During the four years of the "ASI-AGI" project (Analisi Sistemi Iperspettrali per le Applicazioni Geofisiche Integrate) funded by the Italian Space Agency (ASI), INGV developed specific algorithms and products for various geophysical applications to be applied to PRISMA mission (Hyper-spectral Precursor of the Application Mission) data launched on 21st March 2019. In this context, from 6 and 11 May 2014, the INGV carried out an analysis for the identification and characterization of a vicarious remote site for the CAL/VAL activities in support of ASI hyperspectral space missions. This campaign was conducted in Algeria in the Western Sahara region, in the Bechar province, more precisely in the area that goes from Ougarta and Beni Abbes. Furthermore, the identification of a CAL/VAL site could be also of direct interest to the European Copernicus programme with particular attention to the next hyperspectral mission. The identification of Beni Abbes area has benefited from the experience of INGV in the sector of CAL/VAL activities already conducted on Mt. Etna where 10 years of Hyperion data were analysed.

Vicarious calibration (VC) of satellite sensors is an important procedure for any space mission. In the coming years, the new technology of Spectral Imaging (SI) will be mounted in space, with several sensors already in place). Accordingly, VC of SI sensors will play a major role in the satellite operations, and therefore space agencies need to seek out new, appropriate target sites. We propose two sites in southern Israel that can be acquired in the same overpass: one for radiometric calibration and the second for spectral and thematic calibration. The first site is located at “Amiaz Playa” in the Dead Sea rift. The area is 4 km long and 6 km wide, characterized by a homogeneously bright playa made of silty carbonate. The second site is Makhtesh Ramon (MR), located 25 km southwest of Amiaz Playa. MR is one of the most interesting and powerful calibration sites for SI sensors. It is an eroded anticline with steep walls bound the valley, with friable sandstone at the bottom and more resistant limestone and dolomites at the top. The exposed area is rather flat and consists of pure mineralogical exposures of kaolinite, alunite, montmorillonite, calcite, dolomite, marl, gypsum, basalt, tuff, andesite, chert, sandstone, and iron oxides (hematite, goethite), which can all be found within short distances of each other. The area has minor vegetation coverage is mostly cloud-free throughout the year. In this paper, we will present the areas and their spectral and radiometric characteristics, along with some already gained Cal/Val experience.

Remote Visible / ShortWave InfraRed (VSWIR) imaging spectrometers map spectral radiance from 380 - 2500 nm (Schaepman et al., 2009). Surface reflectance features in this interval reveal the chemistry and composition of Earth’s terrestrial domains (Asner et al., 2017; Jetz et al., 2016; Ustin et al., 2004) and aquatic environments (Fichot et al., 2015; Hochberg, 2011). Future space-based spectroscopy missions by multiple nations will challenge the community to upscale airborne techniques, integrating data from multiple spectrometers and locations to address global-scale questions about ecosystem change and biodiversity. However, this data synthesis requires careful uncertainty quantification (UQ) and uncertainty propagation encompassing calibration, atmospheric correction, and the subsequent estimation of ecosystem properties. This can prevent biases from confounding trends due to latitudinal change in observation geometry, difficult regional atmospheric conditions, seasonal atmospheric patterns, or diverging instrument performance, any of which could lead to incorrect scientific conclusions at the global scale. We demonstrate methods by which instrument measurement uncertainty and atmospheric correction uncertainties can be estimated, validated, reported and used on a per-acquisition or per-spectrum basis, enabling principled data synthesis for global cross-platform spectroscopy.

Acknowledgements: A portion of this research was performed at the Jet Propulsion Laboratory, California Institute of Technology. We acknowledge the support of the JPL and Caltech Presidents’ and Directors’ Fund Program. We thank other supporting sponsors including the NASA Earth Science Division for the HyspIRI preparatory campaign, the AVIRIS-NG instrument and the data analysis program NNH16ZDA001N-AVRSNG. Copyright 2019 California Institute of Technology. US Government Support Acknowledged.

An important aspect to using imaging spectrometer data is the radiometric characterization and calibration of the sensors and validation of their data products and doing so with error budgets with known traceability. The radiometric accuracy of a given sensor is important for demonstrating the expected quality of data from the sensor. Known traceability allows data from multiple sensors to be directly comparable as will become more important in the near future with the expected launches of multiple imaging spectrometers from multiple countries, agencies, and commercial entities. The current work describes the state of pre- and post-launch radiometric absolute and relative uncertainties and their role in harmonising on-orbit data. Examples of prelaunch uncertainties based on the calibration of EnMAP and the calibration planned for the CLARREO Pathfinder Mission are presented highlighting recent work in the area of detector-based approaches using tunable laser sources. Post-launch calibration approaches for Pathfinder, EnMAP, CHIME, and DESIS including traditional vicarious calibration methods and the challenges of working with commercial data are presented. The vicarious calibration discussion relies on the example of the recently-available RadCalNet data to describe typical methods and challenges that will be faced when harmonising data between imaging spectrometers as well as with multispectral sensors.

The proliferation of affordable spectral sensors for UAV’s, potentially CubeSats, and in-situ spectral observations has led to many groups in Australia making observations of interest to this working party. Industry, Government (mostly State and Federal) and research organisations have been making spectral measurements for a variety of applications. For example, Australia’s Eddy Covariance network, TERN OzFlux, has a number of sites fitted with multi-spectral sensors and other imaging systems (mostly phenocams) in operation, have had hyperspectral flights in years past and have a variety of spectral measurements taken periodically. Combined with net radiometry, typically standard on Eddy Covariance systems, these measurements provide useful information for future mission calibration and validation and simply understanding some of the challenges. Ancillary data available at TERN OzFlux sites, such as vegetation structure and composition, soil type, moisture and temperature, as well as the flux data itself provide additional insight. Each of Australia’s TERN OzFlux sites also have an ongoing site researcher (Principle Investigator) who can provide valuable insight into the nuances of the ecosystem. Using examples from recent work, I will provide a brief overview of the TERN OzFlux sites used in published and un-published work. I will discuss the challenges and opportunities for using these sites in future studies. I will discuss specific spectral calibration techniques for drone based Hyperspectral Imager, the CUBERT S199 and their relevance to the image processing conducted in collaboration with VITO, Belgium. My aim: stimulate discussion on how these technology can be leveraged for future satellite missions.