• DocumentCode
    1004713
  • Title

    Forest Height Inversion Using High-Resolution P-Band Pol-InSAR Data

  • Author

    Garestier, Franck ; Dubois-Fernandez, Pascale C. ; Champion, Isabelle

  • Author_Institution
    CESBIO, Toulouse
  • Volume
    46
  • Issue
    11
  • fYear
    2008
  • Firstpage
    3544
  • Lastpage
    3559
  • Abstract
    In this paper, a high-resolution P-band Pol-InSAR data set acquired by the airborne RAMSES system over pine forest stands of different height is investigated. A significant penetration depth in all the polarimetric channels and a wide range of polarimetric phase center heights are observed, attesting of an interaction of the radar waves with different forest structural elements. The main objective of this paper concerns forest height inversion at P-band. First, forest-modeling assumptions are evaluated using a priori information, such as ground-level and forest height measurements. The full extend of the forest height is shown to be responsible of the volume decorrelation, and a significant orientation effect is clearly identified over the highest stands. As a consequence, the Oriented Volume over Ground model (OVoG) is determined to be the most appropriated model for the forest height inversion. At P-band, the ground contribution is present in all the polarimetric channels due to the important penetration at this frequency. To overcome this difficulty, a time-frequency optimization method based on sublook decomposition is developed to separate the pure ground and canopy contributions, allowing forest height estimation with OVoG with an rms error on the order of 2 m. In the last section of this paper, a sensitivity analysis of the inversion with respect to two important system parameters, the signal-to-noise ratio and the resolution, is presented, leading to a discussion on the inversion robustness in spaceborne conditions, where these system parameters are the most deteriorated as compared to airborne configurations.
  • Keywords
    airborne radar; height measurement; radar interferometry; radar polarimetry; synthetic aperture radar; vegetation; Oriented Volume over Ground model; airborne RAMSES system; canopy contributions; forest height inversion; forest structural elements; frequency 427 MHz; high-resolution P-band Pol-InSAR data; pine forest stands; polarimetric phase center heights; radar waves; signal-to-noise ratio; time-frequency optimization method; Decorrelation; Ground penetrating radar; Optimization methods; Radar polarimetry; Robustness; Sensitivity analysis; Signal resolution; Signal to noise ratio; Spaceborne radar; Time frequency analysis; Forestry; interferometry; polarimetric radar; radar imaging;
  • fLanguage
    English
  • Journal_Title
    Geoscience and Remote Sensing, IEEE Transactions on
  • Publisher
    ieee
  • ISSN
    0196-2892
  • Type

    jour

  • DOI
    10.1109/TGRS.2008.922032
  • Filename
    4685935