Publications des agents du Cirad

Cirad

Validation of an empirical model, LAI~VI, to force a grass growth model on Reunion Island, France

Alexandre C., Tillard E., Salgado P., Lajoie G.. 2018. Photo Interpretation European Journal of Applied Remote Sensing, 54 (2) : p. 4-10.

The Leaf Area Index (LAI) is a parameter of many growth models used to predict biomass. LAI was used, for instance, in the Mosicas (Martiné, 1999) and Gamede (Vayssières et al., 2009) growth models for sugarcane and grass respectively, on Reunion Island, France. Those models have exhibited some limitations and prediction error can be significant. The aim of our study was to estimate LAI from satellite imaging in order to force a grass growth model. Around 430 samples were obtained from nine experimental plots situated around Saint-Pierre and Plaine-des-Cafres (South-West of the island) from April to August 2015. The sample LAI values were averaged by plot and associated with average vegetation indices (VI), computed from Spot5take5 data. Due to the testing of different VI we observed a stronger correlation between NDVI (Normalized Difference Vegetation Index) and LAI compared to other VI. The relation was so great that we found correlation values of 0.87 and 0.92 for temperate and tropical grass, respectively.

Mots-clés : réunion

Thématique : Méthodes mathématiques et statistiques; Physiologie végétale : croissance et développement; Méthodes de relevé

Documents associés

Article de revue

Agents Cirad, auteurs de cette publication :