Numbers in context

4 December 2007

I am well accustomed to having difficulty finding a sense of scale for certain things (e.g., low temperatures, travel distances, the price of gasoline) as a result of living in The Other Two for twelve years. I was surprised to hear this morning, however, that I-5 is closed in Southern Washington today due to flooding. My surprise originates not from the closure itself, but from the descriptions of the storm that caused it:

A severe storm smacked the region Monday with hurricane-force winds and several inches of rain, and was blamed for four deaths.” (Yahoo News),

Rescue boats were used to grab flood-stranded residents, and GPS-equipped helicopters were used at night. ” (CNN, but from later in the same AP story)

The National Weather Service was certainly more sober: FLOOD WATERS WILL CONTINUE TO DRAIN OVER THE NEXT 12 HOURS BUT MANY PROBLEM AREAS STILL EXIST. THE FLOOD WARNING HAS BEEN EXTENDED INTO TUESDAY AFTERNOON.

My point in all of this was that the total amount of rainfall was somewhere between four and ten inches in something like twenty-four hours. Hilo barely even opens its collective umbrellas when that happens. I guess I’ll be adding ‘twenty-four hour rainfall’ to my list of ungrounded comparisons.

 

On a more scholarly note, I’ve been reading more Edward Tufte, and am greatly enjoying Beautiful Evidence now. In the chapter on corrupt techniques in evidence presentations, he talks briefly about fitting models to data. The part that struck me was this paragraph fragment:

For k explanatory variables, there are 2^k-1 possible fitted models, then multiplied by notions and so on through the [data] slack. Routinely 10^4 to 10^7 fitted models are available; all can be quickly computed and sorted over. These models are not independent and may look pretty much alike, but which few to publish from millions of possibilities? This latitude for evidence selection makes it difficult to distinguish between reliable findings and cherry-picking. Now and then it may matter.      

 Makes me remember fondly the Gemini science meetings on galaxy evolution. Also, latke party at our house tonight–if you’re in the area, stop by for the first night of Hannukah (sp? anyone?). 

Advertisements

One Response to “Numbers in context”

  1. Roberto Says:

    Yup. If Las Cruces got 1 inch of rain we would get all sort of flash flood warnings and the like. This place has no drainage system, which still surprises me. Man was not meant to live in the desert.
    Which model to publish. You see, what you have to do is minimize the sum of the squared residuals. This can sometimes require inverting matrices…….actually, I just publish the plot that looks the prettiest. I suspect many other “scientists” do the same thing.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: