Context and objectivesMonitoring of permanent meadows at the region level is closely linked to the knowledge of environmental parameters and indicators on the state of the canopy. In the case of permanent meadows, these variables are characterized by an important spatial and temporal heterogeneity. This feature is connected to the diversity of management systems and to plants-ruminants interactions which are not easily controlled. In practice, this diversity makes the evaluation of the impact of agro-environmental measures difficult.
The objective of this study is to show that remote sensing and hyperspectral analysis allow to answer those needs by supplying a continuous spatial and temporal monitoring of parameters that characterize the canopy structure of every grassland parcel as well as its biochemical or biophysic properties. It is based on the study of relationships between spectral data and ground truth observations collected over 30 representative permanent meadows located in Belgian Lorraine.
Expected scientific results
The analysis of hyperspectral images was realized following 2 approaches: (i) the calculation of spectral indices, combining spectral responses of narrowbands to characterise grass canopy; (ii) the use of statistical analyses to determine the wavelengths which characterize best each biophysical and biochemical grassland parameter.
To reduce background noise and non relevant data, additional mathematical treatments like the calculation of the first reflectance derivative and the reduction of the number of useful wavelengths were realized.
The results of this study showed the potentialities of hyperspectral imagery for the characterization of grasslands. The quality of relations between physico-chemical parameters and various spectral components (e.g. reflectance curve, first derivative, spectral indices) allowed establishing a discrimination between the various types of meadows (pasture, mowed meadows, etc.). They allowed also to estimate the quality of grass canopy and so to establish regional inventories on grass production potentialities, which constitute a potential decision tool for local field actors.
Finally, this study also showed the interest of data collected with the CASI sensor compared with the SASI sensor. Qualitative differences observed in results can be explained by the geometrical quality of images received by the project and the SASI sensor sensibility regard to background noises generated by environment (eg. soil, atmosphere).