It was getting warmer throughout the week. Air temperature is around 0°C and it rains on Saturday. That is unusual. The snow surface is quickly covered by ice and everything becomes slippery. The meteorologists had predicted that, we sleep late. The missing sun is really missing on weekends.

We discuss the situation in the evening. The wind is extremely gusty. The aircraft could take off and fly but the landing would be risky in the lee of the mountains. Especially for the important 5-hole probe and the platinum wire thermometer. The “normal” sensors for temperature, humidity, pressure and wind estimation are not enough.

The tethered balloon can ascend only with little wind anyway. So we put everything together and leave. The lead-acid batteries at the measuring masts are changed. The three-dimensional wind speed is measured using finger-shaped ultrasonic transmitters/receivers that are installed pointing at each other.

We meet at UNIS around noon (an awesome building) and start to pack. Logistics were an important part of the measurement campaign. Many people, equipment and clothing needed to be transported here, then brought back and forth.

The autopilot controlled the aircraft reliably the entire week and data was collected. That is now being worked on. We took off and landend manually as we always do. This allows to gradually enable/test the autopilot and have a precise landing next to the street near the ground station. Every step counts in the snow.

We have to stay in Oslo over night. In the morning the sun rises above the clouds over the Baltic Sea. So bright.

The classroom suddenly turns dark at the daily debrief in the morning. The power is off. That does not happen often but every now and then. There is a coal power plant at the edge of Longyear which supplies all houses with electricity and heat. Extra heating stations are distributed over the village to help out in such cases. An airlift evacuation is planned in case of a serious emergency. We help ourselves with flashlights and chalk.

The rescue helicopter is doing exercises with the winch over the apron while we await the return of the last commercial aircraft.

Our hangar was formerly used by Aeroflot. There are still boxes with Cyrillic letters standing around. A huge and noisy compressor makes each stay a little uncomfortable. It hisses constantly and turns on with a lot of drama every few minutes.

We head to the beach to fly ovals over the fjord and the ground north of the airport. The plane measures 70% relative humidity. This is not much in absolute amounts, only about 15% at room temperature in the buildings. It makes you want to drink constantly. The skin is dry and rough. The aircraft thrower must be careful that the fuselage does not slip through his (bare) hands during launch. The plane flies one track after the other without any issue.

The last measurement campaign was about the meteorological processes in Adventalen. This time it is about the air above and water in the fjords. In addition to the weather balloon and SUMO there are masts set up on the coast. Some students go out with the UNIS research vessel Viking Explorer and let sensors into the water

A 2m long “glider” (a constantly ascending and subducting autonomous submersible body) also takes part. The laptop in charge tinkles an alarming ship’s bell whenever new data arrives through Iridium satellites.

We set up our SUMO-station close to the water to fly as far as possible over the fjord. The swell gives us a strange feeling of summer vacation. There is routine in the flying. We empty 4 batteries in 4 hours. Every now and then we search the area around us for polar bears with a bright lamp.

Around nine in the evening the walkie-talkie asks for volunteers to go out with a speed boat. The Glider has detected moisture inside and will be brought in as a precaution. A little later the boat races down the fjord past us and brings the glider back safely. False alarm.

The glider does not have a propeller, the forward movement is generated through its stub wings and weight shift during the descent and ascent in the water. The system has many similarities with autonomous aircraft. The conditions are rougher, though. The time intervals are much longer – it can navigate through in the sea for months.

If you set you camera to a very, very long exposure time you get a hint of the northern lights, stars, the moon and a SUMO.

If you increase your cameras exposure time in the deep dark blue at noon you even get some mountains into the picture. Otherwise it is just dark. There will only be the moon in a few days. Ultimately there is no big difference to a Central European working day in winter. You go to work in the dark and leave in the dark. The only thing missing is the enlightened way to the canteen. There might be long-term effects, though.

The tethered balloon is located a few hundred meters west from our SUMO station. It is continously going up to 1000m and pulled down. The balloon ground station is now built up in our rented space in the airport. That’s where we have to go after the first SUMO flight, hack some software.

There are three computers in the aircraft: the autopilot, a data logger and a (commercial) data converter for the 5-hole probe (diameter: 3mm). It suddenly does no longer supply data, unfortunately. We never saw that in any of the campaigns. The laptop shows a “SD card not found” error from the converter. It works ok when the card is removed (huh?), then fails again. We add a status bit so that we can detect a malfunction and possibly reboot the converter before take-off.

