Gaussian Plumes - how far out can you be?

["Tucker, David" <tuckerd@aecl.ca>]

Radsafers:

Some coworkers and I are doing some analysis of accident consequences at
our site.  One aspect we were looking at is the width of area
contaminated at or above a given level after the plume has passed over.

Has anyone done some analysis (or found a good reference) that indicates
how far out from centerline the Gaussian plume model is valid?  I
realize it is only an approximation, and it seems to me that I remember
one of the problems being "edge effects" - the Gaussian plume never goes
to 0 but the real plume does.  Is it OK to be several sigma out  - or is
there some limit.

To illustrate the problem - we wanted to identify the area contaminated
above 1E4 Bq/m^2.  At 500m from the stack in Class B weather, the area
was 800m wide.  This seems wrong.

If anyone has useful info or insight - please advise.
Thanks
Dave Tucker

_____________

Dave Tucker, CHP
Health Physicist
Radiation Protection Branch
AECL - Chalk River Labs
Chalk River, Ontario
Canada
KOJ 1JO
Ph (613) 584-8811 X4505
tuckerd@aecl.ca
Tucker