Calibration and characterisation of four chlorophyll meters and transmittance spectroscopy for non-destructive estimation of forest leaf chlorophyll concentration

Abstract

Chlorophyll meters enable efficient and non-destructive estimation of leaf chlorophyll concentration (LCC), but require calibration against destructively-determined values to provide an absolute quantity that is comparable between different studies and species. Well-established instruments can provide accurate LCC estimates, but the performance of recent low-cost devices is less clear. Questions also remain over the choice of generic or species-specific calibration functions. Additionally, little attention has been paid to transmittance spectroscopy, which offers substantially increased spectral sampling, as a potential alternative. We investigated the well-established Konica Minolta SPAD-502+ and Opti-Sciences CCM-200 instruments and the low-cost atLEAF CHL PLUS and PhotosynQ MultispeQ V1.0 devices for non-destructive estimation of forest LCC. We calibrated each chlorophyll meter against destructively-determined LCC values for eight temperate deciduous broadleaf forest species, and characterised relationships between the different instruments. We also assessed whether transmittance spectroscopy could provide improved accuracy. All investigated chlorophyll meters demonstrated similarly strong relationships with destructively-determined LCC, indicating that once calibrated, even the low-cost devices represent a suitable choice for non-destructive forest LCC estimation. With the exception of oak, chlorophyll meter – LCC relationships were consistent between species, indicating that for these species, a generic calibration function may be suitable depending on required accuracy. Specifically, LCC values provided by the generic calibration functions fell within the prediction uncertainties of species-specific calibration functions for most considered species. The generic calibration functions explained between 2% and 16% less variation in LCC than the species-specific calibration functions, resulting in a mean increase in RMSE (NRMSE) of just 0.01 g m−2 to 0.02 g m−2 (2% to 5%). Transmittance spectroscopy was able to provide improved performance over the chlorophyll meters, indicating that they may miss some relevant spectral information at blue and green wavelengths. However, this improved performance comes at the expense of reduced practicality in the field.