Earthnet Home 30-Jul-2014  
Sessions and Session Summaries
First Annoucement
Scientific Committee
ESA Specific Links
Organising Committee
Round Table Discussion Questions
Conference Photos
List of Participants
All papers

POLinSAR Workshop 2003

Session Title: Forest

Chair(s): Dirk Hoekman and Irena Hajnsek

In the forest session four papers dealing with multi-parameter SAR have been presented. Three of them concern the biomass estimation using forest height or structural classification and one investigates the influence of topography on forest height estimation. The essential statements of the presentations are summarised in the following:

§         Polarimetry improves forest classification considerably (tropical biomass estimation requires prior forest type classification);

§         Pol-InSAR: forest height accuracy seems to be dependent on underlying topography and complexity vertical structure;

§         Dual-Frequency (X- and P-Band): good estimates under certain circumstances, and this was confirmed by LIDAR;

§         s0 saturates with biomass making a global biomass inventory impossible; Pol-InSAR forest height estimation shows to be related to biomass under certain circumstances (using allometric equations);

These statements have to be considered in relation with forest type. For boreal forests and plantations results for forest height and biomass estimation often have a high accuracy. For tropical forests height estimations are lacking, whereas biomass estimation has been done by means of forest structure classifications.

The current applications for quasi operational use with existing airborne and satellite sensors are land cover change and fast illegal logging response for tropical regions. For these applications there is a need for frequent observation and fast delivery from spaceborne systems, which is not sufficiently achieved by today?s satellite configurations. To improve the operational use and the estimation accuracy of the two mentioned applications multi-parameter radar systems would be desirable, including polarimetric systems and frequency diversity. For example L-band is the preferred frequency for observation of land cover changes and X-band for fast illegal logging response.

Several applications have been identified for global monitoring:

§         Forest regeneration and fire risk

§         Timber certification

§         Kyoto protocol (biomass estimation and change)

§         Biodiversity treaty

§         Wetland convention (RAMSAR)


Future multi-parameter satellite systems will have an important impact on forest parameter estimation accuracy because of additional frequency and polarisation capabilities. Full polarimetry capability and frequency diversity will considerably improve geophysical parameter estimation and trigger the capability for new applications. It has been proven that L-band is a frequency which covers a wide range of applications and is therefore the preferred frequency for the existing forest applications as well as for new developments.

There is a need for more airborne and spaceborne experiments with respect to forest type diversity. in order to prove the robustness and accuracy of the proposed methodologies on a global scale. Tropical forests are the most challenging ones, because of its generally complex structure and high density. It has been suggested to define a common test site over different forest areas (tropical, boreal, temperate) where multi-parameter radar data and ground measurements will be available for interested working groups.


Tropical Forest Mapping using Multiband Interferometric and Polarimetric SAR Data
Kemal Unggul Prakoso

Evaluating PolInSAR Tree Height and Topography Retrievals in Glen Affric
Dr Iain Woodhouse

 Vegetation Parameters using TOPSAR and GeoSAR Systems
Dr Scott Hensley

Above Ground Forest Biomass Estimation using Fully Polarimetric / Interferometric Radar Data
Tobias Mette


  Higher level                 Last modified: 09-Feb-2010