The Map

I mean come on, guys, did you really expect me to not touch this one?

Well we made it to Friday, and naturally in the not so serious we have to cover the sharpie map. Because, if the data does not agree with your opinions, clearly the correct response is to just make shit up.

By now you have probably all heard the story about how President Trump tweeted an incorrect forecast about the path of Hurricane Dorian, warning how Alabama could be “hit (much) harder than anticipated”. Except that the forecast at the time was that Alabama wasn’t going to be hit. Cue this map, days later. As in days. As in this news story continued for days.

Note the sharpie weirdly extending the cone (in black, not the usual white) into Florida and onward into Alabama.
Note the sharpie weirdly extending the cone (in black, not the usual white) into Florida and onward into Alabama.

So to be fair, I went to the NOAA website and pulled down from their archive the cone maps from the date of the graphic Trump edited, and the one from the day when he tweeted about Alabama being hit by the hurricane.

Important to note that this forecast dates from 29 August. This press conference was on 4 September. He tweeted on 1 September. So in other words, two days after he used the wrong forecast, he had printed a big version of a contemporaneously two-day old forecast to show that if he drew a sharpie line on it, it would be correct.

Here is the original, from the National Hurricane Centre, for 29 August. Note, no Alabama.

No Alabama in this forecast, the OG, if you will (and if I'm using that term correctly).
No Alabama in this forecast, the OG, if you will (and if I’m using that term correctly).

And then Trump tweeted on 1 September. So let’s take the 02.00 Eastern time 1 September forecast from NOAA.

By 30 August the forecast was already tracking northward, not westward. So by 1 September the idea that the hurricane would hit Alabama was just nonsense.
By 30 August the forecast was already tracking northward, not westward. So by 1 September the idea that the hurricane would hit Alabama was just nonsense.

Definitely no Alabama in that forecast.

This could have all gone away if he had just admitted he looked at the wrong forecast and tweeted an incorrect warning. Instead, we had the White House pressuring NOAA to “fix” their tweet.

Now we can all chalk  this up as funny. But it does have some serious consequences. Instead of people in the actual path of Dorian preparing, because of the falsely wide range of impacts the president suggested, people in Alabama needlessly prepared for a nonevent.

But more widely, as someone who works with data on a daily basis, we need to agree that data is real. We cannot simply change the data because it does not fit the story we want to tell. If I could take a screenshot of every forecast and string them together in an animated clip, you would see there was never any forecast like the sharpie forecast. We cannot simply create our own realities and choose to live within them, because that means we have no common basis on which to disagree policy decisions that will have real world impacts.

Credit for the photo goes to Evan Vucci of the AP.

Credit for the National Hurricane Centre maps goes to its graphic team.

Snowfall in Philadelphia

Today, 9 February, it finally snowed significantly here in Philadelphia. In Chicago it probably snowed shortly after I moved out in September. Today’s graphic is a forecast map from philly.com using National Weather Service (NWS) data.

Snowfall in Philadelphia
Snowfall in Philadelphia

I fail to understand the divergent palette—to be fair this is not the only instance of it throughout the meteorological world. There is a split at the six-inch mark—but why? If anything, my eye would think that the 4–6 range is the heaviest, not the yellow. Snowfall is usually more of a continuous range, and so both within the blues and yellows you get that through a softer edge as the colours become more intense. And then you hit the six-inch mark and a violent shift.

I am also curious as to why the choice to use a coloured map background. Especially if the colour, a lightish green-blue is so close to the lightish blue used by the map to forecast snowfall.

In short, I think this Philadelphia map could use some attention from some designers to make the message a bit clearer.

Credit for the piece goes to the philly.com graphics department.

The Polar Vortex…Or Not

If you live in the United States, you probably have heard the term polar vortex by now. People have been using the term to describe the bitterly cold temperatures affecting the eastern two-thirds of the country. But the term polar vortex is a meteorological term that means a specific phenomenon. In other words, it’s more than a hashtag along the lines of snowmageddon. The Washington Post explains what the polar vortex is.

The Post explains the polar vortex
The Post explains the polar vortex

My caveat for this piece is that from my understanding, it’s not entirely correct. Or perhaps not detailed enough. In brief, we turn to the National Weather Service out of New York for a more technically correct, but more poorly designed graphic.

National Weather Service explanation
National Weather Service explanation

Poor type, poor colours, poor hierarchy. Those are abundantly clear, but the important part is that the NWS wants to correct many of the popular misconceptions. Somewhere between the well-intended but less-than-entirely-clear Post piece and the accurate-but-also-unclear National Weather Service piece is an opportunity to explain the concept to the public.

Credit for the Washington Post piece goes to Bonnie Berkowitz, Richard Johnson, Katie Park, and Gene Thorp.

Credit for the National Weather Service goes to the graphics team at the National Weather Service of New York.