Welcome to CornerPost

Welcome to Cornerstone's blog, which is devoted to providing useful information to our clients and industry professionals.

Our talented staff of bloggers will offer brief, thoughtful pieces on the topics facing our industry today.

CornerPost will feature tips and tools, lessons learned, observations about major conferences, and much more.

We welcome comments and suggestions for topics you would like to read about.

Visit our website:
Cornerstone Environmental Group

Stay Connected

LinkedIn
4 May 2017

How to avoid prematurely committing a landfill to NSPS compliance – carefully plan sampling to avoid positive bias

Landfill operators must conduct Tier 2 testing to calculate a landfill’s non-methane organic compound (NMOC) emission rate if the design capacity is greater than 2.5 million megagrams (MG). Favorable results of a Tier 2 test allows landfills to opt out of the New Source Performance Standards (NSPS) requirements to install and operate a gas collection and control system (GCCS).

Unfortunately, the method frequently used to gather the data has a tendency to over-estimate NMOC concentration levels. That can prematurely commit a site to constructing or expanding a GCCS to maintain NSPS compliance and all its additional operating, maintenance, monitoring and reporting costs. What’s more, EPA recently reduced the NSPS threshold from 50 MG/year to 34 MG/year, so a greater number of landfills are close to the threshold. That makes it doubly important to prevent over-estimating emissions. Here’s a few examples of where this happened and what I think can be done to prevent it.

Sampling methods matter

There are two acceptable methods for getting the samples needed. You can use landfill gas header sampling points to obtain composite gas samples from the entire area covered by the existing GCCS, or you can take discrete samples from individual probes.

Composite sampling saves time and money because you take fewer samples. It’s used frequently where an existing composite cap or interim cover systems is in place to avoid damaging the cover with probe points. But it over-estimates NMOC concentrations levels compared to multiple local samples from the same disposal area. Why? Because composite samples from landfill gas collection system header pipe is more heavily weighted to areas of higher gas production rather than an “average” landfill condition.

Positive bias in header samples – and what to do about it

Most wellfields have higher rates of gas flow from newer, higher gas producing areas of the landfill so they can either control odors, or collect more gas for landfill gas to energy use. This results in a higher weighting toward these areas in the header pipe and the resulting gas samples from the landfill gas header. Since these areas typically have higher NMOC results as well, it can lead to a positive bias in the NMOC sampling results for a Tier 2 event.

Geoprobe sampling

Here’s an example, taken from a landfill in eastern Ohio that had a GCCS system on about 82 percent of the site. We took three composite header samples to represent the area; the NMOC as carbon results were 1900, 2400 and 1800 parts per million by volume (ppmv), or an average of 2033 ppmv. We sampled the remaining 18 percent of the required sampling area with nine pilot probes. We got results of 820, 560, and 1000 ppmv, for an average of 793 ppmv – or only about 39 percent of what we observed in the header samples! The findings projected that the landfill would be over the 50 MG/yr limit because the header results were weighted by the 82 percent of the total.

The landfill operators decided to retest a year later without sampling from the header. Instead, they selected 50 sampling points either from a pilot probe or from an individual gas well. The samples revealed an average of 594.4 NMOC as carbon – about a third of what was observed in the previous year. This allowed the landfill to hold off being subject to the NSPS requirements.

Correcting the bias

There are two ways to correct for this bias. You can sample individual sampling points as our Ohio example did, OR you can try to even out, or “normalize,” the flow from each wellfield collector. Since it is often cheaper to sample from the header compared to taking this high number of individual samples, it can make sense to try to normalize the flow from each individual collector and take samples from the header.

You can only do this if the wellfield is tuned so each collector provides the same amount of flow during the sampling. To do this you have to tune wells from higher productive areas down to match the flows of wells in the lower producing areas. You also have to tune down wells that are flowing with oxygen greater than five percent, or have temperatures greater than 131 degrees, and shut down wells located in areas of known non-degradable waste.

This approach was used by a landfill in Central Ohio, which took three sets of NMOC samples from the common header pipe. The existing gas collection system covered the entire landfill. Test 1 analyzed one can, which showed 2200 ppmv NMOC as carbon with no adjustments to the wellfield. The system was operating to control odors and most wells were being pulled as hard as possible without regard gas quality.

Test 2 analyzed 2 cans; the test showed 1200 and 1500 ppmv NMOC as carbon, after trying to normalize the wellfield flows the day before sampling. Test 3 (two weeks after Test 2), got results of 1100, 1300 and 1500 ppmv NMOC as carbon, after repeating the procedure to normalize wellfield flows.

If a landfill thinks they will be close to the threshold, it may make sense to attempt to normalize the wellfield flows and perform a preliminary Tier 2 sample from the header to see where the NMOC concentration ends up. Depending on results, they can make an informed decision on which method may yield the best results.

If preliminary header sampling shows favorable results, they would be able to perform the Tier 2 test at a lower cost, and prevent bias from higher producing wells. If a landfill wants to prevent this bias altogether, they should sample at individual sampling points or try normalizing the wellfield.

Try, try again

All landfills are unique and there is no one size fits all approach to Tier 2 testing. But remember, there is nothing in the regulations that prevents trying a different method if you don’t get the desired results on the first try. The costs of additional testing may be worth the investment because the costs associated with the monitoring, recordkeeping, and reporting requirements to comply with the NSPS subpart WWW or XXX are substantially higher.

Have you tried either of these approaches to eliminating bias in Tier 2 testing?

Brad Albrinck, PE, is a Project Engineer with seven years of solid waste industry experience. He focuses on landfill gas system design, gas flow analysis, operation and maintenance, data analysis, compliance reporting, and Tier 2 NMOC testing.

Categories: Air Quality, Biogas and Landfill Gas, Environmental Planning & Compliance, Landfill Engineering and Design, Solid Waste
Posted By Brad Albrinck, PE at 2:20 PM  |  No Comments on How to avoid prematurely committing a landfill to NSPS compliance – carefully plan sampling to avoid positive bias

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>