How did you join the European Space Agency?
I studied fundamental physics and astrophysics at the Université Paul Sabatier and aerospace engineering at the Ecole Nationale Supérieure de l’Aéronautique et de l’Espace in Toulouse. I first joined the Agency in late 2000 as a Young Graduate. Upon my arrival I got involved in the pre-launch activities of ENVISAT, and helped specifying a software tool which today allows ENVISAT users to visualise and manipulate the Earth Observation instrument data which ESA provides them.
What were your first experiences on the job and how did you develop professionally?
I was involved in one of the most exciting moments of the ENVISAT mission: looking at the quality of the first data dumped by the satellite and presenting them to the public and to the scientific world. Over the next 10 years, I have supported the ESRIN group responsible for the improvement of the quality of the ENVISAT data in general and of one of its 10 instruments in particular: the Medium Resolution Imaging Spectrometer (MERIS). Together with the MERIS user community, we found ways to diversify and improve its products.
In addition to this contribution to ENVISAT, I became involved in the feasibility studies of new Earth Observation missions, with two missions that did not make it through the ESA selection process and one that made it through and should fly in few years time: called EarthCARE. Currently, I also provide support to the feasibility studies of FLEX, another new exciting mission aiming at measuring vegetation fluorescence from space!
For each one of these missions, I have carried out a mix of engineering, scientific and management tasks, such as prototyping algorithms to exploit the payload data, modelling of the instrument performance, running contracts with European companies and universities to develop new algorithms and software for each one of these missions.
What did you have to learn on the job?
Learning on the job has been necessary and essential to me: each one of the new missions I get involved with has its specificities, and it takes time and efforts to get familiar with it and to understand how to contribute to it efficiently. By the way, I actually acquired most of my IT knowledge on the job. A general IT knowledge is an asset, but since most software is being developed in the frame of ESA missions that software is very specific.
Elsa's daily routine...if there is such a thing!
How much time, from first design to launch and operations, does an average Earth Observation satellite project take?
This is quite variable. When it comes to a complex Earth Observation mission, it can take up to a decade between the first investigations and the launch. This was the case for ENVISAT which is the largest Earth Observation satellite ever flown. ENVISAT is a satellite that carries 10 scientific instruments and weights 8000 kg. Smaller Earth Observation satellites, carrying an instrument already flown, can be developed in much shorter time, say a few years.
All about algorithms - what is your typical day like?
In short, my job is to provide algorithms that allow the exploiting of data generated by scientific instruments on board of ESA Earth Observation satellites. I find this most exciting. It implies digging into a scientific topic, understanding what the instrument measurements are made of, understanding how its measurements relate to a physical quantity in the Earth atmosphere, its oceans or its vegetation.
Sometimes, however, we neither have the in-house expertise ourselves nor the resources to fully develop such algorithms: we then commission the work to industry, laboratories or universities through a contract. This involves essentially the following tasks:
- We write down our needs in a Statement of Work and publish it on the web
- We then assess the technical and financial proposals we receive in response to it
- We follow the work being done by the winning consortium and ensure it is in line with our expectations. In the end this means a lot of managerial work.
It is, however, very important to me to find the right balance between hands on work and running contracts.
Depending on the project and its progress: How often and where do you need to travel to?
My interactions with other ESA centres in the Netherlands, Germany or Spain are limited. My working place is the European Space Research Institute (ESRIN), located nearby Rome. ESRIN is the ESA centre which is entirely dedicated to the exploitation of the Earth Observation satellites. There, I mainly give support to the office in charge of Earth Observation data quality and algorithm improvements. For each running project, I nevertheless need to travel. This happens about once per quarter, but increasingly I try to rely on videoconference tools. Up to now, I am normally off between 1 and 2 full months a year, cumulating all travels.
Do you physically check satellite instruments while they are being built?
My work involves modelling rather than building hardware. Building models, like building hardware generally requires me to interact with colleagues, industry, labs and universities. In order to develop a retrieval algorithm for a new instrument I need for example:
- to have the specification of the instruments
- to know the quantity which needs to be retrieved and with which accuracy and under which conditions
- to have the specifications of the system in which such an algorithm will be run.
I therefore need to interact with the ESA instrument engineer, the scientists who have proposed the mission and defined its scientific goals, and the ground segment engineer who will run the retrieval algorithm.
Which recommendations can you give to students or young graduates if they want to apply at ESA?
I would recommend to first check out on the ESA portal what it is that ESA does that most raises his or her interest: human spaceflight, Earth observation, space science, launchers… If there’s anything you fancy, look out for a Young Graduate Trainee position, or for an internship or check out which national trainee programmes there are. There are plenty of opportunities, good luck!
The Milestones of a typical Launch Campaign
'Launch campaign' is the term used for the time frame starting three to four months prior to launch until a few days after launch, when the spacecraft enters the commissioning phase. This is a very intense period for the entire mission team, but for us flight operators, this period is dominated by the last interface test with the spacecraft, used to verify the closure of flight-critical anomalies, and the simulations campaign.
The sims campaign occupies the last three to four months before launch, with up to two simulations per week. Its objective is to train the mission control team, composed of the flight control team, the ground segment support teams, and the spacecraft expert team. The campaign starts with familiarisation with nominal scenarios, and then trains the team on contingencies by injecting failures at various levels of the ground and space segments. In this campaign, a software simulator replaces the spacecraft and the ground stations network, all other ground equipments are the operational ones.
For our team, the launch count-down typically starts 8 hours prior to launch. From that moment onwards, the mission control team is on-console and follows a joint timeline with the launch site. The mission control team will remain in operations in shifts covering 24 hours a day until a few days after launch, when the spacecraft starts its commissioning phase. At this point operations are moved to a single daily shift and to a mission dedicated control room. Space segment experts are then only required on site for specific operations. This marks the end of the so-called launch and early orbit phase (LEOP).
Elsa’s current project: BepiColombo
There are several types of meetings which are necessary during the development phase of a new project:
- progress meetings
- science working group meetings
- review collocations
- operations interface meetings
In the development phase we hold progress meetings with all contractors. A mission like BepiColombo involves more than 70 contractors in various areas. Meetings typically take place every 4 to 8 weeks. As members of the Flight Control Team however, we mainly attend System and Avionics discussions on the space segment side. These meetings take place at the site of the company in charge. On the ground segment side, we attend all progress meetings with the companies developing the ground systems for the mission. These meetings are normally held in ESOC.
Twice per year, the project organises meetings to discuss mission scientific issues with the teams in charge of developing the scientific instruments to be embarked on the spacecraft. The Flight Control Team attends these meetings.
At all levels, the developments are subject to regular reviews, from requirements specification to readiness for launch. We contribute to these reviews at system/mission level, for subsystems that are key to spacecraft operations (e.g. on-board computer, mass memory, transponder, on-board software), for the instruments, and for the ground systems. The comments raised by the reviewers are normally discussed in one face to face meeting, normally held on-site at the company under review.
On BepiColombo, we have regular operations interface meetings with the Japanese Agency and with the ESA colleagues in charge of science operations. When necessary, we also set up operations interface meetings with the instrument teams.