070c8d10-ffb8-11db-83ab-0014385046ad Sven Gilliams VITO NV Researcher +32 14 33 6827 +32 14 32 2795

Boeretang 200 Mol Antwerpen 2400 Belgium sven.gilliams@vito.be

2007-05-21T16:35:36 ISO 19115:2003/19139 1.0 WGS 1984 2006-06-28 Version 1.0 Map of the cropped areas, season 2005 (Meher) in Test-site 5. The map was generated purely based upon remote sensing data (optical and sar) and served as an input to the joint FAO/EC/WFP/CSA/MoARD workshop in Addis Abeba, 20-22 june 2006. Workshop on Calibration/Implementation Strategies for Agricultural Area Estimates of FAO and the government of Ethiopia, held from 20-22 of June 2006. Purpose of this workshop is to identify an improved method for area estimates in Ethiopia. The present map is preliminary and purely based upon the analysis of ASAR AP data, compared to older satellite imagery and ancillary data. There was no prior knowledge of the territory or what could be expected. The purpose is to identify the potential of the use of satellite imagery in contributing to area estimates in Ethiopia, comparing this to fieldwork as collected by the government of Ethiopia. As for the present map no calibration (field) data set has been available, it is expected that the classification can significantly be improved when making use of a fieldwork-based calibration data set. Lastly, it has to be noted that the current classification is based upon about 80% of the data, the remainder of the dataset has not been provided by ESA. Francesco Holecz Sarmap CEO +41 91 600 9366 +41 91 600 9369

Cascine di Barico Purasca Purasca 6989 Switzerland francesco.holecz@sarmap.ch

eth_site_7_8_s.png thumbnail png eth_site_7_8.png large_thumbnail png Ad hoc map SAR, agricultural area Test site Ethiopia Chore Gedesma No Use constraints provided appropriate acknowledgement is given. Unknown en farming 2005-06-01T16:30:00 2005-12-30T16:30:00 32.9918 47.98824 3.40667 14.88361 ASAR AP (Alternate Polarisation) data (spatial resolution =15 meters), acquired from May 2005 to December 2005, have been inputted to the sarmap classifier. ASAR AP acquisitions have been planned in order to have a site revisiting period of 14 to 35 days. In some cases, due to spacecraft related problems, the planned acquisitions have not been performed; this eventually affected the classification results, sometimes producing an underestimation of the cropped area. Yet the late start of the ASAR acquisitions, with respect to the crop season begin, has also caused an overall underestimation of the actual cropped area (this effect is very well visible in the upper right corner of this map) . Based upon the ASAR AP multi-temporal data series (images acquired on different dates during the crop season) and with the contribution of an Interferometric Coherence map generated from ASAR WS (Wide Swath) products (spatial resolution =100 meters), a decision-tree classification algorithm was run, and the present crop map was generated. It must be noted that, due to the intrinsic limitation of the SAR systems to properly imaging steep slopes, the reliability of such crop maps over site 2 should be considered limited as well. ftp://ftp.gmfs.info/Service/ETH/maps/GMFS_ETH_CA_SAR_7_8_2006.pdf WWW:LINK-1.0-http--link Available through the GMFS ftp site. Request login to Point of Contact