Discussion: swedish rainfall

I just found out from SMHI that swedish rainfall statistics aren't available to punters like me for free, is that wierd? I'd been searching their website for ages looking for info on rainfall until I got desperate and sent them an email, 1,20kr per day it costs... Pity I didn't bring my BOM rain gauge I got from Terry over here with me.

A lot of the European meteorological services are pretty commercialised - and it doesn't necessarily match up well with a country's political leanings - some of the most commercialised are in places like Sweden and Germany. It's slowly getting easier to get European info but it's still very poor compared to what's on line for Australia or North America.

There are a couple of international sites I often use to get data for places where the national sites aren't of much use - www.ogimet.com and www.weatheronline.co.uk. However, I've just looked at a number of Swedish locations on those sites (including Sundsvall) and found very few that report precipitation (there's some good temperature info though, especially on ogimet). My guess is that a lot of the Swedish observing sites are automated, and automatic stations are notoriously bad at measuring snow, even if the gauge is heated - I'm currently writing a paper comparing the readings from two sites (one automatic, one manual) at Falls Creek.

If you can wait until the end of the month, there are graphs of daily data for a number of sites around Sweden in the monthly report on the SHMI website. There's a decent news feed on the Norwegian Meteorological Institute's website too (I look at it quite regularly, and am slowly learning bits of Norwegian from it - although I still send the tricky bits to Cassie to translate), although it's obviously focused on Norway.

It seems to me people would use the data a lot more if it was free on their website. That's interesting you are writing that paper, I had been wondering how accurately snowfall could be measured and how it was done. Thanks for the tips.

That's what we think too (and if our experience is anything to go by, collecting small amounts of money to process data requests is more hassle than it's worth). You'll probably recall that the ABS drastically changed its policy on that a couple of years ago too.

The manual site at Falls Creek records about double the precipitation in winter that the automatic one does, but as that includes some rain, I suspect the ratio for snowfall is more like 3 or 4 to 1 - I still need to do the bit of the analysis which involves separating out the days which were cold enough for snow. The two sites record almost the same amount of rain in summer so I don't think differences in site exposure can explain it.

Measuring snowfall is tough because it's so hard to separate blowing and falling snow, especially when it's windy - the type of gauge used and how it affects the local wind field around it makes a big difference. The Finns did an experimental study a while back with about 15 different types of gauges set up in the same field and came out with a ratio of something like 2:1 between the largest and smallest readings. It's probably impossible to achieve absolute accuracy, so the best you can hope for is to have consistency over time so you're comparing apples with apples. (Most of my scientific life - at least the bits which aren't spent talking to the media - is spent making sure that when it comes to climate change that we really are comparing apples with apples, and not oranges or watermelons).