Ice on puddles when low temperate is above zero?
February 9, 2015 4:09 PM   Subscribe

I've been seeing ice on the tops of puddles when the low temperature for the day is forecast to be low (2 or 3 degrees) but not freezing. Why is that?

I had assumed that this was just because a forecast is a prediction, and in reality the temperature went below freezing. However, it seems to happen a lot, and, come to think of it, soil is usually cooler than the air above it1, and I've been to ice caves, where there are icicles despite the above-ground temperature being in the teens, so is it actually that the water is freezing because the ground is actually colder than the low air temperature for the day?

1: But maybe soil isn't actually cooler than air, but just seems that way due to having a higher thermal coefficient?
posted by Bugbread to Science & Nature (7 answers total) 3 users marked this as a favorite
At night you get radiative heat loss to the cold sky that can bring the temperature of the ground to below the air temperature. From wikipedia "The same radiative cooling mechanism can sometimes cause frost or black ice to form on surfaces exposed to the clear night sky, even when the ambient temperature does not fall below freezing."
posted by doctord at 4:20 PM on February 9, 2015 [3 favorites]

This may help a little (at least in terms of understanding how temperatures are measured).

It's my understanding that surface temperature tends to fall below air temperature during the night, then rise above air temperature during the day, and that this is mostly a result of the ground's greater thermal mass vs. that of the air, which results in the ground heating up more slowly than the air above it.
posted by pipeski at 4:23 PM on February 9, 2015

The clear black sky is very cold. You won't see this happen when it's cloudy.
posted by ryanrs at 4:24 PM on February 9, 2015

If the air is dry, in addition to radiative cooling, puddles will also be affected by evaporative cooling.

In addition to these other factors, it's also entirely possible that your particular area is consistently cooler than wherever it is your forecast is being predicted for. You can test this by measuring the air temperature in the shade for a while and seeing how your highs or lows compare to the predictions.
posted by aubilenon at 4:34 PM on February 9, 2015 [1 favorite]

Air temperature is not the same everywhere. You can get local differences for example due to shade. Where they're measuring is also not right where you are, so where you are could be a couple degrees colder. Those are a couple reasons the actual temperature may be colder. However there are many other factors.

Solids (the ground) heat up and cool down more slowly than gases (the air). There is the latent heat of melting/freezing, which basically means it takes more energy to change phases from solid to liquid, so ice takes longer to heat up enough to melt, and an air temperature temporarily above freezing may not be enough to melt ice.
posted by DoubleLune at 4:58 PM on February 9, 2015 [1 favorite]

Official temperatures are taken at about 5 feet above the ground level. The temperature at the ground can be colder than the official air temperature during clear, calm, nights. If there is a breeze, the air tends to be more well mixed, eliminating these differences.

Temperatures at night also vary quite a bit over very short distances. These differences are again greatly amplified on clear calm nights with good radiational cooling and also are dependent on landscape (elevation and development). I did an experiment last fall on a clear calm night. I live about 2 miles from the ocean in a rural low elevation area. There is a small urban area about 5 miles away, also on the ocean. At about midnight, I got in my car and drove around....down by the water, into the urban area, up to higher elevations and back down into lower areas. My car thermometer was going crazy. There was a difference of about 12 degrees total--all within a 30 minute drive and all within about a 5 mile radius. So if your official forecast says the low will be 35, it is entirely possible that many areas within that area will drop well below freezing. As an example: compare the temperatures at Dulles Airport with Reagan National. During the day, they are more or less the same. But at night, Dulles is significantly colder on average. Example: For January 2015, the average low at Dulles airport was 21, but at Reagan National not too far away it was 29. That is a large difference due mostly to National being located within the urban heat island.
posted by Seymour Zamboni at 5:14 PM on February 9, 2015

Thanks! Mystery solved!
posted by Bugbread at 5:14 PM on February 9, 2015

« Older What to do after Tenure?   |   Acknowledgement of research work in professional... Newer »
This thread is closed to new comments.