Determining Surface and Vegetation Heights form Interferometric Radar

Introduction

There is a critical need to measure and monitor land surface topography over large areas to assess the threat and impact of natural hazards such as flooding. Interferometric and stereo SAR data can be used to determine topography over large areas in the presence of clouds, but the superior vertical resolution of INSAR data provides the best option for making primary topographic measurements in many low-relief areas of the world. Although the capability of INSAR for mapping topography has been demonstrated, success has been primarily limited to areas where the surface is not obscured by significant vegetation. For many applications, topographic features of interest often occur in heavily vegetated regions, e.g. forests. Unfortunately, remotely sensed observations do not provide direct measurements of the true surface topography in vegetated areas, but instead yield a height Zs that depends on the sensor characteristics, the surface elevation Zg, and the vegetation height Zv, as shown in Fig. 1.

Vertical height accuracies of 2-5 m can be obtained in non-vegetated regions with airborne INSAR data processed to 10 m x 10 m terrain patches (pixels), but the presence of vegetation can lead to errors in the computed surface topography of tens of meters in forested areas. A method is needed to distinguish surface elevations Zg and vegetation heights Zv from Zs. A functional relationship has been derived to relate the INSAR observations to Zg and Zv using an electromagnetic scattering model [2]. Estimating the parameters Zg and Zv from the observations is then equivalent to inverting the model. To improve the estimates of the surface and vegetation heights from INSAR images we are developing two complementary methods: (1) exploiting spatial correlation in the images and (2) fusing LIDAR data with the INSAR data.

 Fig. 1. The relationship between the nominal height derived from INSAR Zs, ground elevation Zg, and vegetation height Zv.

A conference paper was accepted for the IEEE Southwest Symposium on Image Analysis and Interpretation 2000. (Abstract).

References

[2] R. N. Treuhaft and P. R. Siqueira, "Vertical structure of vegetated land surfaces from interferometric and polarimetric radar," Radio Science, vol. 35, no. 1, pp. 141-177, 1999.

Last Modified: Mon Sept 20, 1999
CSR/TSGC Team Web