In this project, we inspect fronts crossing the country associated with day-to-day variations in the weather using real-time atmospheric observations. In the laboratory we create fronts by allowing salty (and hence dense) columns of water to collapse under rotation and gravity. We discover that the observed changes in winds and temperature across our laboratory and atmospheric fronts is consistent with Margule's formula (a discrete form of the thermal wind equation) and see that the dynamical balance at work in the atmosphere is the same as in the density fronts created in the rotating tank.
Project Description (PDF)
Notes on relevant theory: thermal wind (PDF)
Tank Experiments
You can read about the experiment in the Weather in a Tank project: Fronts: An Introduction
Atmospheric Data
Relevant Notes at the Weather in a Tank Web site: The Polar Front and Synoptic-scale Fronts
Polar Front
Mean fields: Use climatological data to verify thermal wind balance across the polar front.
- Plot your data (3 files total). See Instructions.
- Use the MATLAB script (M) to compute the temperature gradient and vertical wind shear.
Mid-latitude cyclones and the polar front
Instantaneous fields
1a) Plot the 500 mb temperature over the Northern Hemisphere, using GDCNTR (area: nhem)
1b) Use Any-section to do a cross section from pole to equator.
2) Using the same date, plot the Southern Hemisphere 500 mb temperature: using GDCNTR (area: shem)
3) Compare your plots to the appropriate satellite image
Satellite Images (Polar view)
Warm and cold fronts
3) Find the corresponding Surface Analysis
- Suggested dates: 11/22/07 (18z), 2/6/08 (18z), 3/2/08 (18z), 12/15/08 (00z), 3/11/09 (00z)
4) Plot the 850 mb temperature using GDCNTR
- Suggested changes — GDFILE: regional, GLEVEL: 850, GFUNC: tmpc, CINT: 2
- To zoom in, pick state in the center of interest – enter the state abbreviation in GAREA with a dash at the end, e.g. MA - (the dash is to zoom out)
5a) Produce sections through a warm and cold front, using Any-section
5b) Using gdcross might be helpful in determining the slope of the front, - gfunc=hght; cint=200 (height is in meters)
Other links: