Phenology related measures and indicators at varying spatial scales: investigation of phenology information for forest classification using SPOT VGT and MODIS NDVI data

Abstract

The main objective here is to investigate if leaf phenology indicators as derived from SPOT and MODIS NDVI time series can provide useful information for the detection, characterization and mapping of habitats, with specific reference to the General Habitat Category and Annex I (Natura 2000) schemes. The report is divided into three main parts. Part I ‘’Extraction and analysis of phenology indicators’’, using the Phenolo model of the Joint Research Centre (JRC, Ispra), and a phenological characterization and classification of test sites using Random Forest Classification, and an intercalibration of GHCs with MODIS-derived phenometrics. A set of 31 leaf phenology indicators (phenometrics) was extracted using JRC Phenolo model from time series of NDVI 10 day Maximum Value composites with 6 years of MODIS satellite data and 11 years of SPOT data. Classifications to discriminate deciduous and coniferous forest were performed in selected regions using MODIS satellite data. The main sources identified for low classification accuracy are both the large heterogeneity allowed by the GHC scheme for forests (tree cover proportion), and the low number of training points currently available from field survey. Part II ‘’Multi-temporal analysis of NDVI for grassland mapping and classification’’, focusses on two specific case study areas for grassland mapping. Part III ‘EO time series analysis to identify Annex I habitat types’ describing the processing of MODIS medium resolution time series with the HANTS algorithm and the use of it’s mathematical derived components for the identification of specific Annex I habitats, such as H9150 and H4060. A major limitation to identify those habitats is that the used times series have a minimum spatial resolution of 250 meters which can still be considered as too coarse for most habitats. Next to the fact that not all Annex I habitats have their unique phenology. A GHCs intercalibration strategy integrating EO-based pheno