Evaluating the performance of the MODIS Leaf Area Index (LAI) product over a Mediterranean dryland planted forest M. SPRINTSIN†, A. KARNIELI*†, P. BERLINER†, E. ROTENBERG‡, D. YAKIR‡ and S. COHEN§ †Jacob Blaustein Institute for Desert Research, Ben Gurion University of the Negev, Sede Boker Campus 84990, Israel ‡Weizmann Institute of Science, Rehovot 76100, Israel §Institute of Soil, Water and Environmental Sciences, ARO Volcani Center, Bet Dagan 50250, Israel The launch of the Moderate Resolution Imaging Spectroradiometer (MODIS) onboard the Terra and Aqua satellites improved the ability to evaluate several surface biophysical parameters, including Leaf Area Index (LAI), which is pro- vided as an operational MODIS product, available at 1-km spatial resolution and at 8-day intervals. However, for heterogeneous and sparse planted forests that are common to the semi-arid eastern Mediterranean region, the data at low spatial resolution may be significantly biased by the contribution of different background elements to the total surface reflectance received by the sensor and cannot there- fore correctly reflect the real forest phenology. In the current paper the perfor- mance of the MODIS LAI product was examined over a dryland Mediterranean forest in southern Israel. The study found a significant discrepancy between ground-based and MODIS LAI datasets. In general, MODIS LAI values were c.51% of the ground-based LAI measurements. In addition ground based LAI peaked in the summer due to the natural growth cycle of the pine trees, while MODIS values peaked in the winter. The MODIS seasonal course could be explained by the development of annuals and crypto- and micro-phytes in the understorey and the clearing areas during the mid winter months that are included in the MODIS LAI product but not in the ground based measurements. However, for that period MODIS estimates should have exceeded ground-based estimates while in fact they were still lower. The relationship between MOD12C1 Land Cover Type 3 and MOD15A2 products is discussed. 1. Introduction Leaf Area Index (LAI) is one of the main factors driving eco-physiological processes responsible for forest growth, and is used as an input to various ecosystem models. Ground-based methods of LAI estimation are widely used; however, they yield accurate information only for the close vicinity of the measured point. Consequently, satellite remote sensing systems have been proposed as a solution for extensive spatial LAI assessment (e.g. Andersen et al. 2002, Colombo et al. 2003). Currently, LAI is globally and routinely provided by NASA as an operational Moderate Resolution Imaging Spectroradiometer (MODIS) product at 1-km spatial resolution and at 8-day intervals. However, such low spatial resolution data, for which each pixel may be made up of several land cover types, might be of limited *Corresponding author. Email: karnieli@bgu.ac.il International Journal of Remote Sensing ISSN 0143-1161 print/ISSN 1366-5901 online # 2009 Taylor & Francis http://www.tandf.co.uk/journals DOI: 10.1080/01431160903032885 International Journal of Remote Sensing Vol. 30, No. 19, 10 October 2009, 5061–5069 Downloaded By: [Ben Gurion University of the Negev] At: 20:34 23 September 2009