Survival parameter estimates & model averages - known fa

[Tom]

I have a small number of models describing survival for radio-marked animals of four groups (Adult males, adult females, subadult males, subadult females) and the models with the lowest AIC values do not have a group effect, but rather have an effect of time of year (i.e., trapping season, not trapping season). The "derived parameter' estimates only give me "the probability of surviving the duration of the study" for the 4 groups, but not by trapping, non-trapping. How can I get these estimates?

What are the differences between these "derived paramters" and the "real parameters" from the results browser menu?

I thought that if you model averaged across models that you would have estimates of survival for each group for each survival period; however, these estimates are suspectly high, and do not correspond at all with the derived estimates of survival.