1 Journey of the Trace Gas Orbiter (TGO)

The ExoMars TGO mission, launched in March 2016, together with the later ExoMars Rover and Surface Platform (RSP) Mission for launch in 2020, are elements of the European Aurora Programme dedicated to long-term robotic and human exploration of the Solar System. The ExoMars programme is managed by the European Space Agency (ESA) Directorate of Human and Robotic Exploration (D-HRE). The Science Operations of the TGO missionFootnote 1 are implemented under the responsibility of the ESA Directorate of Science (D-SCI). The Trace Gas Orbiter (TGO) mission is ESA’s second mission to operate an orbiter at Mars, and is pursued as part of a broad cooperation with the Russian Federal Space Agency, Roscosmos. Its investigations focus on Martian surface environment conditions and atmospheric composition.

The 2016 spacecraft was launched on 14 March 2016 from Baikonur and arrived at Mars in mid-October of that year. The first in-flight activity involving the SGS was the Mid-Cruise Checkout (MCC) performed in June 2016. Mars Orbit Insertion (MOI) of the TGO orbiter took place on October 19 2016.Footnote 2 The TGO successfully entered a 4-day-period Mars Capture Orbit (MCO). A series of manoeuvres in Jan.-Feb. 2017 using on-board fuel changed the orbital inclination to 74 degrees and reduced the apoapsis, to yield a 24-hour orbit. Thereafter, an extended period of aerobraking lasting just under a year, and some final propulsion manoeuvres, left the orbiter in its 2-hour approximately circular science and relay communications orbit at an altitude of about 400 km above the Martian surface.

The TGO is now entering its Nominal Science and Relay Phase (NSRP) baselined to last for one Martian year (i.e. 687 Earth solar days, or 668.6 Martian sols).

2 The TGO Mission

The TGO is a three-axis stabilised platform with a science payload consisting of four composite instruments which collectively provide spectro-imaging capabilities from the UV to the near- and thermal-infrared.

From its science orbit the TGO will perform detailed characterisation of the composition of the Martian atmosphere with a sensitivity three orders of magnitude better than previous measurements, studying the distribution of trace gases, their temporal and spatial evolution, and seeking to identify their geographical and seasonal variability, and in turn their sources and sinks. At the same time it will provide greatly improved mapping of the subsurface water distribution.

Fundamental observational modalities are:

  • Solar Occultation: In which the Sun is observed through the Martian atmosphere for a few (typically about 5) minutes around sunrise and sunset limb/terminator crossings, yielding a profile of the atmospheric composition from the top of the atmosphere down almost to the surface

  • Nadir Observation: In which the instruments observe the planetary surface passing beneath the spacecraft in reflected sunlight or thermal re-emission

  • Limb Observation: In which elevations in the atmosphere above the surface are studied looking towards the limb of the planet to study atmospheric emissions.

During the Nominal Science and Relay Phase of the mission the TGO scientific measurements will be coordinated with those of other assets at Mars.

Along with its scientific mission, the TGO will provide communication, command and data relay support to the ExoMars 2020 Rover and Surface Platform, following their early-2021 arrival at Mars, and to NASA assets operating on the surface.

Table 1 provides a breakdown of the principal functional periods of the mission.

Table 1 Principal functional periods of the TGO mission
Full size table
Table 2 Scope of the major SGS Validation, Verification and Commissioning periods on the ground and in flight, leading up to the Nominal Science and Relay Phase. (*) The MCC was an engineering checkout of the health of the TGO science payload performed about three months after launch en route to Mars, and was the first in-flight activity conducted though the SGS, as distinct from earlier in-flight activities conducted directly from the ESOC MOC. The MCC basically repeated the set of tests earlier performed during the post-launch Near Earth Checkout (NEC), but exercising the more flexible planning tools, and the processes, of the SGS
Full size table

2.1 Science Payload

2.1.1 The ACS Instrument

ACSFootnote 3 (Korablev et al. this issue) consists of three separate infrared spectrometers (NIR, MIR, and TIRVIM) intended to observe the Martian atmosphere in solar occultation, nadir and limb geometry, and which share common mechanical, electrical, and thermal interfaces.

