Hyperspectral and multispectral satellite sensors for mapping chlorophyll content in a Mediterranean Pinus sylvestris L. plantation

A new generation of narrow-band hyperspectral remote sensing data offers an alternative to broad-band multispectral data for the estimation of vegetation chlorophyll content. This paper examines the potential of some of these sensors comparing red-edge and simple ratio indices to develop a rapid and cost-effective system for monitoring Mediterranean pine plantations in Spain. Chlorophyll content retrieval was analyzed with the red-edge R₇₅₀/R₇₁₀ index and the simple ratio R₈₀₀/R₅₆₀ index using the PROSPECT-5 leaf model and the Discrete Anisotropic Radiative Transfer (DART) and experimental approach. Five sensors were used: AHS, CHRIS/Proba, Hyperion, Landsat and QuickBird. The model simulation results obtained with synthetic spectra demonstrated the feasibility of estimating Ca+b content in conifers using the simple ratio R₈₀₀/R₅₆₀ index formulated with different full widths at half maximum (FWHM) at the leaf level. This index yielded a r²=0.69 for a FWHM of 30nm and r²=0.55 for a FWHM of 70nm. Experimental results compared the regression coefficients obtained with various multispectral and hyperspectral images with different spatial resolutions at the stand level. The strongest relationships where obtained using high-resolution hyperspectral images acquired with the AHS sensor (r²=0.65) while coarser spatial and spectral resolution images yielded a lower root mean square error (QuickBird r²=0.42; Landsat r²=0.48; Hyperion r²=0.56; CHRIS/Proba r²=0.57). This study shows the need to estimate chlorophyll content in forest plantations at the stand level with high spatial and spectral resolution sensors. Nevertheless, these results also show the accuracy obtained with medium-resolution sensors when monitoring physiological processes. Generating biochemical maps at the stand level could play a critical rule in the early detection of forest decline processes enabling their use in precision forestry.