Moreover, three new hyperspectral satellite missions (i.e., Environmental Mapping and Analysis Program��EnMAP , Hyper-spectral Imager Suite��HISUI  and PRecursore IperSpettrale of the application mission��PRISMA [29,30]), of the German, Japanese and Italian Space Agencies, respectively, have recently started and the launches of these satellites have been programmed to take place in the near future.Every step of the processing of the remote data is focused on the improvement of the characterization of areas of interest [31,32]. In fact, the calibration [33,34], the atmospheric [35,36] and the geometric  corrections of the remote data are all devoted to improving the remote data accuracy. The correction of sun glint effects is also dedicated to reducing the uncertainties in the characterization of the water body in open sea and in coastal areas .
In particular the amount of the uncertainties, which are related to the calibration, can be evaluated to range from 5% to 15% as a function of sensor type [32,34] and the amount of the uncertainties, which are related to the atmospheric correction, can be evaluated at 1% to 2% as a function of spectral bands and surfaces type . Moreover the selection of the methodology, which characterizes and classifies of areas of interest, is performed in order to obtain the best accuracy of the products. In the literature the amount of the uncertainties, which are related to the characterization the optical water parameters of the coastal area obtained by using the radiative transfer theory, can be evaluated at 5% to 20% [23,39,40].
All these remote data processing steps are applied to the identified and the acquired data. Therefore, the accuracy with which remote data characterize a specific surface depends, in the beginning, on the characteristics of the sensor. The proposed FWHM methodology evaluates the accuracy as a function of a spectral characteristic of the remote sensor. This methodology is focused on the evaluation of the number Entinostat of the bands in the specific spectral range of the remote sensor. This methodology does not compare the capabilities of different remote sensors, because each remote sensor presents specific spectral characteristics (i.e., spectral range, spectral resolution, etc.). On the contrary this methodology explores the spectral characteristic of the each remote sensor (i.e., hyperspectral sensor or multispectral sensor, satellite sensor or airborne sensor) in order to lead the identification of the remote data which improve the characterization of a specific surface. In fact, this methodology was developed on one multispectral (i.e., Landasat5 Thematic Mapper��TM) and four hyperspectral (i.e., CHRIS acquired in mode 1 and mode 2, MIVIS and PRISMA) datasets.