Projects for Students

This is a list of areas for students who may be seeking research and development experience in the form of projects. These projects might take the form of independent studies, masters projects/theses, or even PhD work in collaboration with a PhD-granting department.

These projects generally revolve around the numerous data processing issues related to real-time weather modeling. For an overview of the general work I'm doing, you can refer to some of the following links

• Alaska Weather Modeling
• Some Weather Model Output
• Some of My Publications
• WRF

Some possible areas to pursue:

• Robust acquisition of remote data and postprocessing of model output - currently, I have scripts that automatically go out and ftp input data for the model, and then post it to web sites in various forms. It all works great when network connectivity exists, but periodic network glitches can cause this to fall apart, and we end up losing a full day's forecast. So, it's necessary to develop a scheme that gracefully recovers from a bad network connection, allowing the data processing to continue once the network is back up.
• Investigation of the use of Google Earth as a platform for the 3d, time-dependent visualization of weather model output.
• Investigation of the use of Integrated Data Viewer (IDV) as a platform for the 3d, time-dependent visualization of weather model output.
• Extraction of selected data from the output files. The output files are in NetCDF format. I'm currently using some NCL scripts for extracting forecast variables at specific locations, but this is a very slow process. I would like to see more efficient means investigated.
• Build scripts that ease the task of extracting key data from the weather model output and compare with observations available over the Internet. This is currently a very tedious task and error-prone - an integrated system that would do this for us would be very useful.
• Rigorous verification and validation of model output - if you're more of a geophysicist type than computer scientist, there are plenty of opportunities for evaluating the output of our model runs.

Independent Study Guidelines

Students interested in any of the above are encouraged to pursue an independent study under my supervision. The following guidelines apply to all independent studies supervised by me:

• Independent Studies are just what the name implies - work done independently by the student in order to gain experience and insight in an area outside of our normal course sequences. The projects listed above are typically areas that I have a little bit of knowledge in, but for the most part don't have the time to pursue in any depth on my own.
• Independent Studies are not meant for me to teach you material; rather, it is a wonderful opportunity for you to practice learning an area on your own. This is a valuable skill.
• If you choose to do one of these Independent Studies, you will write up an outline of what you hope to accomplish and a timeline for project milestones. I will help you refine this to insure your goals and expectations are realistic.
• In some cases, you will be pursuing work where the outcome is uncertain, and you may reach an unexpected dead end forcing you to redefine goals and directions during the semester. That's OK, and it's important to understand that reporting on failures is just as valuable as reporting on successes. You will not be penalized just because you pursued a direction that you thought would work, only to find that it didn't. The expectation is that you will try your best to recover from the setback and try to pursue alternate courses of action.
• Your grade from the study will come completely from a project paper and public presentation that you make. The public presentation will be a 30-60 minute seminar announced in the department so that your peers and any interested faculty can attend.
• During the semester, it is expected that we will meet periodically to touch base on where you're at. If you're running into problem areas I will try to offer helpful advice, directions to pursue, and motivation, but again, this is an Independent Study, and it's up to you to do your best in an independent fashion.
• No incompletes except under very extenuating circumstances. You should pace yourself during the semester and work on this regularly to insure that you're not stuck at the end of the semester with little to show for it. A primary purpose of our periodic meetings is to encourage you to make regular progress.