Retired senior lecturer in the Department of Meteorology at Penn State, where he was lead faculty for PSU's online certificate in forecasting.
By: 24hourprof , 6:35 PM GMT on October 03, 2013
I was raking leaves yesterday afternoon as temperatures flirted with 80 degrees Fahrenheit here in central Pennsylvania. It's been quite a run of warm days and cool nights during late September and early October. I think about weather all the time, so you shouldn't be surprised that the mindless job of raking allowed me to come up with a good idea to blog about basic principles in meteorology.
After I finished raking, I started looking at data (you know me). The recent spell of cool nights in central Pennsylvania came courtesy of light winds, a clear sky, relatively low dew points, and increasingly longer nights. The high-pressure systems responsible for this early-autumn weather came from the west, and so I expanded my purview to include the northern-tier states east of the Rockies. I eventually found these classic plots (below) of upwelling infrared radiation (top) and air temperature (bottom) at Sioux Falls, South Dakota, on September 21, 2013. Let's learn some basic meteorology from this case...here's the corresponding meteogram that I'll refer to along the way. The meteogram spans from 00Z (6 P.M. CDT) on September 21 to 00Z on September 22).
The plots of upwelling infrared radiation (top) and air temperature (bottom) at Sioux Falls, South Dakota, from 00Z (6 P.M. CDT) on September 21 to 00Z on September 22, 2013. Units of upwelling radiation and air temperature are watts per square meter and degrees Celsius, respectively. Courtesy of Earth System Research Laboratory.
For the record, upwelling infrared radiation is the component of IR energy that's emitted upward by the earth's surface. Quantitatively, upwelling infrared radiation from the earth's surface is proportional to the fourth power of its absolute temperature (expressed in Kelvins). This relationship is often referred to as the Stefan-Boltzmann Law.
To get your bearings for these two plots, the units (vertical axes) for upwelling IR radiation and air temperature are watts per square meter and degrees Celsius, respectively. I note here that, although upwelling IR was calculated using values of absolute temperature, we're using degrees Celsius on the bottom graph because these units are more intuitive than Kelvins. Last but not least, the horizontal axes designate time...00Z (6 P.M. CDT) on September 21 (left) to 00Z (6 P.M. CDT) on September 22 on the right.
On this clear, mostly calm night at Sioux Falls (revisit the meteogram to verify weather conditions), the values of upwelling infrared radiation fell "in-step" with the decrease in air temperature. With no change in air mass on this night (check out the 06Z surface analysis), air temperatures decreased simply because the ground steadily cooled, emitting more infrared radiation (upwelling IR) than it received from the cloudless nighttime atmosphere. Put another way, the ground ran a negative energy budget at Sioux Falls during the night. The bottom line is that you can readily see the Stefan-Boltzmann Law at work...upwelling IR decreases with decreasing ground temperature (I'm using the air temperature as a proxy for ground temperature).
The 06Z surface analysis on September 21, 2013. Courtesy of WPC.
During the daytime hours, the increase in upwelling infrared radiation nicely mirrored the rise in air temperatures, which was attributable to solar heating (revisit the meteogram...there was no change in air mass as the 18Z surface analysis indicates and the sky was sunny).
If you look really closely at the data on the top graph around the time of the daytime maximum air temperature (between 21Z and 22Z...3 P.M. and 4 P.M.), upwelling infrared radiation has already started to decrease even though air temperature continues to increase slightly. What's up with that? As I stated earlier, the air temperature is only a proxy for the ground temperature (with regard to the Stefan-Boltzmann Law, upwelling IR depends on the temperature of the ground, not the air, but the two temperatures are pretty much in sync). Also keep in mind that the air temperature is measured two meters above the ground. So there's a bit of a lag between these two temperatures, which explains the slight "out-of-step" we observed around the time of maximum air temperature.
(Top) The plot of net infrared radiation at Sioux Falls, SD, from 00Z (6 P.M. CDT) on September 21 to 00Z (6 P.M. CDT) on September 22, 2013. Units are Watts per square meter. (Bottom) The corresponding plot of air temperature, in degrees Celsius. Courtesy of Earth System Research Laboratory.
