METR 104:
Our Dynamic Weather
(Lecture w/Lab)
FINAL PROJECT (An Investigation):
Why Does West Coast Precipitation
Vary from Year to Year?
Dr. Dave Dempsey,
Dr. Oswaldo Garcia,
& Denise Balukas
Dept. of Geosciences
SFSU, Fall 2012

Part III: Jet Stream Patterns during El Niño and La Niña Events

Part III of the Final Project completes the background investigative work needed for the Final Project. (A separate document describes the format of a report summarizing the results of your investigation.)

The Final Project is a research project broken into four distinct parts:

  1. Part I: Analysis of precipitation data (done for you, but you need to understand how it was done in order to interpret it properly).
  2. Part II: Analysis of Pacific equatorial sea-surface temperature and statistical connections to precipitation data (completed in lab on Wed., Dec. 12 or Fri., Dec. 14)
  3. Part III: Jet Stream Patterns during El Niño/La Niño Events (in lecture Mon., Dec. 17)
  4. A final, summative report (due on Friday, Dec. 21; we provide you with a template)

Overall Objectives:

Objectives for Part III:



In Parts I and II of the Final Project, you probably discovered the following:
  1. The amount of rainfall that West Coast weather stations receive during the five wettest months of the year can vary quite a bit from year to year.

  2. For stations in some regions, at least, there might be statistically significant connections between rainy season precipitation totals and the occurrence of some types of El Niño and/or La Niña events (that is, ENSO, or El Niño/Southern Oscillation, events).

However, statistically significant connections don't, by themselves, demonstrate cause and effect—for that, we have to demonstrate that there is also a physical connection.

One possible physical connection between ENSO events and West Coast rainfall is through ENSO's influence on atmospheric temperature patterns in the lower troposphere. These influences most directly affect the tropical Pacific Ocean but can affect other areas indirectly, too. Here's how the connection might work:

  1. During El Niño events, the higher-than-normal sea surface temperatures (SSTs) in the central and eastern tropical Pacific should warm the lower atmosphere there through:
    1. increased conduction of heat into the atmosphere from the sea surface, and
    2. increased evaporation from the ocean surface (which converts heat in the ocean into latent heat in water vapor, cooling the ocean surface), followed by condensation of the increased water vapor to form clouds (which converts latent heat back into heat, warming the atmosphere where the clouds form); and
    3. increased emission of longwave infrared (LWIR) radiation from the surface, and hence increased absorption of LWIR radiation (especially in the lower troposphere, where most of the water vapor is).

    During La Niña events, the colder than normal sea surface temperatures (SSTs) in the central and eastern tropical Pacific should produce cooler than normal temperatures in the lower atmosphere there through:
    1. reduced or even reversed conduction of heat between the atmosphere and the sea surface; and
    2. reduced evaporation from the sea surface, and hence reduced cloud formation, and hence reduced latent heat release in the atmosphere; and
    3. reduced emission of LWIR radiation from the surface, and hence reduced absorption of LWIR radiation in the lower troposphere.

  2. Recall that the polar front is a narrow zone of relatively large temperature contrast (large temperature gradient) in the lower troposphere between the tropics and the poles, normally found at midlatitudes. Warming of the lower troposphere in eastern tropical Pacific during El Niño events should create a temperature gradient between the tropics in that region and the midlatitudes, a region where the temperature gradient is normally very weak or absent. This should shift the latitude of the polar front farther south, or perhaps create a second, more southern branch of of the polar front.

    Cooling the lower troposphere in the eastern tropical Pacific during La Niña events should weaken the (already weak) temperature gradient between the tropics and midlatitudes, which might leave the polar front farther north than usual.

  3. The pattern of pressure aloft between the tropics and midlatitudes should shift along with the polar front, because temperatures in the lower troposphere largely determine the pressure aloft. In particular, the narrow zone of large pressure gradient aloft that occurs directly above the polar front, might shift southward or form a southern branch during El Niño events and shift northward during La Niña events, following the polar front.

  4. As the pattern of pressure aloft shifts, the pattern of winds aloft (in particular, the location of the jet stream, which forms in the zone of large pressure gradient directly above the polar front), should shift as well. During El Niño events, we might see the jet stream shift southward or form a southern branch in the eastern Pacific.

  5. Since midlatitude cyclonic storms track along the jet stream, and midlatitude cyclones bring most of the rainfall received on the West Coast, any alteration in the jet stream position might affect rainfall patterns on the West Coast.

