@techreport{oai:jaxa.repo.nii.ac.jp:00004554,
 author = {中山, 幹康 and Langner, Andreas and Nakayama, Mikiyasu},
 month = {},
 note = {Selective logging is applied throughout the tropics causing high levels of forest degradation. Sustainable forest management has the potential to conserve biomass but is not widely applied in the field. Reduced impact logging can promote sustainable forestry but for that sake the degradation stage of a forest must be evaluated. Given a certain forest type, above-ground biomass values can be deduced from that data. Due to the limitation of any kind of satellite sensor to directly measure the biomass especially in complex and carbon-dense ecosystems such as the humid tropics, we analyzed the application of ALOS PALSAR multi-mode L-band radar data to what extent canopy openings can be detected, which describe the degradation stage of a forest. Such information can then be used to deduce the above-ground biomass. In comparison to optical data, which are hampered by the frequent cloud cover in the tropics, radar sensors have the great advantage to derive data irrespective of weather conditions. In this study, Phased Array type L-band Synthetic Aperture Radar (PALSAR) level 1.5 data with 12.5m spatial resolution, 34.3 degree incidence angle and HH-HV polarization were analyzed for the purpose of forest degradation monitoring. For that sake, the different polarization modes were compared with each other. Study site was a production forest in Sabah, Malaysian Borneo. High resolution optical data from the Advanced Visible and Near Infrared Radiometer type 2 (AVNIR-2) and the Panchromatic Remote-sensing Instrument for Stereo Mapping (PRISM) sensor onboard the ALOS platform in combination with ground-truth data were used to validate the PALSAR data. Our results showed that in comparison to the HH data, the HV band shows a higher accuracy in separating degraded forests from pristine forests. In the HV polarization mode it was possible to distinguish highly degraded secondary forests with an open crown cover dominated by regrowth from pristine forests but it turned out that a separation of forests, which have been degraded using conventional selective logging techniques - the prevailing logging type in Borneo - was not feasible. Another limitation was caused by foreshortening and radar shadowing due to hilly terrain. Summarizing, it has to be stated that HV polarized L-band data from the PALSAR sensor can be used to screen for degradation processes occurring before forest conversion, but are too restricted to distinguish between pristine and conventionally logged-over forests. However, the application of PALSAR L-band data in combination with other sensor data, such as for example AVNIR-2 seems to be a very promising approach., 形態: カラー図版あり, 形態: DVD-ROM1枚, Physical characteristics: Original contains color illustrations, Note: One DVD-ROM, 資料番号: AA0065135084, レポート番号: JAXA-SP-11-007E},
 title = {ANALYSIS OF MULTI-MODE PALSAR DATA TO DETECT FOREST DEGRADATION IN THE HUMID TROPICS},
 year = {2011}
}