I remember when I first started blogging at Wunderground (many thanks, Jeff Masters), and I wrote that a cloudless nighttime atmosphere emits infrared radiation toward the ground (downwelling IR). A typical radiating temperature of a clear nighttime sky is 250 Kelvins. To get a sense for infrared emissions from a clear atmosphere, check out (above) the corresponding plots of net infrared radiation at the ground and air temperature at Sioux Falls on September 21. By "net" I mean, "emitted IR by the ground (outgoing) minus IR from the nighttime sky absorbed by the ground)." Note that the net loss of infrared radiation at the ground was greatest around the time of maximum air temperature (during the mid-to-late afternoon).
In summary, radiational cooling (which I define as the net loss of infrared radiation at the ground) was greatest at Sioux Falls around the time of the maximum temperature of the ground (between 3 and 4 P.M. CDT on September 21). Earlier in the afternoon, the ground's temperature had been increasing because the ground absorbed more solar radiation than the infrared radiation it emitted. The lagging air temperature (measured at two meters) followed the lead of the ground temperature, lending credence to our observation that the maximum air temperature occurs around the time the net loss of infrared radiation from the ground is greatest. I know it's not intuitive, but that's the real deal here.
(Top) The plot of net solar (red) and net infrared (blue) at Sioux Falls, SD, from 00Z (6 P.M. CDT) on September 21 to 00Z (6 P.M. CDT) on September 22, 2013. Units are Watts per square meter. (Bottom) The corresponding plot of air temperature, in degrees Celsius. Courtesy of Earth System Research Laboratory.
To better visualize the individual roles of solar and infrared radiation, check out the plot (above) of net solar (red) and net IR (blue) and air temperature at Sioux Falls, South Dakota, on September 21. Note that the vertical axis (watts per square meter) has a more extensive scale to account for the large values of net solar radiation. By the way, net solar radiation means "Downwelling Solar" (incoming) minus "Upwelling Solar" (reflected).
I know I'm repeating myself, but the basic meteorology here is important...even though radiational cooling of the ground (net IR loss) was greatest at the time of the ground's maximum temperature (Stefan Boltzmann at work), the ground temperature was still relatively high because net solar gain was so much larger than the net IR loss. And the air temperature, which lags the temperature at the ground a bit, followed suit in fairly quick fashion..
What else can we learn from this case? Below, I inserted the Sioux Falls' meteogram so I can refer to them more easily. Okay, sometimes you'll here forecasters talk about the dew point being a rough lower bound for the expected nighttime low temperature on clear, calm nights. Focus your attention on the top rectangle of data and note how the temperature (the magenta plot) gradually decreases toward the dew-point plot (green), the two eventually "merging" around 12Z (6 A.M. CDT)...about the time when the low temperature of 36 degrees was measured. You can read the hourly temperatures off the scale shown on the left in the top rectangle of data. As it turned out, the low occurred between hourly observations during the six-hour period ending at 12Z (see the "T6NF" line of data).
The meteogram at Sioux Falls, South Dakota (KFSD), from 00Z on September 21 to 00Z on September 22, 2013. Courtesy of the University of Wyoming.
Another basic but interesting observation comes from the dew-point plot (green in the top rectangle of data). During the period from 00Z on September 21 to 00Z on September 22, the dew point at Sioux Falls, South Dakota, was pretty steady, vacillating only slightly around 40 degrees during the period (again, read temperatures and dew points off the scale shown on the left of the top rectangle of data). But once the sun rose and air temperatures started to climb (around 13Z or 7 A.M. CDT), the relative humidity, which was close to 100% during the wee hours of the morning, abruptly decreased after 13Z in tandem with the sudden increase in temperature after sunrise. To get your bearings, relative humidity is the blue plot in the top rectangle of data...you can read percents off the scale shown on the right of the top rectangle of data).
At any rate, my point here is that, even though the dew point (and the amount of water vapor) remained relatively steady, the relative humidity changed dramatically in response to changing temperature. The sensitivity of relative humidity to temperature is the primary reason I never showed relative humidity whenever I was on-air during my 20-year stint on public television.
Here endeth a lesson on basic principles in meteorology.
Comments will take a few seconds to appear.