ACS will provide a global view of the spatial and temporal variability of trace atmospheric gases, penetrating down to the surface even under conditions of high dust obscuration. Concurrent mapping and vertical profiling of climatic variables (e.g. dust, cloud, temperature) will relate the trace species to their environmental context throughout at least one Martian year.

2.1.2 The NOMAD Instrument

NOMADFootnote 4 (Vandaele et al. this issue) consists of three spectrometers that survey the Martian atmosphere at Ultraviolet, Visible and Infrared wavelengths.

Like ACS, NOMAD studies spatial and temporal variability of trace gas concentrations and key isotopes in the Martian atmosphere. Among other goals it seeks to place constraints on models for the origin of atmospheric methane, to study gases related to possible ongoing geophysical and volcanic activity on Mars, and their breakdown processes, sources and sinks, and to explore Martian climatology and atmospheric dynamics, throughout the Martian year.

2.1.3 The CaSSIS Instrument

CaSSISFootnote 5 (Thomas et al. this issue) is a high-resolution (4.5 metres per pixel) stereo colour imaging camera dedicated to imaging the Martian surface. CaSSIS observations may target specific locations on the surface, which requires the yaw steering of the spacecraft to be suspended while CaSSIS pointing is prime, or they may be non-targeted, in which case CaSISS “rides along” with the default spacecraft power optimised nadir pointing. Stereo image pairs permit building accurate digital elevation models of the Martian surface. Stereo imaging is performed through the employment of a rotational capability which allows the instrument to alternately look forward or backward along its trajectory over the surface.

2.1.4 The FREND Instrument

FRENDFootnote 6 (Mitrofanov et al. this issue) is devoted to surface and subsurface studies, measuring the flux of neutrons ejected from water deposits in the Martian soil by incoming cosmic ray particles. This allows estimation of water content in the soil to a depth of 1 meter and with a spatial resolution of about 40 km on the surface. More generally, FREND records the radiation environment in space and on the surface of Mars, useful for spacecraft radiation analysis and manned flight radiation safety analysis.

3 Organisation of the TGO Science Ground Segment (SGS)

Figure 1 depicts a generic organization of mission operations with reference to the science component. A mission consists of the Space Segment (typically the in-flight spacecraft) and a Ground Segment which has two main sectors:

  • the Operations Ground Segment (OGS), comprised of the Mission Operations Centre (MOC) at ESA’s European Space Operations Centre (ESOC) in Darmstadt, Germany, together with elements of ESA’s Estrack network of satellite tracking Ground Stations, and

  • the Science Ground Segment (SGS), comprised of the Science Operations Centre (SOC) at ESA’s European Space Astronomy Centre (ESAC), near Madrid, Spain, together with Instrument Team Centres (ITCs) connected with the science payload instruments, and typically one or more Partner (Science) Centres (PSC), collaborating in the science operations of the mission.

There may be more than one ITC per payload instrument, and ITCs are not restricted to ESA member states. PSCs may be key contributors to any aspect of the mission, most typically contributing to data analysis systems, and are often major centres of collaborating non-member states.

Fig. 1
figure 1

Representation of a generic Science Ground Segment, indicating key centres internal to the SGS, and showing the SGS relationship to principal management entities and with the Mission Operations Centre (MOC)

Full size image

Overall responsibility for a project in development rests with the ESA Project Manager. After successful commissioning in-flight, responsibility passes to an ESA Mission Manager for the remaining lifetime of the mission.

The Project Scientist, assisted by and chairing the Science Working Team (SWT), establishes all requirements on the scientific accomplishments of the mission, and approves the science operations plan. The SWT membership includes the science payload instrument Principal Investigators (PIs), and representatives of the Scientific Community.

Figure 2 depicts the specific organisation of the ExoMars TGO SGS. The ITCs of the four TGO science payload instruments are indicated, as well as the Science Operations Centre of the Russian partner (IKI SOC).

