These are measures APCD staff take to ensure the quality of data gathered using manual samplers of PM2.5. (Speciated PM2.5 sampling is done under a different set of procedures.) A manual PM2.5 monitoring system consists of a sampling system at the site, an analysis system at a central location and a transportation system to get the sample filters from one to the other.
The sampling system consists of a sampler (one or more at each site) with attached temperature and pressure instruments and a human operator who loads new (unexposed) filters into the sampler, programs it to sample each filter for 24 hours (normally midnight to midnight) on a specified day, and retrieves the exposed filters afterward. The samplers APCD uses are "sequential" samplers, capable of automatically changing filters to sample for several days in a row. At two sites, the main samplers run continuously, except for maintenance and QA checks.
The gravimetric analysis system consists of a microbalance in a climate-controlled weighing room and a human analyst trained to operate it. The weighing must take place within a narrow range of temperature and humidity. Before being weighed, both unexposed and exposed filters must be conditioned for at least 24 hours in this environment.
The exposed sample filters are transported in a cooler to limit the loss of volatile material, and are then kept in a refrigerator if they cannot be weighed within 10 days of sampling. They still must be weighed within 30 days of sampling to be valid. Unexposed filters are taken out to the site under similar controlled conditions, usually in a cooler, although this is not necessary).
- Data requirements:
- 75% data recovery (or at least 11 readings per quarter if the average will not meet the standard).
- Data quality objective: 10% coefficient of variation. The coefficient of variation, CV, is the standard deviation as a fraction of the mean:
- Measures of Accuracy and Bias:
- Check of flow rate (quarterly)
- Collocation of a federal reference method (FRM) sampler with the main sampler (see "Collocation" below)
- Audit by collocation with a portable FRM sampler: Each year, an EPA contractor audits 25% of SLAMS (for Jefferson County, that currently means one site). Each such site is audited 4 times during the year. Since these samples are weighed at a separate laboratory, this measures the accuracy of the whole sampling and analysis system.
- Measures of Precision:
- Check of flow rate (every 2 weeks)
- Collocation (see below).
- 25% of all SLAMS monitors have a second monitor collocated at the same site.
- The second ("duplicate") sampler runs every 6 days.
- Collocated particulate samplers must be less than 4 m apart but more than 1 m apart.
- If the main monitor uses the FRM, it must be collocated with an FRM monitor.
- 50% of federal equivalent method (FEM) monitors must be collocated with an FRM monitor, 50% with a monitor using the same method.
- After an area is found to be in violation of the standards, 80% of collocated monitors will be at sites with concentrations greater than 90% of the NAAQS.