That’s it. At 01:30 UTC on 17th July, we finished our DXpedition to Svalbard. So, how did we do?

The raw numbers

On the face of it, this wasn’t our most successful DXpedition in terms of QSO count. During the six full days we were QRV, we only made 6305 contacts between the six of us.

That hides a significant fact, however: this far north we never really expected to make a huge number of contacts. I’ve written more about propagation below, but to summarise only 30m, 20m and 17m were really open.

We had decided before we came to split into two groups, therefore: On Sunday, M0VFC, M1ACB and myself went on an excursion to Barentsberg allowing G3ZAY, M0HSW and M0TJH to stay in the shack. We did the opposite on Monday. Then, on Tuesday and Wednesday, we adopted the same pattern for our excursion to Pyramiden. This means that for most of the time, there were only three people in the shack.

Perhaps more surprising for us was the 44:56% SSB:CW (2858:3447) split. As it happened G3ZAY, M0VFC and I made almost exclusively CW QSOs, leaving M1ACB, M0HSW and M0TJH to do almost all the voice operation. That wasn’t planned as such, but worked well from our point-of-view.

Propagation

Propagation in Svalbard in the summer is weird. First of all, there is no nighttime during the months of Midnight Sun, so nothing below 30m ever really works. We made just 29 QSOs on 40m, and none on anything lower.

30m had some good openings during morning and evening but was pretty dead during the main part of the day. Throughout what would be the ‘night’, there was often good propagation to the US, which we could see from the 30m APRS beacon we operated.   965 (15%) contacts were on 30m.

20m was the most reliable band and was often open to JA, EU and NA at the same time. Even it went through a quiet period from mid-morning until mid-afternoon however. 3534 (56%) contacts were on 20m.

17m was often open during the afternoon to JA and NA, and then EU in the early evening. There was almost nothing here during the mornings. 1772 (28%) contacts were on 17m.

15m and higher were pretty much constantly dead. We made just 5 CW QSOs on 15m.

Overall, however, the main thing we noticed was that propagation was hugely susceptible to the K-Index. At a K of 5, most bands were pretty unusable. On days when K went down to 2, we made many more contacts, except for our final day which seemed to buck that trend. It might be that flux collapsing to 100 was responsible for that.

Logging

We were logging with Win-Test. With six operators using their own callsigns, you might have assumed we had six laptops but with only three stations we didn’t do that. We instead adopted the approach we’d used successfully in VP9: all copies of Win-Test were networked together and we all logged into a single log. We ‘logged in’ to Win-Test for each of our shifts, however, so we could then separate the logs based on the operator callsign.

During the latter part of the trip, we were uploading logs automatically every 30 minutes to Clublog and LotW. We’ll blog more about how this was done later (and I think a presentation for the RSGB Convention is planned), but basically we had a program listening to Win-Test’s network, capturing the QSOs, compressing them, and then batching them up.

APRS

We always like to innovate in some way on our DXpeditions, and this year (in addition to the live uploads), we also experimented with running an APRS beacon using MFSK16 on 30m. This was less successful than we had hoped given propagation on that band, but it was useful in that I could check how open 30m was to the States when I woke up. I could have used WSPR for the same result, though, assuming internet connection.

That’s kind of the point, though: had we not had internet access (which we weren’t expecting to have), APRS could have given us a route to do basic messaging if required. I think it’s something we will continue playing with to see whether the technology can bring anything extra to DXpeditions.

Thanks to members of the Cross Country Wireless email reflector for running IGates for our packets, and especially to Chris G4HYG for writing APRS Messenger, which we used for the beacon.

And finally…

Photos

My photos from the trip are in a set on Flickr.

Aerial photography

Hugo, M0HSW, brought a drone with him as his hand-luggage. We might have ridiculed him at the airport a little bit, but the film of aerial shots of Svalbard (Longyearbyen and Pyramiden) that he has produced is spectacular. Viewing it is highly recommended:

After our trip to Barentsberg a couple days previously, yesterday we went to Pyramiden, named after the Pyramid-shaped mountain above the town. Like Barentsberg, this is a Russian coal-mining settlement. Unlike Barentsberg, however, Pyramiden is a ghost town.

Or almost a ghost town. The mine closed in 1997. Under the Svalbard Treaty, if a settlement is abandoned for 10 years, the country responsible has to remove all trace of the town and return the area to its natural state. In early 2007 – 9 years and 10 months after the town had been abandoned – the Russian government decided it was cheaper to pay for 2 people to stay there, repairing buildings, than removing all trace of a town for over 1000 people and the associated mine. Today, the population has increased to 8. They get all their supplies from Barentsberg and, apart from one square metre on the harbour where there’s mobile telephone coverage from Longyearbyen, the only communication is by letter twice a month or satellite phone for emergencies.

Nevertheless, to all intents and purposes, this is a ghost town, and a fascinating relic of a remote outpost of the Soviet Union. Recent polar bear visits to the town meant that we weren’t free to explore; we had to travel as a group with our guide, rifle in hand, just in case.

For anyone wishing to do this trip themselves, this excursion was organised by Spitsbergen Travel and comes highly recommended. The guides are excellent. If you’re wondering whether to do both Barentsberg and Pyramiden, I’d say emphatically, yes. The towns might both be Russian, but that’s where the similarities end. I found Pyramiden felt more comfortable as its ghost town nature is explicable with the collapse of mining after the fall of the USSR, unlike Barentsberg which just feels out of place.

After two flights (London – Oslo, then Oslo – Longyearbyen), we’ve arrived in Svalbard. This is the view from the shack, taken just after we arrived (about 00:30 local time). The midnight sun is very real!

The first surprise, we have mobile phone signals and the shack has internet. So, our fears about lack of connectivity were slightly misplaced. Nevertheless, we’re still going to be uploading logs via the Iridium satellite link, and we’ll be doing APRS over HF, because we can.

The JW5E shack has been slightly reconfigured since the video W2NAF took last year but some of the antennas no longer work efficiently. As I type, this is the current situation (which may change over the week):

• We’ve installed a new 30m dipole. Rob M0VFC is on 30m CW with the shack’s FT-1000MP Mark 5.
• Steve M1ACB is using the 40-6m vertical to operate 17m SSB with his Elecraft K3.
• Hugo M0HSW is on 20m using the 15/17/20m beam using Rob’s K3 and a KPA500.
• My IC-706 is set up as the data station for later on. It will use any antennas that are free.

We’ll be on during the IARU contest – we’ll be giving reports but we aren’t entering: with so few bands open this far North and multiple callsigns between us, we wouldn’t be able to be competitive.