Fig. 2
figure 2

Connections among the centres involved in ExoMars TGO Science Operations

Full size image

Within the ExoMars TGO SGS, the parties have the following roles:

  • The ExoMars 2016 Orbiter Science Operations Centre (SOC) at ESAC (Cardesín Moinelo et al. 2014, and this work) is responsible for:

    • overall coordination of SGS activities

    • serving as sole interface to the instrument teams and the ESOC Mission Operations Centre (MOC) in the execution of the science planning tasks

    • delivery to the MOC of the science Payload Operations Requests (PORs)

    • development and/or operation of core science operational systems (e.g. the MAPPS planning tool (see Sect. 6), or the SPICE observation geometry libraries derived from the SPICE toolkit (Acton 1996) developed and distributed by NASA JPL NAIF)

    • data handling: In the case of ExoMars TGO this involves the production, from the Spacecraft (S/C) and payload telemetry (T/M), of low-level science products (PDS4-raw products), which are an input to the higher level data processing done by the instrument teams. (See Sect. 6.2.)

    • data archiving in the ESAC-developed ESA Planetary Science Archive (PSA) (Besse et al. 2018; Macfarlane et al. 2018; Zender et al. 2009)

  • The Russian Science Operations Centre of the Russian Ground Complex (NNK) housed at the Space Research Institute of the Russian Academy of Sciences (IKI) in Moscow, maintains:

    • full visibility into the science planning process

    • redundant installations of SOC-based Data Processing S/W

    • mirroring of ESAC science archive related data

    • special services to Russian PIs or to the Russian community

    • selected agreed enhancements over the nominal system, shared with the ESAC SOC, extending mission efficiency or science returnFootnote 7

  • The Payload Instrument Team Centres (ITCs):

    • develop, test and support payload instrument observing modes

    • provide science observation requests to the SOC

    • ensure successful operation of their instruments in flight

    • verify & validate the instruments’ timeline (ITL) and commands

    • monitor & optimise instrument performance & housekeeping

    • perform instrument calibration and trend monitoring

    • investigate instrument anomalies

    • perform maintenance of the On-Board Software (OBSW)

    • produce instrument & software user manuals and

    • generate data products up to the highest level science products and deliver them to the SOC for archiving in the PSA.

The Member States usually contribute the bulk of resources involved in the ITCs and ESA provides overall coordination (rather than direct management of the components) of the SGS. However, ESA has overall responsibility for the scientific output of the mission.

4 Science Operations Processes

4.1 The ExoMars2016 Science Ground Segment (SGS)

Process flow within the mission Ground Segment is depicted in Fig. 3.

Fig. 3
figure 3

Process flow in the Science Ground Segment (SGS)

Full size image

The SGS, together with the Mission Operations Centre (MOC) at ESOC constitutes the mission Ground Segment overall. The ESAC Science Operations Centre receives science planning requests from the Instrument Teams, reflecting consultations within the Science Working Team (SWT—chaired by the Project Scientist).

The SOC, respecting payload instrument and spacecraft operational constraints and resource limitations, prepares the timeline of instrument operations for execution on the spacecraft in conformance with a set of planning template files defined by the MOC which embody all spacecraft, orbital, dynamical and ground station link constraints. After checking by the PI teamsFootnote 8 this observation timeline, in the form of a Payload Operations Request (POR), together with an associated Pointing Timeline Request (PTR), is delivered by the SOC to the MOC for uplink to the spacecraft, and execution.

Telemetry flows back from the spacecraft and undergoes basic data processing at the SOC and the resulting low-level (PDS4-raw) products are archived in the ESA Planetary Science Archive (PSA) and are made available to the Instrument Teams and to the Russian partner at IKI. (If circumstances require it, both ITCs and IKI can also receive telemetry directly from the MOC.) The Instrument Teams perform the PDS4-raw to Calibrated Data Processing step, and deliver the resulting calibrated data products, and higher level science products, to ESAC for archiving in the PSA. Throughout these stages both the ESAC and IKI SOCs perform a range of activities in support of the payload teams (e.g. observational geometry computation using the SPICE toolkit, support with data standardisation, cross-checking of command sequences etc.).

