This week we will spend less time on the mechanics of downloading and graphing data and more on trying to understand some fundamentals about the earth's energy balance. You should begin by reading chapters 4 and 5  in their entirety, as indicated on the syllabus. Read chapter 4 for Tuesday's class and chapter 5 for Thursday's class. The first, long part of this assignment is NOT REQUIRED but explains some of what we will be doing in class.

Pages 87-93 focus on the different forms of energy and the ways they are transmitted in the atmosphere or at the earth's surface. Pages 93-97 address the energy balance or energy budget for both shortwave and longwave radiation. It is important to distinguish between radiant energy and energy that has been absorbed and transformed into heat, which may take the form of either sensible heat (H) or latent heat (LE). The summary diagram in figure 4-11 illustrates how the entire budget can be balanced for the earth as a whole. As figure 4-13 shows, however, there are latitudinal variations in the balance that lead to poleward transport of energy through the circulation of the atmosphere and oceans.

Pages 97-103 focus on understanding the local energy balance and the concept of net radiation. This is calculated by summing up all of the incoming and outgoing (or downwelling and upwelling) shortwave (solar) and longwave (terrestrial) radiation. The part that has been absorbed and turned into heat energy is not included in this balance; actually, net radiation is what is left after you have subtracted all outgoing radiation from all incoming radiation, and therefore net radiation is what is available to be converted into heat energy. That heat energy may take three forms: latent heat (associated with melting of ice or, most commonly, evaporation of water), sensible heat (associated with temperature change), and ground heating and cooling (which we usually neglect as it tends to balance out to zero over time).
If you compare figure 4-4 (global insolation) with figure 4-17 (global net radiation), you will find that the places with the highest insolation are not necessarily the places with the highest net radiation. Furthermore the global patterns of latent heat (figure 4-18) and sensible heat (figure 4-19) are very different from each other. This will be discussed in class.

1. The components of net radiation and the daily cycle of radiation as it relates to the daily cycle of temperature will be explored on Tuesday in class by looking at the web site of the SURFRAD (surface radiation) network, which provides detailed field measurements of each component at locations from different climatic regions of the U.S. We will use the following procedure. You may try this on your own in advance if you wish; it is relatively easy and should take no more than about 15 minutes.

• select RAD/MET PLOTS to look at daily data, or MONTHLY MEANS to look as monthly averages over the year. For now let's look at the daily pattern but we'll try both in class.
• select a location and a date (if you don't enter a date, you will get data for the most recent day)
• select "direct solar" and "diffuse solar" and plot them together for each location you decide to examine. (Select "View Plot" rather than "Download Postscript" and then hit "Plot Data.") Note the daily cycle and the comparison between these two components. The sum of these two should be equal to the total downwelling solar, but if you add this as a third curve you will notice that sometimes the direct solar actually plots as being higher than total downwelling solar, particularly at the sites that are drier and have more direct sunlight.These three components are actually measured by different instruments, and therefore the discrepancy may be due to instrument error; I'm not really sure.
• now go back to "data display" and select "downwelling solar" and "upwelling solar" (upwelling solar is reflected shortwave radiation). If you plot these together with "net solar" you wil note that "net solar" is the difference between what comes down from the sky and what is reflected from the surface. At least it should be.
• again go back to "data display" and select "downwelling infrared", "upwelling infrared", and "net infrared"; when you plot these together, you should see that net IR = downwelling-upwelling and that typically there is more infrared going up than down, thus leaving you with a negative set of values for net IR.
• now plot "solar net radiation", "infrared net radiation", and "total net radiation"; total net radiation is the sum of solar net radiation and infrared net radiation. Since the infrared net radiation is generally upward and therefore negative, total net radiation will be less than solar net radiation. Print this page so you can compare it with the daily pattern of temperature, which cannot be plotted on the same axis and will have to be printed separately.
• finally, go back to "data display" and plot air temperature for the same location and date where you were working before. Look at the daily cycle of air temperature and try to relate it to the daily cycle of net radiation, focusing particularly on the timing of maximum and minimum temperature.  You can lay one graph over the other and hold them up to the light in order to compare them. How would you expect these two curves to be related?  You may also want to look at the "meteorological data plot" because it will give you some idea as to whether there are local shifts in weather pattern that are also affecting the daily temperature cycle independent of the local radiation balance. Would you expect the relationship between the pattern of net radiation and the pattern of daily temperature to be the same in a humid area as it is in a desert? Why or why not?

