GEOL 1100: Global Physical/Descriptive Oceanography

Paper #1: 3 Kinds of Lies

Describe quantitatively the amount and quality of ocean data available to describe a region, watermass, variability, or process of interest. Some examples:

  1. Subsample a dataset to estimate the sampling error in a bulk measurement, e.g., global mean temperature. How many subsurface observations are needed? Does it matter where they are? Show QUANTITIATIVELY.
  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), or in a repeated section, etc. How do you distinguish variability from error? Can you distinguish cycles from trends?
  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)? Can you distinguish instrumental noise from "aliasing noise", that is, effects from processes you don't want to deal with?
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.
Feel free to use the Matlab references, and refer to similar projects from my past classes:
  1. "3 Kinds of Lies" from 2010's class. username: atoc5051_10 password: Oshuns?Noshuns!
  2. "3 Kinds of Lies" from 2008's class. username: atoc5051_08 password: 2OshunNoshun?
  3. "Ready to Roll" and "What Do You Mean Mean?" from 2007's class. username: atoc5051_07 password: 4OshunNoshun! (Note: these were not revised papers, so they are first drafts.)
The stages of the full Paper 1 project will be:
  1. Plans: (little time, concurrent with revisions) A short pdf you'll send to me, with tentative title, region, dataset, preliminary hypotheses, preliminary figures, and a list of important references you have identified. The main goal of this exercise is so that I can get you on the right track early, so you'll have good resources/code to work with.
  2. Initial Submission (3 weeks) This is *not a draft*, it is a finalized as far as you can do on your own paper. I give the most points to this stage of the process to emphasize the importance. In the real world, it's this version that gets accepted or rejected by a journal.
  3. Shuffle and Review (1 week) You will read and review (anonymously) 2 other students in the class. As soon as I get all the reviews, I'll send them out to authors.
  4. Revise (1 week) Here you will decide how & what to respond to from the reviewers. If you have time, you should fix all the easy to fix mistakes. Anything too hard you should refuse to do. You will learn to use good judgement!
To be most successful, you should:
  1. Pick a region that interests you
  2. Match the quantity you attempt to diagnose to a key process in that region
  3. Develop hypotheses as you make calculations
  4. Challenge your results--Do the units, magnitudes, and behaviors make sense?
  5. Do background reading to help!!! Months in the lab will save you from having to spend hours reading.
  6. Use the textbook to help identify regions/processes that interest you.
  7. This time around, you will not be required to use any dynamical conceptual knowledge, but you must challenge the ability of the data to tell you something meaningful.