5 Development of the Science Ground Segment

A Science Ground Segment is developed over a period of several years preceding deployment of the operational system. An initial Science Operations Concept (SOCD) is elaborated, from which detailed requirements are generated leading to system design and implementation, strongly modulated by testing, validation and verification processes.

Figure 4 depicts the distribution through time of nine focussed, end-to-end test and simulation campaigns conducted in the TGO SGS through the years of its development. It also identifies other key milestones of the development which place those Operations Rehearsals (ORs—as they have been called) into context. Conducted at various stages throughout the development of the mission, they have been executed to progressively validate the system concept, design and build, and in the process train the SGS Operations Teams, and debug processes and interfaces to reach operational readiness.

Fig. 4
figure 4

Development timeline for the ExoMars TGO SGS, built around the major test campaigns and simulations involving the SGS. See Table 2 for further explanations

Full size image

A Science Operations Development Manager (DM) steers this process, handing over in principle to a Science Operations Manager (ScOM),Footnote 9 normally well before nominal science operations begin.

6 Science Operations

There are three main components to the science operations of a typical science mission:

  • Preparation and uplink of efficient, safe, observational timelines

  • Generation of all levels of science data products

  • Archiving and distribution of the science data products.

6.1 Uplink System Architecture: Science Command Timeline Preparation and Verification

The ExoMars TGO uplink system re-uses expertise and tools from previous planetary missions, such as Mars Express (Pischel and Zegers 2009; Rabenau et al. 2012), Venus Express (Titov 2006; Svedhem 2007), and Rosetta (Wirth et al. 2007; Ashman et al. 2016; Costa Sitjà et al. 2016a; Geiger 2016; Vallat et al. 2017), with a centralized planning system at ESAC for the preparation and generation of operational products that are then delivered to ESOC for uplink to the spacecraft. The Mission Analysis and Payload Planning System (MAPPS) is a key component of the ESAC observation planning system used by all ESA planetary missions. MAPPS, originally developed in ESA’s Technical Directorate (D-TEC) in ESTEC, is now maintained and developed at ESAC as a multi-mission support tool. MAPPS is provided to all science payload Instrument Team

Centres and its use by them is supported by ESAC. The nature of the task addressed by the MAPPS tool is discussed in Sect. 6.1.2.

Another key component of the observation planning system is the SPICEFootnote 10 observation geometry computation toolkit developed at NASA’s Navigation and Ancillary Information Facility (NAIF) at JPL (Acton 1996). A cross-mission SPICE service at ESAC (Costa Sitjà et al. 2016b) supports ESA’s Planetary mission SOCs, their associated science payload Instrument Teams, and to some degree the scientific community to, quoting NAIF: “assist … in planning and interpreting scientific observations from space-borne instruments, and to assist … engineers involved in modelling, planning and executing activities needed to conduct planetary exploration missions.”

6.1.1 Science Planning Cycles

Figure 5 provides a graphical representation of the most significant steps and interactions in the SOC uplink planning process described in the text.

Fig. 5
figure 5

Main sub-system components of the SGS Uplink System described in the text

Full size image

The TGO science planning process, spread over several ESAC TGO SOC team members, can be broken down into the following elements:

  • Opportunity Analysis: Identification, calculation and analysis of the science opportunity windows defined by the geometrical conditions of observation specified by the instrument teams, and the operational conditions and events specified by the Mission Operations Centre.

  • Planning and Scheduling: This spans the Long, Medium and Short Term Planning processes:

    • Long Term Planning (LTP): With a look-ahead of six months to a year, a baseline science timeline, or Science Activity Plan (SAP), is established, that contains the instrument pointing and operations requests, in line with resource profiles, meeting the SWT science goals and establishing operational feasibility of the core science plan.

    • Medium Term Planning (MTP): With a look-ahead of one to a few months, the pointing and commanding timelines, having been iterated with the Instrument Teams and the SWT, converge to the agreed attitude sequence and resource allocation for all observations in the plan.

