Vol. 11, No. 1, 64-73, 2012

Abstract
The foliage chlorophyll content can be used as one of the indicators of tree stress caused by adverses oil conditions with heavy metal contamination being the major stress factor. Linear regression models showing the relation between spectral indexes derived from spectroscopic measurements of Norway Spruce foliage and its chlorophyll content were successfully applied in previous studies. Silver Birch (Betulapendula Roth), a pioneer tree species, and Scots Pine (Pinus sylvestris L.) used for revegetation after mining activities are typical tree species in the mine reclamation area of the Sokolov Basin (West Bohemia, Czech Republic). In August 2010 hyperspectral images with a resolution of 5 x 5 m² were acquired with the HyMap sensor. In four selected areas with a different level of soil contamination samples of Silver Birch leaves and Scots Pine needles were collected during the field campaign. Spectral reflectance curves of samples were measured in a laboratory with an ASD Field Spec3 spectrometer using the contact probe. The goal of this study is to prove the correlation between spectral reflectance data and the chlorophyll content determined spectrophotometrically in the laboratory using different spectral indices (MCARI, TCARI / OSAVI, mNDVI₇₀₅, ANMB_650-725) and to find a mathematical description of this relation. The suitability of different indices for application on coniferous foliage of pine and broad leaved birch is discussed. Although the relations between different indices and the chlorophyll content show similar trends, the ANMB_650-725 index revealed the best results regarding the statistical significance. While the sought relation between spectral indices and the chlorophyll content showed to be statistically significant in the case of Scots Pine, it was rather weak and thus not applicable in the case of Silver Birch. The present work is a part of a study aiming to create a methodology for chlorophyll determination in Silver Birch and Scots Pine from available hyperspectral data.