ATOC 5051

Introduction to Physical Oceanography, 2007

Assignment 2: What Do You Mean Mean?

Describe quantitatively the amount and quality of ocean data available. Some examples:

  1. Subsample a dataset to estimate the sampling error in a bulk measurement, e.g., global mean temperature.
  2. Compare similar measurements from two datasets (e.g., Reid & Mantyla versus eWOCE, or WOCE shipboard CTD versus WOCE drifter CTD). You can do this in a region of interest, or in a chosen bulk measurement (e.g., global temperature, mediterranean temperature).
  3. Compare similar measurements but separated in time. Is there signal above the noise? How can you estimate noise (or has someone else done it for you)?

Don't be afraid to piggyback off of other published work. Use google scholar or Web of Science to locate articles by keyword, and then see what they say.
To do a box average (which comes with standard deviations, too) in ODV, refer to the help manual section 11.9. You basically need to make a text file with six entries per line (which delinineate the central latitude, longitude, and depth, and then the range of latitude, longitude, and depth). I have attached an example. You can edit it with notepad (on PC), or textedit (on Mac) or emacs/vi/stdout (on linux). To get output, just do a selection criterion on the region you want, then Utilities>>Box Averaging. The output you get is another text file, similarly named, which gives:
lon lat dep (same as in box definition file) lon lat dep val s nu nr
The second lon lat dep are the averages of the values found, val is the average value, and s is the standard deviation of the data points. nu is number of data points used and nr is number rejected. You get one line of output back for each line of input. 3d Estimation works in very much the same way (see ODV manual).