AQS "Ask the Experts" Webinar - Nov 2018

AQS “Ask the Experts” Webinar Questions

November 1, 2018

Question 1: Is there a way to join the location of monitors (lat/lon) with the observations at those monitors on a single query? For instance, it would be much more convenient to query daily max ozone values for Regions 1-7 (for example) for one particular date and have the lat/lon locations included for all those observations in that query for easy mapping than having to query a 435 (or 501) in addition to the 500 (monitor location) reports and then join them using outside software.

Answer 1: No, AQS does not presently support this today. However, as we are re-engineering our reports capability, this can be added. The new API will have the ability to grab a geographic box of daily summary data.

Question 2: When is EPA going to start requiring the submission of zero and span check transactions and 5 min SO2 certification and the remaining Part 58 regulatory changes made in FY 2019?

Answer 2: There is no requirement for the submission of zero and span check transactions to AQS and there is no current thinking to require the submission of zero or span checks. Submission of these checks is voluntary. We are currently developing AQS to accept zero and span checks in calendar year 2019; thus, State, local and tribal agencies will have the opportunity to submit that data. Reporting requirements come from our monitoring regulations defined in Part 58 and App A to Part 58, which for QA/QC purposes only requires reporting of the 1-point QC check Annual PE and NPAP for the 4 gaseous pollutants (CO, NO₂, SO₂ and O₃).

Also, there are no plans to require 5-min SO2 data certification.

Question 3: When is Qlik going to come online?

Answer 3: Qlik is new technology for eventually replacing AQS reports. This new system allows for visualization of the report data. The main motivator for considering moving to Qlik is that the current tech we’re currently using is expiring in 2-4 years. Within about 4 years we should have the complete new system on-line to replace reports, but it might be earlier.

Question 4: I am having trouble loading some continuous PM and visibility data to the database. Here in southwest Colorado, our PM2.5, PM10, and visibility data were significantly influenced by forest fires over the summer, so I added qualifier codes to all these data. These seemed to cause a problem, resulting in warnings, and not loading into AQS. I have tried both ‘E’ and ‘IT’ as qualifier codes.

I then tried removing the qualifier codes all together, since our PM data is no longer regulatory. I am still having trouble loading and posting the PM2.5 and PM10 data from June through September. I am attaching an AQS report showing the errors, connected to the sample period. The AQS support I found online does not seem to be very responsive. Any ideas?

Answer 4: You cannot use the qualifier ‘E’ today; it has been deprecated for several years. ‘IT’ is valid and should not cause problems. The load report does not show use of either of these qualifiers. Please discuss the other errors with your regional contact.

Question 5: Who should network modification request forms be submitted to at EPA? Who can I ask, or how can I check to see if my predecessor submitted such a form, indicating the replacement of our continuous PM monitor this past May? Could this be connected to the issue I wrote about in my previous email, connected to submitting our PM data from June – September to the AQS?

I was able to submit our data using the old instrument code, for a Teledyne 602 BetaPlus continuous particulate sampler. Our current instrument is a TeledyneT640X continuous sampler. How can I correct this retroactively in AQS?

Answer 5: Ask your regional contact, you should be able to submit an update to data previously submitted.

Question 6: Are the annual summary statistics, for example arithmetic mean, in AMP450 and AMP450NC reports calculated using the same calculation rules for the pollutant standard selected at the time of reports generation? Arithmetic means from the two SO2 reports, below, seem to be different. Which statistics would you recommend in data analyses for meeting air quality regulations? Similar inconsistencies can be noted between the AMP480 and AMP260 reports particularly for the NO2 98^th percentiles. Not included here. AMP450 report

Quicklook All Parameters Report

Answer 6: The statistics for AMP450 and AMP480 are the regulatory statistics. The ones for the other report like AMP260 are historical and may not be relevant today. The formulas for these are in the AQS Data Dictionary posted on the AQS website and are in the 40 CFR Part 50 Appendix for the parameter.

Question 7: What is the procedure for updating/changing method code on data that has been submitted to AQS and is certified?

Answer 7: Use Maintain>Monitor form, or batch “MM” transactions to change the method for the monitor date range. It does not matter if the data is certified or not and changing the method will not “de-certify” data.

Question 8: How can the Pb 3-month rolling averages be generated from AQS?

Answer 8: Right now, this is not possible; however, we will add this to the list of requested enhancements.

Question 9: With the Discoverer Plus tool no longer available, is there a tool that data users can use to retrieve data from AQS that is similar to Discoverer Plus? For what it’s worth, Discoverer Plus was a great tool for generating data reports tailored to one’s needs.

Answer 9: There is no replacement for Discoverer that will let you build your own queries. We are working on the new API which will have about twenty tables of data and eventually will have much more in Qlik reporting. We ask you to let us build the queries, so they are correct and available for everyone. We still need to set up process that will allow you to request new tables and combinations of data.

