How was this possible?
http://climateapps2.oerc.ox.ac.uk/cpdnboinc/result.php?resultid=16130365

Guessing a bit here but do you have your machine set up so it only communicates with CPDN at certain times?

I have a vague notion that what happens with the trickle ups is that in reality they are all amalgamated in to one trickle if they are not allowed to go when created. That would explain the difference compared with the zip files which on my machines at least, BOINC will only allow two to be uploading at a time.

It isn�t my own computer, I just saw it while browsing the top participants.
http://climateapps2.oerc.ox.ac.uk/cpdnboinc/show_user.php?userid=558670

How was this possible?
http://climateapps2.oerc.ox.ac.uk/cpdnboinc/result.php?resultid=16130365

The column header on the trickle column is a little misleading - it's actually the time when the server processed the received trickles. It doesn't tell us anything about when the user's computer prepared the trickle reports. They could indeed have been created while the computer was offline, as Dave suggests, or there could have been server problems delaying processing of the received reports.

Trickles (which are processed via a scheduler RPC) are technically unrelated to data file uploads, so the 'two at a time' restriction doesn't apply here.

The thing which disappoints me is that the task was allocated in December last year, and has only just been reported - it's been sitting around somewhere for over 11 months, even though it needed less than four days of actual work. CPDN may have relaxed deadlines, but I'm sure the researchers would prefer to have their results quicker than that.