Christof J. Weissteiner

Exploring the Use of MODIS NDVI-Based Phenology Indicators for Classifying Forest General Habitat Categories

The cost effective monitoring of habitats and their biodiversity remains a challenge to date. Earth Observation (EO) has a key role to play in mapping habitat and biodiversity in general, providing tools for the systematic collection of environmental data. The recent GEO-BON European Biodiversity Observation Network project (EBONE) established a framework for an integrated biodiversity monitoring system. Underlying this framework is the idea of integrating in situ with EO and a habitat classification scheme based on General Habitat Categories (GHC), designed with an Earth Observation-perspective. Here we report on EBONE work that explored the use of NDVI-derived phenology metrics for the identification and mapping of Forest GHCs. Thirty-one phenology metrics were extracted from MODIS NDVI time series for Europe. Classifications to discriminate forest types were performed based on a Random Forests™ classifier in selected regions. Results indicate that date phenology metrics are generally more significant for forest type discrimination. The achieved class accuracies are generally not satisfactory, except for coniferous forests in homogeneous stands (77-82%). The main causes of low classification accuracies were identified as (i) the spatial resolution of the imagery (250 m) which led to mixed phenology signals; (ii) the GHC scheme classification design, which allows for parcels of heterogeneous covers, and (iii) the low number of the training samples available from field surveys. A mapping strategy integrating EO-based phenology with vegetation height information is expected to be more effective than a purely phenology-based approach.

Regional yield predictions of malting barley by remote sensing and ancillary data

Yield forecasts are of high interest to the malting and brewing industry in order to allow the most convenient purchasing policy of raw materials. Within this investigation, malting barley yield forecasts (Hordeum vulgare L.) were performed for typical growing regions in South-Western Germany. Multisensoral and multitemporal Remote Sensing data on one hand and ancillary meteorological, agrostatistical, topographical and pedological data on the other hand were used as input data for prediction models, which were based on an empirical-statistical modeling approach. Since spring barley production is depending on acreage and on the yield per area, classification is needed, which was performed by a supervised multitemporal classification algorithm, utilizing optical Remote Sensing data (LANDSAT TM/ETM+). Comparison between a pixel-based and an object-oriented classification algorithm was carried out. The basic version of the yield estimation model was conducted by means of linear correlation of Remote Sensing data (NOAA-AVHRR NDVI), CORINE land cover data and agrostatistical data. In an extended version meteorological data (temperature, precipitation, etc.) and soil data was incorporated. Both, basic and extended prediction systems, led to feasible results, depending on the selection of the time span for NDVI accumulation.

Enhancing Remotely Sensed Low Resolution Vegetation Data for Assessing Mediterranean Areas Prone to Land Degradation

An enhanced long term remote sensing based data set for Green Vegetation Fraction (GVF) was created for the Mediterranean area. The dataset contains 10-day composites of GVF for the time period 1989–2005 on a scale of 0.01°, covering the Mediterranean basin. The MEDOKADS data set was employed to create mixture triangles of NDVI and surface temperature, of which three abundances, the “vegetated”, “non-vegetated” and “cold” abundance were derived. The vegetated abundance was eventually converted to GVF. Compared to NDVI, clear improvements have been made for GVF, in particular in respect to the mitigation of undesired effects of bad atmospheric conditions. GVF can be derived in an almost fully operational way, which enables it as base data for monitoring vegetation and related purposes. The data has been successfully employed in two case studies on olive farming intensity and rural land abandonment.