Repeated measures - appropriate for my study or not?

Howdy folks - first time poster here. I'm a grad student at CU-Boulder working on grassland ecology and need some assistance.

I've got six transect plots- two control, four treatment - that have been sampled twice a year (in spring and fall) for the past 11 years. The sampled variable has been plant cover, expressed as a plant's proportion of absolute ground cover and plant cover. I am interested in seeing if linear directional trends exist in the data when regressed against time, a heating degree day index and yearly precipitation.

Up to now, I've been performing correlations (using SAS' "PROC CORR") and multiple linear regressions (PROC REG) to evaluate my data. My adviser and I have discussed the use of a repeated measures approach, thinking that the repeated sampling of each transect at biannual intervals qualifies as a repeated measure. Am I likely to see anything different with a repeated measures approach? Would there be anything to specifically warrant the use of repeated measures, or is a regression fine? Another concern of mine is that regression may be inappropriate due to time and temperature's tendency to autocorrelate.
I am kind of suspicious that you have an error structure that is valid for both, but:

Why you shouldn't want to analyze a repeated measures as an ANCOVA (or regression):

ANCOVA (regression) entertains the notion that the continous covariate isn't always the same level in every subject. So what we are talking about in your example is measuring the response at different times for each of your plots. Repeated measures leverages that fact that a priori you had set times you were going to take measurements across all your subjects into more power. If you discard that information and do the same analysis that you would have done if your measurements were not taken at all the same times then you are loosing something.