One relatively simple test of this possible physical connection is to analyze upper tropospheric wind speed data to see if the average jet stream position during the rainy season during El Niño and during La Nina events differs from the jet stream's overall average position during the rainy season. If it does differ, and in particular differs in ways consistent with changes in observed patterns of rainfall during El Niño and/or La Nina events, then we will have confirmed (but of course not proven) the hypothesis that ENSO events affect the latitude of the jet stream, and hence midlatitude cyclone tracks, and hence rainy season precipitation totals. That's as far as this project will go, but confirming the possible explanation would help justify searching for more evidence, which is how it works in science!

Instructions for Part III

The National Atmospheric and Oceanic Administration's Earth Systems Research Laboratory (ESRL), in Boulder, Colorado, provides Web access to many years of atmospheric observations analyzed originally for use with computer forecasting models. Among other things, the Web site allows you to construct "composites" (by which ESRL means averages over time of spatial patterns) of a variety of atmospheric quantities, including wind speed at various levels in the atmosphere.

We will take advantage of ESRL's Web site to test the hypothesis that ENSO events influence the jet stream along the West Coast in winter in ways consistent with statistical connections between rainfall and ENSO events at some West Coast weather stations.

To do this:

  1. Access ESRL's Monthly/Seasonal Climate Composites Web site at

    (Alternatively, to get to this page step by step:
    1. start with ESRL's Physical Science Division at;
    2. from the menu of links across the top of the page, pull down the "Products" menu and select "Plotting and Analysis", which gives you access to a wide range of different sorts of data and ways of analyzing them;
    3. click on the link to "Monthly/Seasonal Mean Composites".)

  2. Specify the quantity that you want to analyze and plot:

  3. Specify the level in the atmosphere where you want to analyze the wind speed:

  4. Specify the period of particular months of the year (the "season") during which you want to analyze the wind speed at 300 mb:

  5. Specify the range of years for which you want to compute a composite average of 300 mb wind speed during your five-month rainy season:

  6. You are going to create a "color-filled" contour plot, which is a contour plot (of lines of constant wind speed, or isotachs) in which the area between each pair of adjacent contour lines is filled in with a different color. Specify a plot color:

  7. The wind speed data available from ERSL's Web site is in meters per second. One meter per second is almost 2 knots (or 2.24 miles per hour). By convention, the jet stream is defined to be a relatively narrow "tube" of air aloft moving with a speed of at least 60 knots, which is about 30 meters/second. However, the jet stream position can vary somewhat from one day, week, month, and year to the next, so averaging the wind speeds for many months will tend to smear out the position of the jet stream and the winds will be weaker at any particular spot (because sometimes the jet stream will be there and sometimes not). To account for this "smearing out" of the averaged jet stream position and better highlight the average location of the jet stream, you'll want to construct a plot of wind speed that doesn't show winds slower than about 25 meters/second (rather than the conventional cut-off of 30 m/s). To this end, and to help optimize the jet stream plot more generally, change the default wind speed contour interval and the range of values to plot:

  8. Rather than viewing a plot for the entire world, create one for North America (which focuses more closely on the area of interest to us, the West Coast of the U.S.):

  9. Click on the "Create plot" button. This should create the specified plot and display it in your Web browser.

  10. Capture the plot for use in your summary report for the Final Project:

  11. Now you're ready to create another plot, this time a composite of average rainy-season wind speeds during years in which strong El Niño events occurred.

    Click on your browser's "Back" button to get back to the "Monthly/Seasonal Climate Composites" page. Repeat Steps 1 through 10 except for Step 5. Erase the existing entry (if any) for the "Enter range of years" item. This time, instead of a requesting a full range of years from 1951 to the more recent "rainfall season", refer to the option immediately above that, called "Enter years for composites (from 1 to 16)". In the text boxes beneath it, enter the years in which strong El Niño events occurred, which you determined in Part II of the Final Project. In Step 10, be sure to give your new plot an appropriate title.

  12. Repeat for years in which strong La Niña events occurred. By the time you're done, you should have three plots altogether.

  13. If you're not working on the computer on which you plan to write your summary of results for the Final Project, then email yourself a copy of the file containing your three captioned plots, or save a copy on a thumb drive, or ask the instructors for advice about how to save the document in a place and form where you can access it later.

You're now in a position to see if the hypothesis is confirmed, that ENSO events affect the position of the jet stream in a way that can help account for any statistical connections that you saw in Part II of the Final Project.

At this point, you should be ready to write your summary report for the Final Project (see "The Summary Report" for guidance).