Accuracy Vs. Operability: a Case Study Over Barrax In The Context Of The DEMETER Project

The use of a physical approach in an Earth Observation based model for surface energy budget calculation leads to a complex parameterization, and hence to multiple needs in inputs. These inputs concern ground and atmospheric variables that can be measured or estimated with various techniques at various levels of difficulty. It is however possible to get most of them during major field campaigns. In the context of the DEMETER project, the demonstration of a routinely usage of such approaches implies to take into account with operability constraints. A new Multi‐Scales Energy Balance System was developed to include a panel of possible algorithms to obtain intermediate results involved in the evaporative fraction calculation. It is therefore possible to use either external observations, obtained with advanced methodologies and fitted to the area and period of simulation, or different surrogates, depending on availability of canopy geometric and/or radiometric observations. As it cannot be possible, in operational context, to get detailed, validated inputs continuously, the quality of the resulting evaporative fraction may vary from a simulation to an other. It is hence necessary to quantify the impact of the variability of quality in input on the accuracy of the final results. The experiment presented is based on different scenarios that are assumed to be representative of different operational contexts, and illustrates problems that can occur for practical use of evaporative fraction in irrigation management, and the balance between accuracy and operability.