What about this: Get some cheapo network enabled temperature/humidity sensors (i like to use these for server room monitoring) and put them up on various places in a building. Then go ahead and implement some data logging, statistics and maybe real-time warning.

This is not that hard to implement, you get real-world data, plus you may save your university some real money by helping them optimize the use of the climate control (or show them the need to install one).

This could be a multi-year project for many students, extending the system. Maybe implement a web page showing current and historical data and predictions based on that. Play "connect the dots" from your access logs and try to find out what the people are looking for.

While this system might not have a real "wow!" effect at first, it can really be helpful. They may ask themself many question that such a system can answer: Is it cooler in the cafeteria than outside (so, should i go early and get one of the limited places inside)? How warm is the water in the swimming pool? Is the heating in the chemistry lab still broken (do i need to long johns again)?

Of course, beeing students, there are of course the uhm, not so legalized questions that the system will also help to answer, like "Is the humidifier at our dorms pot vegetable plantation working correctly and is climate control putting out the optimum temperature?". Not that i would endorse such activity, but i know how groups of young, bored people tend to think. ;-)