MODIS NDVI time-series allow the monitoring of Eucalyptus plantation biomass Guerric le Maire a, b, ⁎, Claire Marsden a, c , Yann Nouvellon a, d , Clovis Grinand e , Rodrigo Hakamada f , José-Luiz Stape g , Jean-Paul Laclau a, h a CIRAD, UMR Eco&Sols, 2 Place Viala, 34060 Montpellier, France b CIRAD, UMR TETIS, Maison de la Télédétection, 34093 Montpellier Cedex 5, France c SupAgro, UMR Eco&Sols, 2 Place Viala, 34060 Montpellier, France d USP, Atmospheric Sciences Department, Rua do Matão 1226, 05508-090 São Paulo, Brazil e IRD, UMR Eco&Sols, 2 Place Viala, 34060 Montpellier, France f International Paper do Brasil, Rodovia SP 340, Km 171, 13.840-970 Mogi Guaçu, SP, Brazil g Department of Forestry and Environmental Sciences, North Carolina State University, Raleigh, NC 27695, United States h USP, Ecology Department, Rua do Matão 321, 05508-900, São Paulo, Brazil abstract article info Article history: Received 23 February 2011 Received in revised form 23 May 2011 Accepted 29 May 2011 Available online 1 July 2011 Keywords: Aboveground biomass Moderate Resolution Imaging Spectroradiometer CBERS WorldClim Brazil Forest Fast-growing plantations The use of remote sensing is necessary for monitoring forest carbon stocks at large scales. Optical remote sensing, although not the most suitable technique for the direct estimation of stand biomass, offers the advantage of providing large temporal and spatial datasets. In particular, information on canopy structure is encompassed in stand reflectance time series. This study focused on the example of Eucalyptus forest plantations, which have recently attracted much attention as a result of their high expansion rate in many tropical countries. Stand scale time-series of Normalized Difference Vegetation Index (NDVI) were obtained from MODIS satellite data after a procedure involving un-mixing and interpolation, on about 15,000 ha of plantations in southern Brazil. The comparison of the planting date of the current rotation (and therefore the age of the stands) estimated from these time series with real values provided by the company showed that the root mean square error was 35.5 days. Age alone explained more than 82% of stand wood volume variability and 87% of stand dominant height variability. Age variables were combined with other variables derived from the NDVI time series and simple bioclimatic data by means of linear (Stepwise) or nonlinear (Random Forest) regressions. The nonlinear regressions gave r-square values of 0.90 for volume and 0.92 for dominant height, and an accuracy of about 25 m 3 /ha for volume (15% of the volume average value) and about 1.6 m for dominant height (8% of the height average value). The improvement including NDVI and bioclimatic data comes from the fact that the cumulative NDVI since planting date integrates the interannual variability of leaf area index (LAI), light interception by the foliage and growth due for example to variations of seasonal water stress. The accuracy of biomass and height predictions was strongly improved by using the NDVI integrated over the two first years after planting, which are critical for stand establishment. These results open perspectives for cost-effective monitoring of biomass at large scales in intensively-managed plantation forests. © 2011 Elsevier Inc. All rights reserved. 1. Introduction International agreements and regional markets now give an economical value to forest carbon stocks in order to encourage countries to increase or maintain these stocks for climate mitigation purposes. Such economical assessments require precise and reliable methods to estimate forest carbon stocks at large scales, for instance through the development of remote-sensing applications, which has become an active research field (Baker et al., 2010; Goetz et al., 2009). The use of satellite-based estimations of carbon biomass is a promising solution in terms of (i) cost- and time-effectiveness compared to large-scale field inventories, (ii) integration of forest spatial variability in regional synopses of carbon stocks, and (iii) reactivity to high-impact disturbances such as deforestation and afforestation. Different types of imagery have been used to assess forest carbon biomass from remote sensing. Interferometric radar and lidar data are the most promising techniques for forest biomass estimation, and it is recognized that optical imagery cannot reach the same level of accuracy (Patenaude et al., 2005). However, the immediate need for biomass and biomass change estimates across long time-periods and large scales cannot be satisfied by active remote-sensing techniques, meaning that optical imagery solutions are still essential (Powell et al., 2010). The main limitation with optical remote-sensing is its rapid Remote Sensing of Environment 115 (2011) 2613–2625 ⁎ Corresponding author at: CIRAD, UMR Eco&Sols, 2 Place Viala, 34060 Montpellier, France. Tel.: +33 4 67 54 87 64; fax: +33 4 67 54 87 00. E-mail address: guerric.le_maire@cirad.fr (G. le Maire). 0034-4257/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.rse.2011.05.017 Contents lists available at ScienceDirect Remote Sensing of Environment journal homepage: www.elsevier.com/locate/rse