First Assessment of the Capabilities of BIOMASS Tomographic Data for 3D Forest Structure Mapping

Abstract

Synthetic Aperture Radar Tomography (TomoSAR) reconstructs vertical profiles of the full 3D distribution of the backscattered radar power (or reflectivity) in natural media from SAR images acquired with a slight variation of look-angle [1]. In contrast to lidar profiles, the interpretation of TomoSAR reconstructions in terms of 3D forest structure (linked to the 3D size, location and arrangement of trees, trunks and branches in a stand) is today not fully established. The framework proposed in [2] attempts to interpret TomoSAR profile by means of two indices expressing heterogeneity in the horizontal (indicated with HS in the following) and vertical (VS) direction. These indices are defined empirically based on the spatial distribution of the (meaningful) peaks between ground and canopy top of all the TomoSAR reconstructions within a (larger) structure cell on ground. In particular, the heterogeneity in the horizontal direction can be inferred by evaluating the number of profile peaks in a top canopy layer. The close correlation of HS with the well-established stand density index and thus with basal area has been shown in [2]: large(r) values of HS indicate dense(r) stands, while low(er) values indicate sparse(r) stands. For evaluating vertical heterogeneity (sometimes also called complexity), VS is simply calculated as the product between the standard deviation of the unique peak heights and their number within the structure cell. An increasing VS, corresponds to an increasing heterogeneity (or complexity). The significance of this framework has been demonstrated only in airborne experiments in temperate [2], [3], and tropical forests [4] in correspondence with lidar and ground measurements. The tomographic phase of the ESA BIOMASS mission offers the unique possibility to extend 3D forest structure characterization at large scale as it provides for the first time fully polarimetric data stacks systematically, with penetration until the ground even in the densest forest stands guaranteed by the P-band waves. In this context, the objective of this work is to provide a first assessment of the capabilities of the framework proposed in [2] applied to BIOMASS reconstructions. The assessment will be carried out in correspondence of ground measurement or lidar acquisitions denoting different structure types. Three critical factors will be considered: 1. the 6 MHz bandwidth, limiting the TomoSAR vertical resolution reconstructions; 2. the presence of dielectric changes from pass-to-pass within a stack of BIOMASS acquisitions. This introduces a certain – non-structural – variability within the stack of BIOMASS acquisitions. According to experiments referred in the literature (see e.g. [5], [6]) P-band reflectivity variations in a 3-day revisit time may lead to defocusing in the TomoSAR profile reconstruction; 3. the strong ground contribution at P-band, which makes more difficult the interpretation of structure components close to the ground. These three factors are expected to limit the ground resolution of structure indices and to bias their values, with a reduced sensitivity to structure types and gradients. Polarizations might be critical to recover the dynamic range of the structure indices towards a better separation of structure types and reduce interpretation ambiguities. This additional point will be discussed in the presentation.

[1] A. Moreira, P. Prats-Iraola, M. Younis, G. Krieger, I. Hajnsek and K. P. Papathanassiou, “A tutorial on synthetic aperture radar,” in IEEE Geoscience and Remote Sensing Magazine, vol. 1, no. 1, pp. 6-43, March 2013.

[2] M. Tello, V. Cazcarra-Bes, M. Pardini and K. Papathanassiou, “Forest Structure Characterization From SAR Tomography at L-Band,” IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, vol. 11, no. 10, pp. 3402-3414, Oct. 2018.

[3] V. Cazcarra-Bes, M. Tello-Alonso, R. Fischer, M. Heym, and K. Papathanassiou, “Monitoring of For-est Structure Dynamics by Means of L-Band SAR Tomography,” Remote Sensing, vol. 9, no. 12, p. 1229, Nov. 2017.

[4] M. Pardini, M. Tello, V. Cazcarra-Bes, K. P. Papathanassiou and I. Hajnsek, “L- and P-Band 3-D SAR Reflectivity Profiles Versus Lidar Waveforms: The AfriSAR Case,” IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, vol. 11, no. 10, pp. 3386-3401, Oct. 2018.

[5] D. Ho Tong Minh, S. Tebaldini, F. Rocca and T. Le Toan, "The Impact of Temporal Decorrelation on BIOMASS Tomography of Tropical Forests," IEEE Geoscience and Remote Sensing Letters, vol. 12, no. 6, pp. 1297-1301, June 2015.

[6] Y. Bai, S. Tebaldini, D. H. T. Minh and W. Yang, "An Empirical Study on the Impact of Changing Weather Conditions on Repeat-Pass SAR Tomography," IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, vol. 11, no. 10, pp. 3505-3511, Oct. 2018.

[7] S. Tebaldini, "Algebraic Synthesis of Forest Scenarios From Multibaseline PolInSAR Data," IEEE Transactions on Geoscience and Remote Sensing, vol. 47, no. 12, pp. 4132-4142, Dec. 2009.