The Astronomy ESFRI and Research Infrastructure Cluster, ASTERICS, brings together astronomers and astroparticle physicists of 23 European institutes to help world-leading facilities, such as SKA, CTA, KM3NeT, and E-ELT, work together to find common solutions to their Big Data challenges, their interoperability and scheduling, and their data access. ASTERICS is a four year project, funded through the European Union’s Horizon 2020 Framework Programme. One of ASTERICS' goals is to open up multi-wavelength and multi-messenger astronomy to scientists and to the general public through the Virtual Observatory and through citizen science experiments.

The landscape of ground-based gamma-ray astronomy will drastically change with the perspective of the Cherenkov Telescope Array (CTA) that will be operated as an open observatory for the first time in this energy domain. I will present the current work that aims at ensuring the compliance of Cherenkov data with the Virtual Observatory (VO). VO data models and protocols have been tested in a data diffusion prototype, as well as authentication and authorization (A&A) systems and data processing solutions.

ESO builds and operates most advanced ground-based astronomical facilities, and all scientific data gathered by the diverse suite of ESO instruments are returned to the public. One pillar of ESO's archive strategy is to build up high quality content for scientific data products. Therefore, selected data products are systematically produced in-house, and our community delivers substantial datasets for survey and large programmes. Data product interface standards and content validation ensure compatibility with our archive and warrant their scientific legacy value.
Acknowledging its role as a scientific resource on its own, ESO seeks to further enhance the capabilities of its archive.
Future archive services shall allow users to easily discover, visualize, interact and thus efficiently utilize the breadth of ESO data holdings and connect them with other data resources. ESO is highly interested to involve already existing frameworks and suitable building blocks to deploy its new archive services.
Some areas require additional effort, and their advancement should be beneficial also to other archive / VO users and data centers.
They could become interesting development opportunities in the ASTERICS / DADI framework.

I will present some of the data sets we publish at AIP in Potsdam (RAVE, digitized plates, cosmological simulations), and discuss challenges in curating the data. Further, I will give an overview on the technologies behind the scenes and introduce some useful tools, including plug-ins for MySQL, a UWS (IVOA Universal Worker Service) client, and a UWS validator.

The Italian centre for Astronomical Archives (IA2) is an INAF facility for storing, archiving and providing data resources for the astrophysical community, operating within the Italian national institute for astrophysics (INAF) as well as through international collaborations.
A process of refurbishing its architecture is ongoing, including user access solutions:

• distributed archiving infrastructure
• user authentication and group management
• aided automated generation of user interfaces for data discovery

Ground-based solar observatories are entering the era of the next generation 4m-class telescopes – the US-American DKIST currently under construction in Hawaii, and the European EST, which is embedded in the ESFRI road map. They are equipped with several post-focus instruments observing at an unprecedented high spatial and temporal resolution. Their observations will generate a large, diverse, and rapidly growing body of data, which needs to be made accessible to the solar community. We will review the efforts undertaken within the framework of the CASSDA and the European FP7 SOLARNET projects to maximise the scientific exploitation of ground-based solar observations and to prepare for the data archiving and distribution demands by creating a unified data access method with advanced search capabilities. The developed strategies and acquired insights will form the basis of a Solar Virtual Observatory.

(To be announced)

Radio astronomical data models are becoming very complex since the huge possible range of instrumental configurations available with the modern radio telescopes.
What in the past was the last frontiers of data formats in terms of efficiency and flexibility, is now evolving with new strategies and methodologies enabling the persistence of very complex, hierarchical, and multi-purpose information. Various formats (FITS, MBFITS, VLBI's XML description files and ancillary files) of data acquired with the Medicina and Noto Radio Telescopes are handled within one unique archiving system using a configurable tool (NADIR). Data preservation is foreseen to follow the directives of the Open Archival Information System and the International Virtual Observatory Alliance for the data sharing and publication. A Web User Interface allows for easy and user friendly access to data; this archive is also ready to be Virtual Observatory compliant and as much as possible interoperable with other data sources.

