It is not strictly annual in the chronological sense, but during the preponderance of years this event took place.
there was a recent paper in quaternary research led by julio betancourt and tom swetnam comparing drought reconstructions in new mexico derived by speleothems and tree-rings.
there was no detectable relation between the tree-ring and speleothem reconstructions.
there are many possibilities on why there was no relation between the two types of records.
if anyone is interested in this report, i can supply the reference when i get back to the office next week.
neil Don, You ask for my take on a paper entitled, A Test of the Annual Resolution in Stalagmites Using Tree Rings.
I suppose the easiest thing to do would be to say interesting and ignore it, or to trash the paper completely, but I will try to do a fair evaluation.
One of the main considerations is a difference in mind set between two different groups of scientists.
One group are geologists the other tree-ring researches.
In the paper cited (Polyak and Asmeron, 2001) they found more than 1600 bands in a stalagmite radioactively dated as between 2796 +/- 88 and 835 +/- 25 years in age dated via u series dating.
On the face of it, if you would assume the dates are correct and temporarily ignore the variance, then you have a period of 1961 years +/- 113, with only 1600 bands.
Of course the rings in the stalagmite do not correspond 1 to 1 with the calendar years and hence will not correlate 1 to 1 with tree ring dates.