Over the last twelve years we have found the best predictor around here being rainfall the previous October and December Doesn't make much sense I know but in a natural system a 50% correlation (which this model has) ain't bad.
With martins, mortality for the whole population running about 60% a year, to merely maintain the population each nesting pair has to fledge three young (two parent martins plus three fledglings makes five martins, three die, two survive to return the following year).
Every year has been different: In 2010 we had our best year ever, where few young died, and almost all the young were plump and healthy. 2011 was an unmigitated disaster. It seemed that a switch had been thrown during what was already a hard season such that most young died in the nest, including what had been seemingly healthy twenty-plus day-old nestlings, who's parents were suddenly unable to find food for them.
2012 turned out better than expected, actually more than four fledges on average per nesting attempt. This year we had hopes at least for a break-even year, with three fledges per nest.
Looks like its going to be less than that. Today's nest check revealed just 123 young among thirty-eight nests. In the impossible event (some are runts) that they all survived, we would average 3.2 fledges/nest for just these productive nests.
Realistically, by the time we average in two failed clutches and two missing broods, plus allowing for inevitable nestling losses, I would guesstimate about a 2.0 - 2.5 fledge/nest rate for our colony this year
Mike Scully