Everything works smoothly for the rest of the evening. The SUMO is commanded to fly profiles (circles at different altitudes) over land and sea, then ovals in between that two points. In the end we fly ovals over the sea only. We are gaining more confidence in the batteries and reach flight times of about 45 minutes. That is a pity as it ruins the number-of-flight statistics.

The primary aircraft was somewhat imprecise in flight yesterday. We calibrate the gyro and acceleration sensors during the day. We could only do short (15 minutes) flights as the batteries were too cold. The pressure hoses enter the aircraft at the tip, just in front of the battery. We tight it with extra foam. The heat searation foam behind the autopilot is renewed. We isolate a plastic box with foam for battery storage and put heating pads between the lithium-polymer batteries.

We leave at seven. The two fragile sensors are screwed in at the airport and we head down to the campsite by the fjord. The trunk of a small van is our base. We test new batteries from HK. They cannot proved a high (but adequate) current, but 50% more capacity (at the same weight) than the current favorites from SLS . They seem to work fine even under this conditions. We achieve flight times of about 40 minutes and have a 10% spare. The planes work well and make their rounds over the sea and land reliably.

A lit up fishing trawler moves in the jord in front of us. There is a thin layer of fog above the water surface. The air temperature is -14°C, the ground is -20°C and the fjord is -3°C. This is not incredibly cold, but uncomfortable. Especially if you stand more or less motionless, looking at the sky and actually standing on that ground. It is really nasty when the wind is up. Thermos flasks do not only give you a warm but also non-frozen drink.

It feels a little like a holiday that is spent in the same place each year. We move into our room at UNIS. Some things are still there on the shelf where we have left them. There are three planes here this time. One is already equipped with the 5-hole turbulence sensor and the fast fine wire thermometer. A second aircraft is equipped to replace it in case of issues.

The permit to fly (for now only line of sight) was issued a few weeks ago. We can start immediately. UNIS has rented a small hangar at the airport in which also the balloon is parked. It was accompanied there on foot from the old auroral station on Sunday.

The last aircraft (a white Do-228) lands in front of our hangar at 7 and we are ready to go. The fast temperature sensor is mounted and we head down to the water. This time we received labor suits instead of snowmobile suits. It is easier to move in them (especially launching the plane by hand). But they are a little cooler if you are standing in the wind motionless. They will be replaced tomorrow.

The SUMO completes test circles and is landed on the nearly snow-free concrete lane. The surface is a bit rough but the platinum filament remains in one piece. We command it to fly profiles in three altitudes over land and the fjord, then ovals over land and sea. The platinum wire survives the landing in the snow but the folding carbon propeller gets broken. That had not happened yet. Seems that the tethered balloon winch does not like the fickleness of the current generator. We drive back to Longyear at around 11.

It was like a déjà vu. Everything works exactly the same way as 8 months ago. Start in Oslo, leave the plane in Tromsø, through passport control and back over the apron, landing in Longyearbyen.

You leave the plane and the cold, dry air beats you. Not this soggy-wet snow weather that winter usually starts with. No, really cold air and crunching snow. Every contact with metal ends with an electrostatic lightning.

We bring the bags to the guest house and see the passengers from the plane in the supermarket. It is off-season, no other aircraft on the apron, the bus almost empty.

And it is dark. The sun disappeared half an hour before landing. The city lights are reflected on the mountains a little, the moon barely manages it through the clouds. With an LED lamp you can illuminate some LEGO, nothing more.

We are again on the way to Svalbard to fly the SUMOs with the University. The first leg has brought us back to the airport hotel in Oslo. We will have the final leg tomorrow morning.

It is similar autumnal as in Germany (+8°C), slightly hazy. But not wintry.

We visited the IMAV2014 conference in Delft/The Netherlands. The Paparazzi project is closely connected with this event, the competitions were always important milestones for the progress. It is the meeting point not only for many Paparazzi developers.

The picture shows Piotrs fully autonomous Paparazzi Lisa/S.

The autonomous flight itself is no longer the main topic. The tasks are more complex and require work at higher levels such as image processing. There is a tendency to ready-made aircraft, designing the aircraft is no longer in focus.