6, No. 11. pp. 10750-10772.
@article{81ad9fb6da98477a876c15aa6e3efab8,
title = "Estimation of airborne Lidar-derived tropical forest canopy height using landsat time series in Cambodia",
abstract = "In this study, we test and demonstrate the utility of disturbance and recovery information derived from annual Landsat time series to predict current forest vertical structure (as compared to the more common approaches, that consider a sample of airborne Lidar and single-date Landsat derived variables). Mean Canopy Height (MCH) was estimated separately using single date, time series, and the combination of single date and time series variables in multiple regression and random forest (RF) models. The combination of single date and time series variables, which integrate disturbance history over the entire time series, overall provided better MCH prediction than using either of the two sets of variables separately. In general, the RF models resulted in improved performance in all estimates over those using multiple regression. The lowest validation error was obtained using Landsat time series variables in a RF model (R2 = 0.75 and RMSE = 2.81 m). Combining single date and time series data was more effective when the RF model was used (opposed to multiple regression). The RMSE for RF mean canopy height prediction was reduced by 13.5{\%} when combining the two sets of variables as compared to the 3.6{\%} RMSE decline presented by multiple regression. This study demonstrates the value of airborne Lidar and long term Landsat observations to generate estimates of forest canopy height using the random forest algorithm.",
author = "Tetsuji Ota and Ahmed, {Oumer S.} and Franklin, {Steven E.} and Wulder, {Michael A.} and Tsuyoshi Kajisa and Nobuya Mizoue and Shigejiro Yoshida and Gen Takao and Yasumasa Hirata and Naoyuki Furuya and Takio Sano and Sokh Heng and Ma Vuthy",
year = "2014",
month = "1",
day = "1",
doi = "10.3390/rs61110750",
language = "English",
volume = "6",
pages = "10750--10772",
journal = "Remote Sensing",
issn = "2072-4292",
publisher = "Multidisciplinary Digital Publishing Institute (MDPI)",
number = "11",

}

TY - JOUR

T1 - Estimation of airborne Lidar-derived tropical forest canopy height using landsat time series in Cambodia

AU - Ota, Tetsuji

AU - Ahmed, Oumer S.

AU - Franklin, Steven E.

AU - Wulder, Michael A.

AU - Kajisa, Tsuyoshi

AU - Mizoue, Nobuya

AU - Yoshida, Shigejiro

AU - Takao, Gen

AU - Hirata, Yasumasa

AU - Furuya, Naoyuki

AU - Sano, Takio

AU - Heng, Sokh

AU - Vuthy, Ma

PY - 2014/1/1

Y1 - 2014/1/1

N2 - In this study, we test and demonstrate the utility of disturbance and recovery information derived from annual Landsat time series to predict current forest vertical structure (as compared to the more common approaches, that consider a sample of airborne Lidar and single-date Landsat derived variables). Mean Canopy Height (MCH) was estimated separately using single date, time series, and the combination of single date and time series variables in multiple regression and random forest (RF) models. The combination of single date and time series variables, which integrate disturbance history over the entire time series, overall provided better MCH prediction than using either of the two sets of variables separately. In general, the RF models resulted in improved performance in all estimates over those using multiple regression. The lowest validation error was obtained using Landsat time series variables in a RF model (R2 = 0.75 and RMSE = 2.81 m). Combining single date and time series data was more effective when the RF model was used (opposed to multiple regression). The RMSE for RF mean canopy height prediction was reduced by 13.5% when combining the two sets of variables as compared to the 3.6% RMSE decline presented by multiple regression. This study demonstrates the value of airborne Lidar and long term Landsat observations to generate estimates of forest canopy height using the random forest algorithm.

AB - In this study, we test and demonstrate the utility of disturbance and recovery information derived from annual Landsat time series to predict current forest vertical structure (as compared to the more common approaches, that consider a sample of airborne Lidar and single-date Landsat derived variables). Mean Canopy Height (MCH) was estimated separately using single date, time series, and the combination of single date and time series variables in multiple regression and random forest (RF) models. The combination of single date and time series variables, which integrate disturbance history over the entire time series, overall provided better MCH prediction than using either of the two sets of variables separately. In general, the RF models resulted in improved performance in all estimates over those using multiple regression. The lowest validation error was obtained using Landsat time series variables in a RF model (R2 = 0.75 and RMSE = 2.81 m). Combining single date and time series data was more effective when the RF model was used (opposed to multiple regression). The RMSE for RF mean canopy height prediction was reduced by 13.5% when combining the two sets of variables as compared to the 3.6% RMSE decline presented by multiple regression. This study demonstrates the value of airborne Lidar and long term Landsat observations to generate estimates of forest canopy height using the random forest algorithm.

UR - http://www.scopus.com/inward/record.url?scp=84912104905&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84912104905&partnerID=8YFLogxK

U2 - 10.3390/rs61110750

DO - 10.3390/rs61110750

M3 - Article

AN - SCOPUS:84912104905

VL - 6

SP - 10750

EP - 10772

JO - Remote Sensing

JF - Remote Sensing

SN - 2072-4292

IS - 11

ER -