The Stellar Department of the Astronomical Institute of the Czech Academy of Sciences maintains archives of spectra acquired with the largest Czech national telescope – the 2m Perek Telescope in Ondřejov – as well as CCD frames of the Ondřejov Southern Photometry Survey obtained at the 1.54 Danish telescope at La Silla by several groups of Czech stellar astronomers in remote observing mode.
Both types of raw data are being regularly processed by human-controlled pipelines up to the final calibrated product and published using basic VO protocols. During and after the data ingestion into the VO server, the high-level data products, as continuum normalised spectra and multi-color light curves, are created automatically using advanced processing algorithms and are made accessible for advanced on-the-fly post-processing with the VO DataLink protocol.
We will describe the whole procedure in detail, emphasising the important role of VO standards facilitating easy discovery and advanced analysis of data as well as potential cross-matching with other world-wide VO resources.

Persistent identifiers such as DOIs, which were introduced roughly 20 years ago, are already well established for scientific publications. An extension to datasets started in 2005, following growing awareness of the importance of accessing the underlying data for scientific work.
Using examples from data collections in astronomy and geology, the use of DOIs for identifying data sets will be discussed, and why data centres are ideally suited for deploying these.

External certification is not mandatory for astronomical data repositories, but hosting data from the project in a certified repository can be a very good point in Data Management Plans, which are more and more required by funding agencies for the projects they fund. The CDS succesfully applied for certification in two frameworks for "basic" certification, the World Data System (WDS) and the Data Seal of Approval (DSA). We got an external, positive evaluation of our processes, which was the initial aim, but it was also a very useful exercise for the team. We will share the lessons learnt.
Recently the DSA and WDS joined efforts in the RDA Repository Audit and Certification DSA–WDS Partnership WG to establish a common certification framework, which is being implemented by the two organisations. The outcome of the current work will be briefly presented. The certification criteria can also be used by a data centre to assess its processes and evaluate progress.

Ground-based observations of interplanetary spacecraft provide complementary information to the onboard instrumentations. In recent years Very Long Baseline Interferometry (VLBI), a radio astronomical technique, has been applied to spacecraft tracking to get data on solar wind composition, satellites' ephemerides, and gravity field studies to probe the internal composition of moons. I will discuss the challenges of joining space science and classical radio astronomy, with particular focus on data acquisition, format and handling.

The VO Registry is a structured repository of metadata on astronomical "resources" -- mainly, services or data collections. If you run standard VO services, their metadata should definitely be in the Registry so that VO clients can find them. But even browser-oriented services can be registered for some extra visibility. Also, registering is not really hard. This talk will discuss the main options, along with a very brief introduction into the Registry concept required for a useful registration.

(To be announced)

We present a science-driven discovery portal called ESA Sky for space astronomy missions, including all the ESA Astronomy Missions at ESAC. This first public release of this services features interfaces for sky exploration and for single and multiple targets. Using the application requires no prior-knowledge of any of the missions involved and gives users world-wide simplified access to high level science-ready data products from ESA Astronomy missions plus a number of ESA-produced source catalogues. XMM-Newton data, metadata, and products were some of the first items to be accessible through ESA Sky. In the next decade, ESASky aims to include not only ESA missions but also aims to provide access to data from other space-based and ground-based astronomy missions and observations.
From a technical point of view, ESA Sky is a web application that offers all-sky projections of full mission datasets using a new generation of Healpix projections called HiPS, detailed geometrical footprints to connect all-sky mosaics to individual observations, and direct access to science-ready data at the underlying mission-specific science archives.

The ANTARES neutrino telescope is collecting data since 2008. The neutrino events are distributed for optical and X-ray followup, exchanged with gamma ray, CR, neutrino, and gravitational wave detectors for multi-messenger analyses. KM3NeT is a future km³-scale neutrino observatory in the northern hemisphere. It is in the construction phase now and the first data is about to be taken. Also, ANTARES data from its first four years has been released via the VO, and more data is planned to be released.

The IVOA has designed a structured list of metadata for the description of observations' data files. Obscore DM describes the properties of a data product obtained by an observation process. It is meant to support a global discovery search across multiple archives supporting the VO infrastructure. The data model offers an homogenized description of the physical coverage in space, the electromagnetic domain, time, polarimetry, as well as general features for data set management. Data model elements can be encoded in the ObsTAP TAP schema and used as selection criteria in ADQL queries via the TAP protocols.

