Assessment of Spatial Resolution in Estimating Leaf Area Index from Satellite Images: A Case Study with Avicennia Marina Plantations in Thailand

Abstract

This study aims to test whether a well-established method of high resolution remote sensing can be used for estimating the mangrove leaf area index (LAI) of the Avicennia marina plantation in Thailand. The hemispherical photographs of sixty-two A. marina plantation plots in Bangpu, Samut Prakan, Thailand were used for in situ leaf area index (LAI) calculation using CAN-EYE software. The location of each plot was recorded by dual frequency GPS receiver. The geo-referenced QuickBird image was resampled to 2.5, 5, 10, 15, 20, 25, and 30 meter pixel sizes, then were calculated for five Vegetation Indexes to predict LAI. Regression analysis showed a good agreement of the field LAI and models. The relationship between field LAI versus model derived from GVI, NDVI and EVI which resample to 10 meter pixel size yielding R2 =0.797, 0.796 and 0.794 respectively. The GVI model produced the lowest error when compared against the independent field data (RMSE=0.188). The results confirmed that a well-established method of high resolution remote sensing can be used for estimating the mangrove LAI of the A. marina plantation in Thailand. Additionally, the results also suggested possibility to use moderate spatial resolution satellite imagery with about 10-meter ground resolution for estimating the mangrove LAI with lower cost than the sub-meter high resolution data.