17th August 2013, 10:49 AM
Seedy Girl Wrote:....what if the site formation processes were not clearly understood on site, what if the feature is not/ can't be dated? That is then a waste of time and money and the data wasted. The data is only relevant if the feature can be dated, if not, it's just a pile of stuff!...
...and of course being able to model residuality across your site through time is fairly important too, a pit that's been cut through 300 years of Roman statigraphy and then backfilled with the randomised same isn't going to be a good sample target. For a big RB site we did several years ago a significant proportion of the enviro assemblages ended up in the 'junk' column after the pot people had finished modelling the really quite epic residuality issue across the site - if 90% of the pot's residual it's a fair bet the same can be applied to the bone, seeds etc, the challenge is teasing out the secure stuff