(To be announced)

Ground-based gamma-ray astronomy is a relatively new window on the cosmos. The first source detected from the ground was the Crab nebula, seen by the Whipple telescope in Arizona in 1989. Today, about 150 sources have been detected at TeV energies [1] using gamma-ray telescopes from the ground such as H.E.S.S. in Namibia [2] or VERITAS in Arizona. Until now, almost all of this TeV gamma-ray data was private to the collaborations that built and operated the telescopes, processed and analysed with private software. Soon construction will start for the Cherenkov Telescope Array (CTA) [3], which will be the first ground-based gamma-ray telescope array operated as an open observatory and public observer access [4].
In this presentation, I will report on recent efforts to create open-source science tools as well as an open data specification for imaging atmospheric Cherenkov telescopes (IACTs) like H.E.S.S. or CTA (see [5]), including prototyping activities to export H.E.S.S. data into this format and thoughts on how to publish it. My main motivation to join this workshop is to present the IACT data use case and to learn about and discuss options to publish IACT data in a way that is useful for gamma-ray experts as well as other astronomers.
[1] http://tevcat2.uchicago.edu
[2] https://www.mpi-hd.mpg.de/hfm/HESS
[3] https://www.cta-observatory.org
[4] http://adsabs.harvard.edu/abs/2015arXiv150806078K
[5] https://github.com/open-gamma-ray-astro
  /2016-04_IACT_DL3_Meeting

For the last two decades, atomic and molecular data producers have been taking advantage of the Internet to publish their data. Using various software technologies, they built their own data formats and search system. Users have been accustomed to look for the data they needed across multiple, heterogeneous well-known databases.
The VAMDC infrastructure aims at providing an upper layer to these various databases, so that data can be searched and read through well-known, normalized protocols, capitalizing on previous works on the same subject within the International Virtual Observatory Alliance. We will present an overview of our architecture and what features it provides to simplify the work of both data publisher and data user.

Surveys of interstellar regions require the use of spectroscopic information within the observed range of wavelengths / frequencies. If there is no knowledge of the exact atomic and molecular dataset used for a given analysis, the latter may not be reproductible if the dataset evolves over the years. We also notice that usually there is no citation of the authors who produced the spectroscopic data, since for such large surveys there is a large contribution from many experimental / theoretical spectroscopic papers.
In order to address these issues, we have started a collaboration with the Research Data Alliance: VAMDC became a use-case of the Data Citation Working Group. Through our presentation we would like to highlight how the RDA recommendations are being implemented into the VAMDC infrastructure: the main difficulties we have to cope with come from the fact that the RDA recommendations are proposed for standalone databases or warehouse, whereas VAMDC is a distributed infrastructure with no central management system. We will present our solutions and feedbacks on the data citation theme.

(To be announced)

The Parameter Description Language (PDL) is a standard of the International Virtual Observatory Alliance (IVOA). In this language, parameters are described in a rigorous data model. PDL intends to be an expressive language for self-descriptive scientific web services exposing the semantic and physical nature of input and output parameters, as well as all necessary complex constraints. PDL is indeed suitable for documenting all kinds of scientific web services (from the simplest one to the most complex, exposing heavy simulation codes with hundreds of parameters) and is particularly addressed to scientists or engineers wishing to expose their research codes on-line as public services and / or wishing to interconnect their codes into workflows.
The software framework developed over the PDL grammar allows data / code providers to easily (and quickly) deploy a working PDL service from scratch: the server, client and all the other technical layers are automatically generated by the framework, starting from a PDL description of the service to expose.
Throughout this presentation, we will present the key elements of the PDL grammar. After some details on the components (client, server, workflow engines) of the PDL framework, we will explain how to deploy a working PDL service.

Every photometric survey has some way to produce light curves. The problems come in when differential photometry is ineffective (irregular coverage) or when we want to run light curve extraction many times with different parameters (the process is time consuming).
We will show automatic photometry and light curve extraction from FITS images using the Munipack and HEALPix libraries. We focus on the reusability of this process, on high performance, and on scalability. We will also mention data ingestion and provisioning of all data involved in this process, ensured by the GAVO DaCHS package. The discovery and interoperability of light curve publishing is also important; for this we use Obscore 1.1 and the new Cube Data Model.