Retrieval of Tropical Forest Height by means of SAR Tomography
Ho Tong Minh, Dinh1; Le Toan, Thuy1; Rocca, Fabio2; Tebaldini, Stefano2; Villard, Ludovic1
1Centre d’Etudes Spatiales de la Biosphere (CESBIO), FRANCE; 2Politecnico di Milano, ITALY

Forest height is one of the important parameters for assessing forest biomass at local scale for resource monitoring and at global scale for the global forest carbon stocks, which plays a key role in the global carbon cycle and hence in the global climate. While airborne LiDAR is used routinely for measuring forest height for local applications, there is a need to define spaceborne systems well adapted to forest height at global scale, particularly for tropical forests.

Longer wavelength Synthetic Aperture Radar (SAR) imaging is an important tool for forestry investigations, by virtue of the capability of lower frequency microwaves to penetrate the vegetation layer down to the ground. The availability of multiple baselines provides in addition an important advantage of this technique, which is the possibility to see the vertical structure of the vegetation through SAR Tomography (TomoSAR) (Reigber and Moreira, 2000). For this reason, TomoSAR techniques have been assessed in the last years over vegetated areas, as witnessed by papers recently published in this field (Tebaldini and Rocca, 2012; Ho Tong Minh et al.,).

In this paper, we present an approach based on TomoSAR to retrieve tropical forest height. TomoSAR enables to observe the whole forest vertical structure directly by exploiting the 3D nature of the forest data volume. It can separate the ground contribution in the radar backscatter measurements, can image the vegetation layer by layer and thus can reduce the topographic effects. The experiment results are based on the P-band polarimetric multi-baseline data, collected by ONERA over Paracou, French Guiana, in the frame of the TropiSAR ESA campaign (Dubois-Fernandez et al., 2012). We compare the forest height retrieved from SAR Tomography with LiDAR measurements. We are able to map forest height of 50 m resolution. The estimation appears to be reliable for vegetation layers ranging from 20 m to 35 m, consistent with tropical forest height. For this range height, standard deviation has been assessed in about 3 m.

While results are based on the airborne campaign data, the potential of the future spaceborne P-band SAR on board the BIOMASS mission with tomographic capability is prominent to provide high accuracy tropical forest height maps for global forest carbon monitoring (BIOMASS_report; Le Toan et al., 2011).


A. Reigber and A. Moreira, 2000, ''First demonstration of airborne SAR tomography using multibaseline l-band data,'' Geoscience and Remote Sensing, IEEETransactions on, pp. 2142-2152, Sep. 2000.

S. Tebaldini and F. Rocca, 2012, ''Multibaseline polarimetric sar tomography of a boreal forest at p- and l-bands,'' Geoscience and Remote Sensing, IEEETransactions on, vol. 50, no. 1, pp. 232 - 246, Jan 2012.

D. Ho Tong Minh, T. Le Toan, F. Rocca, S. Tebaldini, M. Mariotti d'Alessandro and L. Villard, 2012a, ''Relating P-Band SAR tomography to tropical forest biomass,'' Geoscience and Remote Sensing, IEEE Transactions on, Accepted for publications.

P. C. Dubois-Fernandez, T. Le Toan, S. Daniel, H. Oriot, J. Chave, L. Blanc,L. Villard, M. W. J. Davidson, and M. Petit, 2012, ''The tropisar airborne campaignin french guiana: Objectives, description, and observed temporal behavior of the backscatter signal,'' Geoscience and Remote Sensing, IEEE Transactionson, vol. PP, no. 99, pp. 1 -14, 2012.

''Report for Mission Selection: BIOMASS'', ESA SP-1324/1 (2012).

T. Le Toan, S. Quegan, M. Davidson, H. Balzter, P. Paillou, K. Papathanassiou, S. Plummer, F. Rocca, S. Saatchi, H. Shugart, and L. Ulander, 2011, ''The BIOMASS Mission : Mapping global forest biomass to better understand the terrestrial carbon cycle,'' Remote Sensing of Environment, pp. 2850-2860, Jun.2011.