I have been running climate prediction HadSM3 for approximately 122 hours according to the "elapsed time" on the task tab. It also shows the "to completion" time as approximately 127 hours.

Is there an explanation as to why on both measurements the time continuously increases?

---

There are two possibilities:

1) The exact time to complete depends on a lot of things that are dependent on the computer and it's usage. The project has made a 'best guess' of the time to complete, which is sent to the computer, but this is based on an 'ideal' situation, and now BOINC is working out the exact amount of time it will take.
Eventually the time will start to decrease.
This is more noticeable on the currently available model, but less so on the shortest model type, which will soon be re-released.

2) The model has turned into an "ice world".
In this case, the time-to-completion will continue to increase.
This is discussed in several places on the Number crunching forum on the cpdn user forum, which explains how to tell if this is the case, and how to deal with it.

Thank you for your thorough explanation. I check it out, as you described, and yes it was in an ice world state.

Thanks
Dennis

---