Question 10: CO at our NCore site…our Precision Check standard is 350ppb. As of 3/28/16, CFR says CO PC range is 0.5-5 ppm (500-5k ppb), but, of course, there is no mention of NCore. Since we haven't been hit for being below the required range in 2016-17, I have to assume that NCore sites are treated differently. The only thing I've found is in the 2007 NCore workshop, where it says CO pc level 250-500.

Answer 10: We understand that most NCore trace level CO data is in a range of 150 – 300 ppb. Despite these low numbers and even with no non-attainment areas we still use the CO data to show attainment of the CO NAAQS. So, while the NCore CO monitor is meeting multiple objectives, from a QA standpoint its most important objective is still to support NAAQS. Therefore, it is intended that a precision check is run at low point of 500 ppb. We suggest that you stay consistent with the regulatory requirements and perform a check minimally at 500 ppb. AQS is still developing range criteria for 1-point QC checks so it should give you a warning in 2019 if the 1-point QC checks are not within the required range. You can always run a second QC check at a lower range. In addition, the Annual PE audit can be conducted at lower audit levels and this is a second way of gathering information on precision and bias at lower concentrations.

Question 11: We have 2 CO-TLE instruments (trace level) that we use for both our NCore site (trace level measurements) and Near Road site (ambient level measurements). We are using the same method code for both…should we be?

Answer 11: Yes, given that these are both CO-TLE methods, they should have the same method code. We work with the EPA’s Research and Development Office’s Methods and Equivalency Program. This program has over last 30-40 years approved a number of method designations, in many cases it’s the same method that gets updated over the years. When EPA and made a push to go to trace level monitoring for CO and SO2 we reached a crossroad. We made decision, to add 500 to the number of the original method designation for purposes of differentiating those methods when the data are stored in AQS so users would know that it is a different detection limit. The official method designation would still be the same whatever the initial designation was but if it’s a trace level we add 500 to it to differentiate it in AQS.

Question 12: The method code list is specific to instrument models but does not keep up with the rapidly evolving changes. The EPA "list of designated reference and equivalent methods" could be the best common-sense guide, then be consistent in submittals. Right?

Answer 12: Monitoring regulations require all data to be measured by approved regulatory methods. The method code list is the complete list of approved regulatory methods.

Question 13: Please clarify guidance on using the "method code plus 500" choice for trace level measurements.

Answer 13: I a few cases, methods approved many years ago, now have a “trace-Level” version of the method with the same method designation. To differentiate these methods from their conventional early version we have adopted a practice of adding 500 to the method code.

Question 14: Please clarify how some qualifier codes seem to be very much alike, such as "AS" and "EC", which is referenced in EPA guidance. Same idea for "1" and "1V".

Answer 14: The qualifier codes in existence today are historical, and often have overlap in applicability. Selection of the best one to use depends on your best judgment and coordinating with your Regional Contact.

Question 15: New flags were created 1C and 1V for QA and data records respectively. The flags are being accepted for transactions other than 1-point-QC transactions but comments can only be entered for 1-point-QC. Are there plans to add a comments section to other QA transaction forms in AQS?

Answer 15: Yes

Question 16: I have a question regarding the use of AQS Method Code 170 with Met One BAM 1020 PM2.5 monitors. My understanding is that the 170 Method Code has always been specific to the BAM 1020 with VSCC FEM monitors and used with Parameter Code 88101. However, AQS now allows the 170 code to be used with BAM 1020s operated with SCC in non-FEM configuration, with Parameter Codes 88501 and 88502. In the past, non-FEM BAM 1020s with SCC were coded as Method Code 731 for Parameter Codes 88501 and 88502.

Although AQS now accepts the 170 Method Code for non-FEM BAM 1020s, is it actually correct to be using it?

How can method code 170 be reported for a bam monitor using an SCC instead of a VSCC. Even if your region allows you to switch from 88101 to an 88502 data was still originally collected with an VSCC so why is the SCC being allowed?

Answer 16: No, it is not correct to be using it. Monitoring agencies should plan to run the Met One BAM 1020 in the PM2.5 FEM configuration and report the data to 88101. Legacy BAM 1020 monitors for PM2.5 that were never intended to be an FEM should use parameter code 88502 with method code 731. In limited cases an Met One BAM FEM for PM2.5 may be approved to report to 88502 with method code 170 if it does not meet certain performance requirements and the Regional office agrees with not using the data for NAAQS. It is true that there are some agencies that bought an FEM version of the 1020 and ran with SCCs rather than the approved VSCC; thus, rendering the method as no longer an FEM. However, no monitoring agency should be doing that anymore. Some legacy data is in there, but this is a reflection of what has happened in the past. There was a recommendation to add a note in the table on allowed combinations. The note would state that the Met One BAM 1020 with SCC is a legacy combination and nobody should be reporting a BAM1020 with an SCC as method code 170.

