Potential of proximal hyperspectral imaging to estimate the spatial and temporal variability of soil microbial respiration

16th International Conference on Near Infrared Spectroscopy, La Grande-Motte, FRA, 02-/06/2013 - 07/06/2013; International audience; The objective of this study was to evaluate the potential of proximal hyperspectral imaging to (i) calibrate a model to predict soil respiration under controlled conditions, and (ii) to determine both spatial and temporal variations in soil respiration at a hundreds of micrometer scale. Calibration models based on spectra and respiration measured on standards were built then applied on hyperspectral images acquired on soil mesocosms (400 cm²), in order to predict their respiration pixel by pixel (0.2 mm²). The quality of the models was acceptable but their application to produce prediction maps faced some technical problems due to hyperspectral technology.