    • Short Term Planning (STP): With a lead-time of about one week before the execution of observations, the detailed instrument commanding timeline for all observations is elaborated, respecting all instrument, spacecraft and mission constraints and resource envelopes.

  • Plan Validation: Beginning with the Long Term Planning cycle referred to above, regular meetings of the SOC planner with the Instrument Teams and the Project Scientist examine successive iterations of the Science Activity Plan (SAP). The SAP is iterated through the planning tools down to the level of the instrument command timeline, ensuring that a version of the plan is realised which implements the required science observations for the mission. As the planning proceeds through the Medium Term and finally the Short Term cycles, the iterations and refinements with the Instrument Teams and the Mission Operations Centre at MOC intensify ensuring that the finally performed observations fully reflect the science aims of the mission to the most detailed level. Exposure durations, filter selection, image cadence, also compatibility of the pointing and commanding timeline with the attitude, thermal, illumination, angular momentum and other resource envelopes of the system, are confirmed. At a number of points throughout the process the developing plan is exposed to the full Science Working Team (SWT) including the Interdisciplinary Scientists, who can adjust and finally endorse it.

  • Uplink Product Generation: All operational products for the Medium and Short Term Planning cycles are generated, for iteration with the PI teams and export of the final observation timeline, including the pointings, command sequences and resource profiles, into MOC compatible observation request formats (POR/PTR).

  • Auxiliary Data Conversion: This step condenses the handling processes of all ancillary data products, mainly generated by the MOC, for the Long, Medium and Short Term Planning processes. A key functionality is generation of the SPICE kernel files using the Auxiliary Data Conversion System (ADCS) from the data delivered by Flight Dynamics and by using tools and processes based on the SPICE toolkit to determine higher-level observation geometry parameters to support both the science planning and the subsequent data analysis activities. The aim is to convert all required auxiliary information from the spacecraft and make it available to the PI teams and the science community in SPICE format.

  • Operations Data Management: This concerns the organisation of all operational files within the Uplink system, making the information available for all other sub-systems and modules within the SOC, and managing the main operational interfaces to receive and transfer information with other parties within the SGS or ESOC. This sub-system is also responsible for some of the basic validation processes that ensure the content, syntax and format correctness of operational files. Additionally this sub-system follows the guidelines defined in the SGS Configuration Management Plan for the configuration control of all information employed.

6.1.2 MAPPS: Reconciling Observations, Resources and Constraints

As indicate above, the execution of the mission’s science programme implies a distribution of available system and mission resources across the payload instruments, while a range of physical constraints limit what can be done in any given period.

Examples of in-demand mission resources are:

  • The share of observing opportunities (e.g. Sun occultations by the Martian atmosphere at the planetary limb; nadir and other imaging opportunities, …)

  • Power

  • Spacecraft angular momentum, which must be held within strict limits, and which constrains the amount and rate of slewing that can be accommodated in any period of time, and of course therefore the number and diversity of pointing opportunities

  • Downlink telemetry data capacity (shared also with the 2020 Rover and Surface Platform Mission which relies on the TGO as a communications relay platform)

  • Uplink telecommand capacity (shared also with the 2020 Rover and Surface Platform Mission)

  • Available on-board solid-state memory (in the Payload Data Handling Unit (PDHU)) for storing science data until download, especially during periods of reduced data downlink capacity when the Earth-Mars separation is large. (shared also with the 2020 Rover and Surface Platform Mission)

Examples of observational constraints are:

  • Requirements of the various observation types and modalities of the instruments

  • Constraints on Solar illumination angle (e.g. on the Solar Panels) and duration—linked to power generation and to battery capacity and lifetime

  • Constraints on Solar illumination on (parts of) the science payloads (for example thermal radiators)—linked to thermal limits within the instruments, or linked to stray-light limits

  • Spacecraft orbit, orbit phasing, orbit inclination, ground track repeat cadence

  • Avoidance of interference with spacecraft platform operations (such as Orbit Change Manoeuvres (OCMs), momentum wheel off-loadings (WOLs), or Ground Station communications passes).