Question 17: When I run the AMP300 Violation report why does the max value given only extend to two decimal places? Can we make that three? The NAAQS is at three.

Answer 17: There is a change ticket in place for that we will get to it as resources allow.

Question 18: What is the plan for Screening Groups in AQS? In Maryland, we have 2 (Maryland and Maryland PM). It would be so much better for us not to have them split out like that. I heard (a while back) that the plan was to eliminate screening groups and I hope that's still the plan.

Answer 18: It is still the plan to eliminate screening groups and move to access control by Site and Monitor Agency Roles. This will be implemented when resources allow.

Question 19: I had a request, which I understood was forwarded by the help desk to the AQS Team back in April, to have “Delete All Selected” buttons added to the Correct QA data forms (i.e.: 1-Pt QC, Semi-annual Flow Audits, etc.) as already exists for the Correct Raw data forms. This would be very helpful when there are a large number of QA records having errors that would best be remedied by deleting all and resubmitting corrected QA records. My question is if there are plans to address this in the near future. I noticed that the recent updates didn’t include this.

Answer 19: You do not need “Delete All Selected” for this. All batch processing is by file. Just stop using the old file and submit a new one.

Question 20: The questions I have as below for PM2.5 in AMP 600 report:

How to calculate CV for the collocated site?
How to calculate CV Est?
How to calculate CV UV?

Could you show an example for each calculation?

Answer 20: If you want to replicate what is in the AMP 600 see Section 4.2 of Appendix A to Part 58. (https://www.ecfr.gov/cgi-bin/text-idx?SID=78e66d7cea6928ea986bc25221f78ed5&node=40:6.0.1.1.6&rgn=div5) Section 4.2 identifies the statistics for the assessment for the various PM and Pb parameters. Also, you can go to AMTIC website, there is a tool, the Data Assessment Statistic Calculator. You can see how those open in Excel and you can see how the equations are run in Excel. Go to AMTIC > Quality Assurance > Quality Indicator Assessment Reports > Data Assessment Statistical Calculator, then you can see how the formulas are run in Excel. To get there follow the link: https://www3.epa.gov/ttn/amtic/qareport.html

Question 21: Can you describe the best way to identify a QA-collocated monitor for PM2.5 in the AQS database? We have several sites with collocated monitors, but not all of them are specifically designated as QA-collocated in our network. For instance, a QA-collocated monitor has to meet certain siting criteria to satisfy the QA-collocation requirement in 58 Appx A. So is there a best practice on how to identify monitors in the AQS database as a monitor whose specific purpose is to satisfy the QA requirement? Is any monitor with a “N” for primary monitor in the QA-collocation tab (under maintain monitor) considered a QA-collocated monitor by default? Is it important to identify the monitoring objective as QA-collocation? In other words, in the AQS system, how would an agency differentiate their QA-collocated monitors from other collocated monitors of the same parameter, which are neither sited, nor intended to satisfy the QA-collocation requirement in 58 Appx. A.

Answer 21: The monitor type for all of these monitors presumably if you are a slams would still be slams the primary, the QA collocated, and a continuous are all still slams. The monitoring objective would tell us which monitor is the QA collocated monitor. For sites that have a QA collocated monitor, in AQS use the Maintain Monitor “QA Collocation” tab to designate the primary monitor (will have a ‘Y’ as noted above), and for the QA collocated monitor, again use the Maintain Monitor, “QA Collocation” tab to designate the QA monitor (with an ‘N’) as noted above. There are tutorials for this on the AQS website. QA EYE issue 23 pages s 7-9 provide additional information on this. The QA EYE can be found at https://www3.epa.gov/ttn/amtic/qanews.html

Question 22: Met One E-FRM method question

from List of Designated Reference and Equivalent Methods, June 15, 2018

^b Method code as available in the AQS data system

but AQS method code list shows

88101 Method Code List

and no 521 method

So, which is it?

Answer 22: The method codes have been updated to account for differentiating the VSCC and WINS configuration of the Met One PM2.5 FRMs. These updated method codes in AQS are now as follows:

Updated Method Code Table
Method	Designation No.	Method Code
Met One Instruments, Inc. E-FRM with WINS	RFPS-0315-221	221
Met One Instruments, Inc. E-FRM with VSCC™	RFPS-0315-221	521
Met One Instruments, Inc. E-FRM with URG	EQPS-0316-235	235
Met One Instruments, Inc. E-SEQ-FRM with WINS	RFPS-0717-245	245
Met One Instruments, Inc. E-SEQ-FRM with VSCC™	RFPS-0717-245	545

Question 23: For AMP 355 - Combined Site Sample Value report, could that be made to use with the same method code (when you have two samplers alternating in the sampler schedule)?

