Improving the Tropical Forest Remote Sensing Services for the Dem. Rep. of Congo inside the EU FP7 ''RECOVER'' Project
Haarpaintner, Jörg¹; Kohling, Miguel¹; Enssle, Fabian²; Pedrazzani, Donata³; Mane, Landing^4
1Norut, NORWAY; ²Albert-Ludwigs University Freiburg, GERMANY; ³GMV, SPAIN; ⁴OSFAC, CONGO, THE DEMOCRATIC REPUBLIC OF THE

ReCover is a 3-year EU FP 7 project (Nov. 2010 - Oct. 2013) and consortium led by VTT Technical Research Centre of Finland, aiming to develop and improve science based remote sensing GMES services to support tropical forest management and activities to reduce emission from deforestation and forest degradation (REDD) in the tropical region. The Observatoir Satellitale des Forets d'Afrique Centrale (OSFAC) is among the users of this service development representing the case of the Democratic Republic of Congo (DRC). Through a specific Service Level Agreement (SLA), OSFAC has requested the following products: a) optical and SAR image mosaics, b) optical and SAR based forest / land cover maps, c) biomass maps, d) forest area change maps and e) forest degradation maps. A first iteration of the products a) and b) for the years 1990, 2000, 2005 and 2010 and the biomass product c) for the period 2003-2009, has been delivered to OSFAC in spring 2012 by the responsible team composed by its lead, Norut (Norway), and partners, GMV (Spain) and Albert-Ludwigs-Universität Freiburg (ALU-FR, Germany). These products have been assessed by the user. This initial approach was based on a first data assessment and off-the-shelf methods with only negligible regional knowledge and an overall accuracy assessment using VHR data, without a detailed validation. The second iteration products, due in summer 2013, will be an improvement through a critical review and validation process and by further method development and research, like a prior statistical data analysis considering temporal and especially seasonal variability. Also pre-processing methods of both optical and SAR data have been improved meanwhile. In this paper we will 1) report on the product and user assessment of the first iteration products (Fig.1), 2) give a summary of the methods and improvements that were applied for the second iteration of the service products a, b, and c, 3) investigate the interoperability between optical and SAR sensors, and 4) present and assess the methods of the new service products of (semi-) decadal forest change (e) and forest degradation (f) for the periods 1990-2000 and 2000-2005-2010. The main challenge of these decadal change products is the fact that the single year products are based on different optical from Landsat and ALOS AVNIR-2, as well as different SAR sensors and modes from ERS-2 SAR, Envisat ASAR and ALOS Palsar.
Fig. 1: Optical and SAR image mosaics (upper images) and their derived forest (green) / non-forest (beige) maps (lower images) for indicated years with water in blue. The reference image mosaic is from Google Earth and indicates the Kompsat-2 VHR scenes (red) and Alos AVNIR-2 (blue). The reference forest map is the 2005-2010 forest change map from the FACET Atlas [OSFAC, 2010]