Figure 6 provides an illustrative example of engineering constraints in action. It shows the relationship among the TGO payload spectrometer apertures projected onto the sky. It is evident that simultaneous observation of a common target—specifically the Sun during atmospheric occultations—is possible for some instrument combinations, but not for others. The resulting trade-offs of observational efficiency, scientific priority and acceptable distribution of observing time among the payload instruments, modulated by considerations of available on-board storage and relative telemetry transmission rates to ground, also as a function of Earth-Mars separation, are examples of typical issues addressed during the science planning process.

Fig. 6
figure 6

Relationship among the projections onto the sky of the TGO spectrometer apertures

Full size image

6.1.3 Science Planning Tools

Accomplishing all these tradeoffs is an iterative process between the SOC and the ICTs, who employ the same planning tools as the SOC.

The management of almost all of these resources is governed by policies and agreements established well before launch. These policies and agreements are encoded in a range of systems that are inherently configurable to implement the policies and agreements, such as the mission System Budget, the Flight Operations and Flight Control Procedures, the rules for placing Ground Station passes and Mars Surface Asset relay communications passes into the timeline, and in decisions taken by the Project Scientist and the Science Working Team (SWT) aimed at optimising the science return of the mission. Very often these policies, procedures and agreements are largely and inflexibly determined by physical and engineering properties of the system.

Mission planning (mission simulation and modelling) tools at both ESOC/MOC and at ESAC/SOC are configured to take account of all of the above constraints to arrive at a compatible, valid, observation timeline. In the course of nominal Science Planning at the SOC, the MAPPS tool, together with SPICE and a range of dedicated stand-alone widgets (all validated through very extensive testing, e.g. in Operations Rehearsals), are applied to generate the Payload Operations Request (POR) and Pointing Timeline Request file (PTR) and a set of related files. These are validated against a MOC-provided web interface tool to determine suitability for delivery to the MOC for combination with spacecraft platform management sequences and uplink and execution on the spacecraft.

In cases where subjective evaluations or qualitative judgements have to be made in the area of science planning, with possible disagreements among major mission stakeholders which are not resolved in dialogue with the SOC mission planners, then the Project Scientist, in consultation with the Science Working Team (SWT),Footnote 11 in which he is advised by the Interdisciplinary Scientists, takes responsibility for deciding how to proceed.

6.2 Data Handling System Architecture

6.2.1 Data Processing

The baseline architecture for the downlink system is divided in two sub-systems: Data Processing and Data Archiving.

The essential architecture of the Data Handling sub-system is summarised in Fig. 7, illustrating the main processes and interfaces within the SGS.

Fig. 7
figure 7

Data Processing architecture in the TGO SGS

Full size image

The TGO SOC data retrieval and processing is an automated process. The ExoMars TGO SOC retrieves the Spacecraft and Instrument telemetry and related files from the ESOC Data Distribution System (EDDS), and processing up to standard Planetary Data System v4 (PDS4) raw levelFootnote 12 is done directly at ESAC, in a collaborative effort with the instrument teams and the Russian SOC. ExoMars was the first planetary mission at ESAC to use operationally the PDS4 standard.

The raw data processing system is written in Java and is developed using an Agile development process, providing the flexibility to adapt to new and changing requirements. A skeleton system was constructed at an early stage and evolved into the operational system using an incremental, iterative, process. Continuous integration and continuous testing ensure that the code base is always in a working state and new releases can be made at any time. New features and bug fixes are implemented in Feature Branches, which are merged into the master branch when they are ready for release. Software releases are created by branching off the master branch and testing in one or more release candidates. The development uses industry standard tools, including Eclipse, Maven, Git, Jenkins, Nexus and SonarQube.

