Multiscale analysis of satellite-derived vegetation parameters for biogenic VOC emission modeling

Volatile organic compounds (VOC) emitted by vegetation play an important role in atmospheric chemistry contributing to tropospheric ozone and secondary organic aerosol formation. Quantification of biogenic VOC may be performed using emission modeling tools that require information on vegetation dynamics. For this purpose satellite-derived parameters such as Normalized Difference Vegetation Index (NDVI) and Leaf Area Index (LAI) are considered. However, spatial resolution of the satellite data could be an important source of uncertainties in biogenic VOC quantification. The main objective of the current study is to accomplish a sensitivity analysis of the biogenic emission modeling to the changes in data resolution of the satellite-derived vegetation parameters. This study is performed for an area of 80x 80km(2) in Portugal for 2011. Satellite observations provided by DEIMOS-1 (22 m resolution) and MODIS (250m and 1000m resolution) are analyzed for NDVI and LAI. Also, meteorological data from the Weather Research and Forecasting (WRF) model and detailed land cover data are considered by the emission quantification algorithm. Multi-scale analysis of LAI and NDVI was performed. Also, the modeling results are analyzed in terms of spatial and temporal variations of the emissions. The results confirm high sensitivity of the emission model to spatial resolution of the satellite-derived data resulting in about 30% difference in total isoprene emissions for the study area.