Tropical forest parameters retrieved from airborne multifrequency radar images
An AIRSAR image acquired in 1993 over the Paracou area in French Guiana is used to explore the capability of the airborne multifrequency SAR to retrieve important tropical forest parameters, in relation with silvicultural practices. The test-site is located in a primary tropical dense forest and includes twelve permanent plots for which dendrometric and position data have been acquired on every single tree of dbh (diameter at breast height) greater than 10cm, once every year since 1984. Nine of the twelve plots (250 m x 250 m each) underwent different silvicultural treatments in 1986, while the other three were left untouched and are used as reference plots. The plots and their responses to the different perturbations were then followed during measurement campaigns every year. The data gathered are here used as ground truth to which indices extracted from radar images are compared. In particular, we investigate the extent to which the difference in levels of woody biomass between the treated and untouched forest plots can be detected or quantified. It is found that even the relatively moderate relief of the area cannot be ignored and that better correlations are obtained with indices based on band ratio (for exemple PHV/CHV) which tend to highly reduce the relief-induced radiometric distorsions. Sensitivitv to biomass levels of more than 300 tons/ha are found. The results suggest that the combination of a very low frequency such as P-band, combined to a higher frequency acquired simultaneously, could be useful for the monitoring of tropical forests.
| Main Authors: | , , , |
|---|---|
| Format: | conference_item biblioteca |
| Language: | eng |
| Published: |
IEEE
|
| Subjects: | U30 - Méthodes de recherche, K10 - Production forestière, F62 - Physiologie végétale - Croissance et développement, forêt tropicale, forêt primaire, productivité primaire, biomasse, télédétection, radar, relevé aérien, Aéronef, http://aims.fao.org/aos/agrovoc/c_24904, http://aims.fao.org/aos/agrovoc/c_28112, http://aims.fao.org/aos/agrovoc/c_34328, http://aims.fao.org/aos/agrovoc/c_926, http://aims.fao.org/aos/agrovoc/c_6498, http://aims.fao.org/aos/agrovoc/c_24071, http://aims.fao.org/aos/agrovoc/c_156, http://aims.fao.org/aos/agrovoc/c_233, http://aims.fao.org/aos/agrovoc/c_3093, http://aims.fao.org/aos/agrovoc/c_3081, |
| Online Access: | http://agritrop.cirad.fr/390376/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | An AIRSAR image acquired in 1993 over the Paracou area in French Guiana is used to explore the capability of the airborne multifrequency SAR to retrieve important tropical forest parameters, in relation with silvicultural practices. The test-site is located in a primary tropical dense forest and includes twelve permanent plots for which dendrometric and position data have been acquired on every single tree of dbh (diameter at breast height) greater than 10cm, once every year since 1984. Nine of the twelve plots (250 m x 250 m each) underwent different silvicultural treatments in 1986, while the other three were left untouched and are used as reference plots. The plots and their responses to the different perturbations were then followed during measurement campaigns every year. The data gathered are here used as ground truth to which indices extracted from radar images are compared. In particular, we investigate the extent to which the difference in levels of woody biomass between the treated and untouched forest plots can be detected or quantified. It is found that even the relatively moderate relief of the area cannot be ignored and that better correlations are obtained with indices based on band ratio (for exemple PHV/CHV) which tend to highly reduce the relief-induced radiometric distorsions. Sensitivitv to biomass levels of more than 300 tons/ha are found. The results suggest that the combination of a very low frequency such as P-band, combined to a higher frequency acquired simultaneously, could be useful for the monitoring of tropical forests. |
|---|