Clive
Seems like a good argument for using an XRF to take several measurements to give you mean and standard deviation.
This is all a bit new to me but I presume dealing with a site where most of the soil is slightly contaminated would be different from a site where all the contamination is localised at one point. Would not mixing samples potentially provide similar results for these , very different, situations?
Regards
Peter Fleming
07958 205920
Skype: petermfleming
-----Original Message-----
From: Contaminated Land Management Discussion List <[log in to unmask]> On Behalf Of Williams, Clive R
Sent: Monday, 02 July 2018 11:18
To: [log in to unmask]
Subject: Re: Composite sampling
What procedures do the labs take now to reduce a 1kg tub sample of soil down to the teaspoon's worth that gets made into an ICP-MS pellet? In my days of being a mine geologist I would take channel samples across the exposed face of the ore body to gain an idea of what the overall quality of the ore was likely to be, being mindful that softer bits were more likely to yield material and also to include the harder bits which were lower grade.
We also coned and quartered large bulk samples to reduce them down but I'm mindful of the impracticalities of doing this with grossly contaminated materials with or without asbestos outside a fume cupboard.
Clive Williams
BSc MSc CGeol SiLC SQP
Associate
Contaminated Land Specialist
T +44 (0)29 2046 7867
[log in to unmask]
-----Original Message-----
From: Contaminated Land Management Discussion List [mailto:[log in to unmask]] On Behalf Of Mike Smith
Sent: 02 July 2018 11:07
To: [log in to unmask]
Subject: Re: Composite sampling
Hi Colin,
I wouldn't disagree with you in any way concerning the inadequate sampling densities used during many investigations for contamination. Unfortunately despite the available UK guidance (e.g. CLR documents) there is often little grasp of sampling statistics. BS 18400-104 will cite some of this guidance in an effort to get users to think first and dig holes later.
I also agree with your comments about the use of XRF and similar onsite methods.
Composite sampling really is only applicable for determination of average concentrations except under very special circumstances.
I wasn't perhaps as clear as I might have been.
"cluster sampling" does not refer to the mixing of samples in the laboratory but to forming the field sample from multiple increments taken over a small area (taking increments at a fixed depth along the side of trial pit is an equivalent process). It smooths out some of the very localised variations in composition. Mike Ramsey thinks it is OK.
In contrast, the five or so samples that are to be combined in the laboratory would be taken from evenly spaced locations from one of say four zones within an allotment plot. The results for the four zones can be compared. In practice the number of samples to be combined depends on the variability at the scale of the sampling - this has to be determined initially for sample zones or assumed.
Regards,
Mike
-----Original Message-----
From: Contaminated Land Management Discussion List [mailto:[log in to unmask]] On Behalf Of Colin Green
Sent: 02 July 2018 10:29
To: [log in to unmask]
Subject: Composite sampling
Hi Mike
Composite sampling can be useful as previously described to get an "indication" about potential contamination issues. The real issue this brings up however is the generally low sampling (and therefore analysis) densities seen when assessing potentially contaminated sites or for classifying stockpiles. Several papers over the years have highlighted this problem, and recently Professor Mike Ramsey from Sussex University has shown on several occasions how the confidence in data quality suffers when too few samples are taken. Even with a grid of 20m x 20m and taking a sample at every 0.5m depth, this equates to a volume of 200 m3, which equates to around 26 standard muck away lorry loads. On most brownfield sites sample homogeneity is poor, so the possibility of a lorry load containing significantly more contamination than the single analytical result obtained for these 26 lorry loads is high. In my experience a 20m x 20m equivalent grid is somewhat rare with 35 - 50 m grid equivalents quite common. A 30m x 30m x 0.5m grid equivalent would require 58 standard muck away lorries to remove.
The new rules that came in force in April this year now significantly increase the costs and consequences of inadvertently (or intentionally) mis classifying soil for disposal as non hazardous when it is actually hazardous. Similarly, disposing of soil as hazardous when overall the soil is non hazardous is also an unnecessary expense at around £2,000 per lorry load.
Appropriate sampling and analysis is therefore very relevant to the overall cost of a remediation. I hope the new sampling standards will help address this issue. The latest generation of in situ analysis methods such as XRF for heavy metals or UV fluorescence for hydrocarbons and PAHs can also be used to provide the increased sampling densities and provide results that are equivalent to laboratory methods, but at a significantly lower cost. Deana Crumbling et al from the US EPA in 2003 (Crumbling, Griffith, Powell : REMEDIATION Spring 2003) showed how higher sampling densities using just low resolution semi quantitative in situ methods produced higher overall data confidence in the CSM.
This approach has been successfully used to assess allotments in the UK.
I am intrigued by the cluster sampling example. I would have thought that if 5 samples were combined, the concentrations of the combined result should be multiplied by x5 to give the maximum theoretical concentration in a single spot. This value can then be applied to the agreed site limits.
I would be interested to see the draft sampling guidelines. I come from a time when the old DD0175 was in use (or not actually used as was the usual case!!)
Regards
Colin Green
########################################################################
To unsubscribe from the CONTAMINATED-LAND-STRATEGIES list, click the following link:
https://emea01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.jiscmail.ac.uk%2Fcgi-bin%2Fwebadmin%3FSUBED1%3DCONTAMINATED-LAND-STRATEGIES%26A%3D1&data=01%7C01%7Cclive.williams%40MOTTMAC.COM%7C5944f2dd47b14fe5845c08d5e003b13c%7Ca2bed0c459574f73b0c2a811407590fb%7C0&sdata=ox3YxzLGzUYkZXMl2Jwq%2F3%2BTJCfP2B7QyeZDZT6aTe0%3D&reserved=0
########################################################################
To unsubscribe from the CONTAMINATED-LAND-STRATEGIES list, click the following link:
https://emea01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.jiscmail.ac.uk%2Fcgi-bin%2Fwebadmin%3FSUBED1%3DCONTAMINATED-LAND-STRATEGIES%26A%3D1&data=01%7C01%7Cclive.williams%40MOTTMAC.COM%7C5944f2dd47b14fe5845c08d5e003b13c%7Ca2bed0c459574f73b0c2a811407590fb%7C0&sdata=ox3YxzLGzUYkZXMl2Jwq%2F3%2BTJCfP2B7QyeZDZT6aTe0%3D&reserved=0
########################################################################
To unsubscribe from the CONTAMINATED-LAND-STRATEGIES list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CONTAMINATED-LAND-STRATEGIES&A=1
########################################################################
To unsubscribe from the CONTAMINATED-LAND-STRATEGIES list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CONTAMINATED-LAND-STRATEGIES&A=1
|
