Lost in the Arctic

In this post we look at a challenging map situation and how some cartographic choices got in the way of a good article.

While perusing the web recently, I came across an article on the lack of ice regrowth in the Arctic this year.  It’s highlighting that at this time in the northern hemisphere, sea ice should be expanding as the temperatures begin to drop, but is not doing so this season.  There was a nice graphic showing the changes in seasonal sea ice extent which was instructive.

No problems so far (Ed. Um, well…). What threw me for a loop was a map that was included in the article.  As usual, the caption for the map was at the bottom so when I first scrolled down to see it, I wasn’t really anticipating what it was showing – here’s the map:

Now I don’t know if it was just me or not, but at first glance, I had noooooo idea what this map was showing.  There was nothing familiar to hang my perception on – what are the big grey areas?  And all that white?  Are they countries?  Or provinces of the “Central Arctic”?  Honestly, I stared at it for a good 30 seconds (a lifetime in most webpage attention spans) before I could make sense of it.  And then finally the penny dropped – we’re looking down on the north pole and each of the coloured zones is a designated section of the Arctic Ocean.  The map caption gives it away: “The regional seas of the Arctic Ocean.”  As I found out in some later searching,  “scientists often refer to the different seas within the Arctic Ocean when they discuss sea ice extent”.  Even knowing what it’s showing, I still find this map very disorienting; there are two main reasons for this.

First, we’re looking at the surface of the earth from a very unfamiliar vantage point – above the north pole.  With most map projections, we’re looking sort of side on and land masses take on more familiar shapes.  But in this map, the North American and Asian/European continents are seen from a very different angle.  Couple this with the fact that the lines of longitude converge on the pole and therefore our familiar rectangular map area becomes circular (even though the map is cropped to a rectangle).  This is a particular challenge of mapping at the poles.  There are polar map projections that facilitate mapping at these extremes, like the one shown below:

(Nice to see New Zealand at the TOP of a map for a change, even if we are upside down. [Ed. who’s to say what’s up or down?])

But  I think this in only part of my dissonance.  The other part relates to symbology, or the colour and symbol choices.  In this case, the way that the map features have been shown contributes to some of my confusion, and I think this mainly relates to the continent/ocean relationship.  The aim of this map is to make it easy for the reader to understand where the different regional seas are.

In my classes I talk a lot about the background and foreground when making maps.  It’s a bit like movies.  While the filmmakers wants us to focus on the actors and the action, they also includes background music and sound effects.  If the filmmaker does their job right, we shouldn’t really notice the background music but we should definitely feel the effect.  Take your favorite movie scene and then remove the music.  Does it have the same effect?  Remove the Blue Danube from 2001: A Space Odyssey, or the soundtrack from Dunkirk and you have a very different effect.  (Or a particular favourite of mine, Gorecki’s Symphony of Sorrowful Songs in Peter Weir’s Fearless – I’m sure you’ve got your own favourites – maybe you could add a comment?)  If you’re conscious of the background music, then you’re not focusing on the action on screen, and the effect is broken.  With mapping, the “background  music” is often the other data layers that provide you with some spatial context.  In this map, I’m feeling very distracted by the background music, which, here, is the continents, and the “negative space” of the oceans.

I dug a little bit deeper into this – the source for the map is the National Snow and Ice Data Center and they hold lots of data from both poles.  I went in search of some of their data to work with and came across this map:

This alone is a massive improvement (IMHO) – and it’s the combination of two things: good colour contrasts between land and sea (with sensible, meaningful colour choices) and labelling.  This map uses different labels than the one we first looked at and labels both land and sea(s). I’m in two minds about the blue colour – the deep blue is quite strong and maybe distracts a bit but could also be a signifier of veeeery cold water temperatures.  Note, there are no country boundaries – a cartographic choice.

In many ways, this map is more effective than the original one we saw above.  It clearly shows the same information but is more informative and easier to understand.  If I were to try and improve on this map (which I can’t really, as I’ve not been able to find the raw data used), I might consider adding in some lines that could further demarcate the seas (but maybe not use separate colours for each) and possibly add in lines of latitude and longitude, to provide a bit more spatial context – this wouldn’t be essential).

I don’t mean to be overly critical of the cartographers here: map making is perhaps a bit more art than science, and a process fraught with choices to be made and their effects to be lived with.  And mapping polar areas poses some very specific problems.  For a site like Mashable, where readers come from a wide variety of backgrounds, the care and feeding of maps like this becomes important.  In my case, I was so focused on trying to understand this map, that I think I missed the overall impact of the article.  This may also be a good reminder that more often than not with maps, less is more.

(And a huge hat tip to NSIDC – they have made a lot of useful Arctic and Antarctic data available on their site – more of that please.)

C