|
The method (derived from one in USEPA
SW-846: http://www.epa.gov/epaoswer/hazwaste/test/pdfs/chap9.pdf
- good discussion on page 9-6) is designed to estimate the number of samples
from a homogeneous, normally-distributed population needed to determine if the
mean value of the population is less than some regulatory limit, with a stated
level of certainty (80% in this case). It’s not a bad tool for that task,
but make sure what it’s answering is what you’re really
asking. Are you interested in the mean value of the pile or the likely
maximum value? Is it reasonable to expect the pile’s concentrations
to be normally distributed? Is the pile homogeneous and amenable to being
sampled in this way? Of course, you may be forced to use this
for at least two reasons:1) regulatory requirement or 2) nothing better. To me, it’s much safer to use the
rule of thumb that you take at least 10 – 20 samples in every “block”
you consider homogeneous. This generally gives you enough statistical
power to estimate most anything needed for investigation. When in doubt –
consult a good statistician who is familiar with environmental sampling (not
all are!). Tim Glover From: Robert Pope
[mailto:[EMAIL PROTECTED] Hi all, I am writing with a specific question
about the In Section 5, Page 5-75, Table 5-13, they
offer a Stockpile Statistics Worksheet that is designed to estimate the minimum
number of samples required to sufficiently evaluate a stockpile of waste soils. http://www.sdcounty.ca.gov/deh/lwq/sam/pdf_files/manual_2004/sections/pdf/section_5.pdf My questions: 1) Does anyone have opinions about the method they utilize?
(i.e., is it good/bad and why?) 2) Does anyone have a better method? Thank you, Robert Pope |
* By using the ai-geostats mailing list you agree to follow its rules ( see http://www.ai-geostats.org/help_ai-geostats.htm )
* To unsubscribe to ai-geostats, send the following in the subject or in the body (plain text format) of an email message to [EMAIL PROTECTED] Signoff ai-geostats
