I think it matters to ask “How fast does temperature track solar change?”.
The essential thesis of AGW is that temperature has a long lag time. There is heat “caught in the system” somehow. Either as “back radiation” from the sky or “whatever”. ( They are a bit vague on it, and I think that vagueness is an indication of error).
In a prior posting I was watching how the temperatures changed “over the spacial domain” around the San Jose International Airport. It looks like there is a clear and easily demonstrated Airport Heat Island.
In looking at various data about the airport thermometer at Wunderground, I thought that maybe I could look at daily data and find some “signature” of back-radiating-IR that ought to be there, but was missing. I’ve not finished that pondering yet (and feel free to suggest “possibles”… things like “Would back radiation show up as a later time of cooling in the evening after the sun has set” or “Would back radiation show up as a higher low in present data vs 1970s data”) But along the way I looked at Just One Day. Graphed above.
Look at it for just a minute. The top line is temperature. The bottom line is “change from the prior hour”. Basically, did it get warmer or cooler in that hour which just passed? My thoughts were about all that “back radiation” we are told exists, and was there some clear skew in the curve of changes to indicate some persistent slowing of the loss of heat, or more rapid onset in the mornings.
Sunrise was at: 5:46 AM
Sunset was at: 8:30 PM
Which if I’ve done it right puts “High Noon” at 1:07 PM local (daylight savings time).
Looks to me like about a 1 hour lag time. For about 1 hour after first sunrise, temperatures are still low. Change is nearly nothing, but perhaps smidgen of cooling still. At 9am to 1 pm we have our maximum rate of heating and the highest temperatures come about 1 to at most 2 hours later ( 2 pm to 3 pm). But by then, the rate of heating has already rolled over. Turning to net cooling by 4 pm with the sun still well overhead and evening sun for another 4 hours. Just at an ever shallower angle with ever less net heat input to the ground. By 10 pm, just 1 1/2 hours after sunset, maximum rate of cooling has passed and we’re in a very slow very minor rate of cooling into the late night.
I think I’ll do similar graphs for other days of the year, and for other places. ( Like dry desert areas without the ocean nearby to add water as a “Greenhouse gas”…)
But just from a simple first look at one day of data only (and a day with near zero wind speed at night and not much more during the day, so not a lot of air mass changes) it sure looks like there isn’t much room for “back radiation” to do, well, anything. At most I can see an argument for a 1 hour delay of maximum impact (though I think that is heating of the air mass not IR related), which would mean that at most it could shift the “almost zero net cooling” from 10 pm to 11 pm. Perhaps moving that “minimum” from 5 am to 6 am or shaving a couple of 1/10s degree of that corner at the start of daily warming. By 2 pm, anything that was done is lost in the daily cycle and the much higher rate of heat loss from the warmer air. That strong 4th power heat loss with temperature of the Stephan-Boltzmann Law just dominates. And does so in minutes to at most an hour. (The rate of heat gain is already dropping as we pass 2 pm and is a net loss closely thereafter).
So this “speaks to me”… but it just seems too obvious. Hasn’t anyone else taken a day with still air and just looked at how heat gain and loss changes over the cycle of the day? Noticed that it very closely tracks insolation in time? That there just isn’t any indications of heat “hanging around” for long periods of time and that the cool / cold nights rapidly stabilize at a point that looks more related to loss of heat from ground and structures than from the air? In essence, that 4 or 5 hours in the middle of the night with almost no change of temperature are when the ground is cooled on the surface enough that it is just holding heat loss to the air steady. In winter I’d expect an even longer flat period in the middle of the night, but with less daily heating and some net heat loss over time. In short, I think it is quantity of sun and absorption into the dirt that matters. The air temperature is a side effect.
We already have air temperature change as quite fast from earlier in the day as the sun is just starting to set, so we know the air changes more quickly than the temperature curve is changing. That implies that with a colder sky and colder air – less ‘back radiation’ – something else is slowing the decent into deep cold. I’d assert that “something” is just the dirt, buildings, and “stuff”. Rather like lunar soil moderates surface temperatures. In short: The specific heat and mass of the soil and buildings are far greater than the specific heat and mass of the air. The air cools rapidly to the point where surfaces dominate and stabilize the temperature.
That’s my thesis on what this graph says. Any other ideas? As I’m going to be poking around looking at other temperature histories and plotting up a few more of these “Daily Profiles”, now is the time to say if there’s somewhere in particular you would like to see. (Nome Alaska at the equinox and solstice dates might be interesting ;-)
For me, there just isn’t anything in that graph to speaks of “back radiation” and there’s a whole lot that speaks to mass of air, mass of dirt, specific heat, and rate of insolation. With what looks like about a 1 hour “lag” for the insolation near the Summer Solstice to soak in to the dirt.
To the extent the ‘dirt and sky’ thesis is true, then changes in ground cover, such as vegetation and snow removal, would be major influences on the temperature record. Airports would be among the worst possible places to measure “global warming” when compared to pre-airport times with grass fields and trees… Yet, that’s what they do…