As mentioned before, calibrated and higher-level science data products are produced by the payload instrument teams within their consortia. The structure and interfaces for the exchange of all levels of data product are set out in great detail in the PI-provided Experiment to Archive Interface Control Documents (EAICDs), which also customarily serve as the Data Processing User Manuals for the science instruments. The EAICDs document all aspects of the science products organisation, including data and metadata organisation, and the specification of how the products should be collected into PDS4 bundles for ingestion into the PSA.

Data is exchanged between the ESAC SOC and the Instrument Teams either as discrete, perhaps observation-specific, PDS4 data products—from the local Data Stores, for day-to-day working interactions, or as archive PDS4 Data Bundles, where data is retrieved from, or delivered for ingestion into, the PSA. The PSA has a set of rules on the organisation of data at the bundle and collection level. The products are ingested into the PSA in accordance with these rules. At lower levels the data organisation is permitted to be determined by mission/instrument needs and this organisation is agreed between ESAC and the Instrument Team Centres (ITCs). The full structure is described in the EAICDs and the PSA rules are specified in the PSA PDS4 Archiving Guide maintained by the ESAC Science Data Centre (ESDC).

While ESAC and the Instrument Teams exchange data products, there is no systematic exchange of Data Processing software with the Instrument Teams in the ExoMars case, although SOC-developed and SOC-operated processing software is freely available to all partners in the SGS for download from the SOC Source Tree GIT repository should they wish to obtain it.

Table 3 Data product levels produced for ExoMars TGO, and their definitions
Full size table

6.2.2 Data Archiving

The architecture, construction and operation of the ESA Planetary Science Archive (PSA) is an ESAC/ESDC responsibility.

Since mid-January 2017 ESA/ESAC serves the data from numerous ESA planetary missions to the worldwide scientific community through the redesigned ESA Planetary Science Archive (PSA)Footnote 13 (Besse et al. 2018; Macfarlane et al. 2018; Zender et al. 2009), employing for the newer and future missions PDS4 standard products, and retaining the PDS3 standard for older missions.

For ExoMars, The PDS4-raw data products derived by SOC are stored in the ESA Planetary Science Archive as described below. Science Instrument Teams may retrieve the PDS4-raw data products from the PSA to derive the higher level data products which are delivered in return to the SOC for archiving in the PSA. The reference to “mirror” in Fig. 7 refers to the fact that the PSA Data Bundles and the ESAC local working Data Store are mirrored with the Russian SOC in Moscow.

Instrument Teams, the ESAC and IKI SOCs, and specific other external entities (e.g. Co-Investigator institutions) are able to deliver data products of any processing level (raw, calibrated or derived) to the SOC Data Delivery Area for inclusion in the PSA. This is reflected by the dashed line in Fig. 7 indicating the ICTs providing all levels of data product to “Delivery Package Creation”. The ITCs may use SOC tools for validation and packaging allowing them to directly send products to the PSA. However, in nominal practice the ITC delivers first to the ESAC datastore and the SOC then prepares the data for delivery to the PSA by generating PDS transfer manifests. On ingestion, the PSA places the data products into bundle structures specified in the delivery.

All the ExoMars 2016 data in the PSA is mirrored to the Russian Science Archive.Footnote 14 The full PSA architecture is shown in Figs. 8 and 9.

Fig. 8
figure 8

The ESA Planetary Science Archive architecture. All ExoMars data in the PSA is delivered to the PSA as PDS4 transfer manifests. The PSA is designed so that users then may access the data via a variety of interfaces, but as the PSA is still at an interim stage of development only Table View and the FTP access are currently offered

Full size image
Fig. 9
figure 9

A view of the PSA process flow, referenced in the text

Full size image

A user-friendly interface allows the worldwide scientific community to have access to all the science data products stored in the PSA. This interface provides a filtering mechanism within a set of menus which allow searches on target, instrument, sub-instrument, time, and data processing level. The search menus are planned to be expanded to include geometry information and eventually mission/instrument specific metadata. Additionally a free search mechanism has been implemented to allow direct metadata searches. A PSA image gallery provides an interface to directly browse images using the same search functionalities. For ancillary data, such as spacecraft housekeeping and SPICE kernels, an FTP interface is also provided. Later a selection of appropriate data from a view of a planetary surface will be offered and this interface will also be a primary interface for the ExoMars 2020 (RSP) mission.

