Forest monitoring by using L-band coherence

Manabu Watanabe, Motoyuki Sato, Masanobu Shimada, Kazuo Ouchi, Haipeng Wang

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

Two ALOS/PALSAR polarimetry data taken in 2006 were used and examined the deforestation stage by using coherence value. Many trees were fallen by a typhoon hit on September 8, 2004 at a test site in Japan. Reforestation has started for the deforestation sites and several kinds of deforestation stages are appeared there. We classified the stands into four groups. 1) Only small stumps and branches are left. 2) Tree trunks have been carried out, but branches and stumps have been left. 3) The fallen trees have been left. 4) No trees were fallen by the typhoon (normal forest). Coherence values of HH-HH and W-W become -0.39 for group 1 and -0.30 for group 4, and show a significant difference. But the coherence values of group 2, and group 3 are -0.3 and show almost same value as group 1. The coherence value of HH-VV and VV-HH are 0.1 lower than the one of the HH-HH and VV-VV. The coherence values, which include cross polarization, such as HV-HV and HV-VV, show no difference for all groups and the value is 0.2 to 0.3. While coherence values show the difference between group 1 and group 4, σ° shows wider dynamic range and is preferable to distinguish the deforested stands from forest stands.

Original languageEnglish
Title of host publication2007 Asia-Pacific Microwave Conference, APMC
DOIs
Publication statusPublished - 2007 Dec 1
EventAsia-Pacific Microwave Conference, APMC 2007 - Bangkok, Thailand
Duration: 2007 Dec 112007 Dec 14

Publication series

NameAsia-Pacific Microwave Conference Proceedings, APMC

Other

OtherAsia-Pacific Microwave Conference, APMC 2007
CountryThailand
CityBangkok
Period07/12/1107/12/14

Keywords

  • Coherence
  • Forest monitoring
  • L-band SAR
  • Polarimetry

ASJC Scopus subject areas

  • Engineering(all)

Fingerprint Dive into the research topics of 'Forest monitoring by using L-band coherence'. Together they form a unique fingerprint.

Cite this