28th August 2013, 09:48 PM
My research background is in site predictive modelling so I have a little experience with predicting sites, negative evals, etc.
Across dozens of countries, countless environments, countless cultures, all manner of sites and time periods you can predict with pretty high accuracy and precision where you will find sites. In the last area I looked at 93% of the land was devoid of sites. If you were to take a random piece of land (no such thing as random in landscapes but imagine for a moment) there was a 93% chance you would find nothing. I can even predict where in that 7% that you find sites you will find a certain type of site e.g. camp, cave, quarry, etc.
Great you say, we don't need to do archaeology for everything. Well 30, closer to 40, years of work on predictive modelling and pretty much everyone has come to the same conclusion, including myself, always always always always conduct an evaluation. Even when you can say with a certain high level of certainty that no site will be at location A you will find something occasionally. Moreover, not finding something is just as valuable as finding something. For site predictive modelling to work you need to have lots of areas with "known" nothing. The best you can do is advise developers the potential costs e.g. if you build there expect to pay X.
Across dozens of countries, countless environments, countless cultures, all manner of sites and time periods you can predict with pretty high accuracy and precision where you will find sites. In the last area I looked at 93% of the land was devoid of sites. If you were to take a random piece of land (no such thing as random in landscapes but imagine for a moment) there was a 93% chance you would find nothing. I can even predict where in that 7% that you find sites you will find a certain type of site e.g. camp, cave, quarry, etc.
Great you say, we don't need to do archaeology for everything. Well 30, closer to 40, years of work on predictive modelling and pretty much everyone has come to the same conclusion, including myself, always always always always conduct an evaluation. Even when you can say with a certain high level of certainty that no site will be at location A you will find something occasionally. Moreover, not finding something is just as valuable as finding something. For site predictive modelling to work you need to have lots of areas with "known" nothing. The best you can do is advise developers the potential costs e.g. if you build there expect to pay X.