March 18, 2010

Diurnal Variations


Now we are in a dead weather period (no precipitation until Sunday), it is good time to look at the typical daily (diurnal) variations of various weather parameters. One of my favorite charts is shown above...a 24-hr plot of the surface observations at a station (in this case Seattle Tacoma Airport) for the past day (ending Thursday afternoon). Time is in UTC (aka GMT or Z time), subtract 7h to get PDT (12 UTC is 5 AM, 00 UTC is 5 PM). The top line is wind speed....sustained (black) and gusts (red). In meteorological parlance sustained winds are usually averaged over 2 minutes, with gusts defined as the highest 3 or 5 second wind during that period. Why are their gusts? Generally from the turbulent mixing of higher winds down from aloft.

Notice something about the winds? They tend to be stronger during the day. Why? As discussed in an earlier post, there are several reasons, the most important being the atmosphere is less vertically stable during the day when the ground is being heated. As a result there is more mixing of higher winds from aloft.

Temperature and relative humidity are mirror images of each of other. The reason? When temperature increases the air has more capacity hold water vapor. Since relative humidity is the amount of water vapor in the air divided by its capacity, increases temperature (and thus increasing capacity causes the RH to drop).

What about pressure at the surface? Here is an interesting thing. If there are no weather systems over an area the surface pressure still changes! There is a typical diurnal cycle with pressure being higher at 10 AM and 10 PM and lowest at 4 AM and 4 PM. The highest high is 10 AM (+or- one hour) and the lowest low at 4 PM. Why? The biggest forcing is the diurnal tide in the stratosphere. It is like a lunar tide in the ocean (two highs and two lows), but instead of being forced by gravity, it is the heating of the stratosphere that does the forcing.

You can see the solar radiation at the bottom. It looks like a cosine function (for those of you who remember your trig), with the few imperfections in the shape due to clouds.

Finally, the second strip of observations is wind direction. Not how it flops around when the winds are light. In fact, you really can't believe wind direction for most wind vanes until the winds get above roughly 2-3 knots.

Finally, you can help in an important study:

A group of UW graduate students created a survey directed at western Washington residents with two main goals:

1) Determine how often, from where, and why residents seek weather information
2) Determine residents' level of understanding associated with weather terminology

The information gathered by the questionnaire will help guide more effective techniques for communicating important weather and flood information. The following link connects to the survey. Thanks for considering this.

http://www.surveymonkey.com/s/G3FKPDG

5 comments:

  1. won't your readers bias this study toward those who are knowledgeable and who check various srouces?

    ReplyDelete
  2. Interesting survey. It motivated me to look for an alerting NOAA radio.

    ReplyDelete
  3. Cliff, I think SLP diurnal variability is a combination of standard (gravity) tides PLUS some stratospheric heating. How else to explain the TWO cycles per day?

    ReplyDelete
  4. Winners of the survey drawing win a tour of a forecasting office? How boring is that? I'd rather have lunch with Cliff Mass and an autographed copy of his book...

    ReplyDelete
  5. Filling out a survey usually grants one the opportunity to see the results. No such courtesy on this survey. No information of if or where the survey will be made public. This really should be addressed.

    ReplyDelete

Please make sure your comments are civil. Name calling and personal attacks are not appropriate.

A Near Perfect Forecast of Yesterday's Event. The Next Windstorm Comes into View

 The next time someone makes a weatherperson joke, remember the nearly perfect forecast for yesterday's wind event over Washington. No l...