All data in the FTP area is public while the science data provided through the user interface may be either public or under proprietary protection. Within the proprietary period (6 months from data downlink) data access is restricted to specific ESAC and IKI SOC technical personnel, and relevant Instrument Team personnel. The proprietary protection is applied on download, so all PSA users can see that there are products in the archive, but only authorised users can download them.

After expiry of the proprietary protection period for new data (6 months), a user-friendly interface allows the worldwide scientific community to have access to all the data and data products stored in the PSA.

An aspect of bringing the data products to their general release is the validation of the whole data processing chain through which they are produced. A major element of that validation is the peer review of the pipeline, through which relevant experts inspect a representative sample of the data products and provide criticism or endorsement. The first such data pipeline peer review for ExoMars TGO is targeted for late 2018 focussed on the early-mission calibrated data products. Further peer reviews will be organised as derived data becomes available, and as dictated by the mission evolution. Post-mission, a final review of all data will be conducted.

7 Comparison with Other ESA Planetary Science Ground Segments

The general structure and organisation of the ExoMars2016 Ground Segment, both SGS and OGS, is quite generic to ESA planetary missions, but it is worth contrasting some key similarities and distinctions with other missions:

  • The organisation of the science planning process into three phases (long, medium and short term—Sect. 6.1), allowing comprehensive preparation of the global science programme with progressive refinement and adaptation to actual mission parameters leading up to execution, is highly characteristic;

  • The MAPPS tool (Mission Analysis and Payload Planning System) has been used through generations of planetary missions, for example SMART-1, MEX, VEX and Rosetta, and is being used for ExoMars and JUICE (the latter at least in its current early development period);

  • ExoMars (both the 2016 and the 2020 missions) joins SMART-1, Giotto, Huygens, VEX, MEX, Rosetta and Bepi Colombo as elements of ESA’s Planetary Science Archive (PSA).Footnote 15

Significant differences, or in some cases evolution in, the approach across missions can be found in other areas. For example:

  • As a component of ESA’s Robotic Exploration Programme TGO began its life as a communications relay platform for the 2020 Rover and Surface Platform mission, and for other surface assets (including NASA assets). Funding for the Science Operations of the TGO mission was mandated by the ESA Science Programme Committee (SPC) in mid-2013, while a first version of the SGS had to be operational in time to support the mid-2016 Mid-Cruise Checkout (MCC) of the science payload, which was the first in-flight task of the SGS. Therefore an important challenge for the ExoMars SGS development has been the requirement to develop and deploy an operational SGS in a relatively short period of time. A key element in achieving that goal was the extensive reuse and commonality of systems across missions, referred to above. But perhaps the most crucial element has been the availability at ESAC of a large pool of highly experienced engineers and scientists who could reapply skills developed and honed in the several earlier or parallel missions already mentioned. One must add to that the wider experience in operational processes at ESAC accumulated through decades in the development and operation of the SGSs of the full complement of ESA Solar System and Astronomy missions.

  • In the area of Data Processing, the distribution of responsibility among the constituent centres of the SGS, and between the ESA SOC and the non-ESA centres varies substantially across missions. The ExoMars SOC limits its involvement in data processing to the production of low-level data products, with the generation of calibrated and higher-level science products performed by the payload instrument (PI) teams (see Sect. 6.2 and Table 3). In contrast, for example, all levels of data processing will be performed within the Bepi Colombo SOC at ESAC.

  • While the ExoMars SOC follows, as mentioned above, a series of ESA planetary missions in rendering its data through the PSA, it is the first ESA mission to employ the PDS4 data standard and to integrate the PSA into the routine operational process by serving, during the Nominal Science and Relay Phase of the mission, the low-level data products to the instrument teams through the PSA interface within 24 hours of their receipt at ESAC.Footnote 16 ExoMars is also the first mission to accomplish the archiving of science products from the earliest phases of the in-flight mission.