 

 

[Note: the graphs shown at this site all have the same time base; rather than local time, it uses UTC. The major drawback is that you are not seeing a complete daily cycle. Instead the graph begins with the tail end of the previous day, then shows the nighttime portion of the cycle, and then the daytime portion of the current day, but only up through late afternoon. If you want to see the rest you have to look at the next day's graph. However for days with relatively stable conditions and sunny weather, this will make less of a difference.]

3. On Thursday in class we will look at patterns of temperature change over time from individual stations, and we will compare these patterns for different geographic locations. In addition we will look at global temperature distributions and try to explain them in terms of the factors discussed in chapter 5.

Your required assignment for Thursday is to read chapter 5 in its entirety. We will do the investigative work in class and you will be asked to write up the results to be turned in by the following Tuesday. What follows is a description of the activities to be completed in class.

Pick a latitude for comparative study. Locate at least one continental and one maritime location at or close to the same latitude. If you're ambitious you might choose one west-coast location and one east-coast location as well as a continental interior location. This does not have to be limited to sites in the U.S.; there are both U.S. and global data sets available.

You may also choose a pair of stations at different latitudes but comparable maritime or continental interior locations, if you so desire.

Visit the Web site entitled "NCDC Climate Visualization". Under the heading CLIMVIS Graphics Choices, you will see options for both National Weather Service Summary of the Day and Global Summary of the Day. Pick whichever one is appropriate. You have the choice of plotting a time series or a contour map. Start with a time series option. Select the option that allows you to plot one parameter for a month from two separate stations in the same region or in separate regions. Pick your stations and plot both stations together on the same graph for a summer month and a winter month. (For the summer month pick either mean or maximum daily temperature; for the winter month pick either mean or minimum daily temperature.) Comment on any patterns that strike you as interesting or noteworthy. (For example, I tried doing this with Anchorage and Fairbanks, both in Alaska, using the minimum temperature for February 1997. The results are shown below. I also tried comparing Asahikawa, Japan with Zamin Ude, Mongolia. Those results are not shown here; you can look them up for yourself if you wish.)

Afterwards, go back and try a different graph type: "display the period of record for one parameter at one station" (only works for U.S. stations) or "display one parameter for a specified time frame" (up to one year; works for U.S. and global data sets). You can use this to plot temporal variations in maximum, mean, or minimum temperatures to show seasonal cycles. In the case of the U.S. data set you can also see multiple years laid out in sequence and get a real feel for year-to-year variation in temperature patterns. Do this for at least two stations, preferably the ones you have already been looking at.  (Here's an example: I compared maximum daily temperature at Phoenix and Flagstaff, Arizona for the period January-December 1995. If you want to make this kind of comparison for non-U.S. stations, you need to choose the "Global Historical Climatological Network Data" option.)

You can also make contour plots illustrating spatial patterns of climatic variables across the U.S. for specified dates. You might want to try experimenting with this as the contour plots may help to understand the relationship between the temperature trends at the two stations you picked earlier (only works if they are both continental U.S. stations. This is not required.)

Write up some general comments on how the observed patterns do or don't conform to what you would expect based on the book's discussion of factors affecting both seasonal patterns and spatial patterns of global temperature.

Next go to the "SSEC Sea Surface Temperatures" Web site and go to "Latest SST".  This is a beautiful color image showing the global distribution of sea-surface temperature. It is updated daily. There is an archive containing daily images for the last two weeks, but unfortunately this site doesn't retain images for long enough to compare patterns showing seasonal changes. Another site sponsored by NOAA does contain monthly SST images for the period between 1984 and 1996. Look here in order to see whether you can pick up any significant changes between images taken a few months apart.  How do they relate to the patterns you observed in the annual cycle of the net radiation balance? (This is a slow site; if the images take too long to load, don't worry about it.) Another, faster site with monthly SST images based on long-term average conditions is NOAA's "Live access to climate data."  Note in particular that although temperature varies mostly with latitude, there are significant anomalies that clearly are affected by factors other than latitude. This shows up as well in the following animation. You can also try looking at the animated view of sea-surface temperatures for the North Atlantic. This will show you the seasonal cycle of changing temperatures over a span of about 1.5 years. You can get a fascinating view of how temperatures reflect not only seasonal cycles of the global radiation balance, but also the dynamics of the ocean surface itself. Can you describe and explain the patterns shown here?