Answer 23: AMP355 will report the daily regulatory data for the site, independent of method. So, yes, it will work for two samplers on an alternating schedule.

Question 24: Where can I find the memo that talks about how to tag/code QC checks associated with ambient data being invalidated due to corrective action process?

Answer 24: The memo is on AMITIC and titled “Steps to Qualify or Validate Data after an Exceedance of Critical Criteria Checks” A QC check that is valid that ends up invalidating routine data should stay in AQS . A QC check where is was discovered that the QC system failed (so the QC check was invalid) should not be reported and a “1C” null code should be included in the QA Transaction qualier field.

The technical memo can be found at https://19january2021snapshot.epa.gov/amtic/policy-memoranda-and-technical-guidance

Question 25: How can I retrieve a set of data that I can import into Excel spreadsheet? I am trying to download AQI data and concentration data for several years> ACHD: Run an AMP 350 and use excel to calculate the 3-month averages

Answer 25: Nick Mangus - AQS Team: @ L W: You can run the Air Quality Index report in AQS. The output is space delimited and can be imported into a spreadsheet (but not as easily as a comma delimited file). Also, we post national level data here: https://aqs.epa.gov/aqsweb/airdata/download_files.html#AQI

Question 26: We have a monitor in the wrong screening group. Can we fix that? Or does someone at the EPA / AQS help center need to fix it?

Answer 26: Yes, groups can move a monitor from one of their screening groups to another of their screening groups. If there is a problem, contact the EPA Enterprise IT Screening Group (EISD) at 1-866-411-4372, choose option 9, or email EISD@epa.gov.

Question 27: How can I find out who my regional contact is? Joanna Casey: I am in region 8. Is there a specific AQS contact for each region?

Answer 27: The regional AQS contacts are posted on the AQS website at: https://19january2021snapshot.epa.gov/sites/static/files/2018-05/documents/aqs_regional_contacts_0.pdf

Question 28: Without Discoverer, we need to use "pre-generated" data files for "site listings" and "monitor listings" (https://aqa.epa.gov/aqsweb/airdata/download files.html) and use in our databases. Is there a way to add "Scale" to the Monitor file? Also, can you put "Revision Date" on the web page for these 2 files, and when can we expect to see the changes to those files made?

Answer 28: Yes, we can add measurement scale to the file. We can also post a revision date. Let us know if you need anything else related to these files. We usually update those files around the end of May and November.

Question 29: How far back can we correct raw data? There used to be a restriction on how far back we could correct raw data in AQS. (I think it was 5 years?) Is the restriction still in place?

Answer 29: The AQS system has a “freeze date” beyond which user cannot make corrections to raw data. That is 20 years. If a user wishes to make changes to data older than that they should contact their regional AQS contact who will make the request to the AQS_Team.

Question 30: What is the current status of updates to AQS pertaining to the 2016 Exceptional Events Rule, especially based off of the workgroup that met in 2017?

Answer 30: We will be implementing this as soon as we have a contractor in place (estimated, Jan. 2019).

Question 31: How do we deal with failed QC checks that have no data associated with them, but it affects the precision and bias calculations in the AMP600 report?

Answer 31: There is not much you can do about this. If the QC check was valid and there was an exceedance and the monitoring organizations invalidated the data that was represented by the check then the QC check should not be used in precision and bias calculations. The QC check is still reported so but will count towards completeness but the data will not be used in the assessment of precision and bias if the routine data has been invalidated.

Question 32: When you fill in a monitor's lat an lon metadata, does AQS automatically fill in CBSAs and related info?

Answer 32: Not currently, we are moving toward that, trying to update all geographic data eventually.

Question 33: Regarding screening groups, we have some legacy monitors that were managed by US EPA (eg, RTP, ERG) that were never completely closed, and we have no authority to do so. Should we just compile a list and submit a ticket?

Answer 33: Yes

Question 34: Ada: from List of Designated Reference and Equivalent Methods, June 15, 2018: MetOne E-FRM with WINS has method code 221. MetOne E-FRM with VSCC has method code 521. However, AQS method code list has MetOne E-FRM with WINS or VSCC has method code 221. So for VSCC is the method code 521 or 221?

Answer 34: See response to question 22 above.

Question 35: For AMP 355 Combined site sample value report - Could that be made to use with manual PM10 with the same method code. (when you have two samplers alternating in the sampler schedule)

Answer 35: Yes, we will add this to the list of requested enhancements.

Question 36: Where is the memo that talks about how to tag /code QC checks associated with ambient data being invalidated due to corrective action process?

Answer 36: The memo is on AMITIC and titled “Steps to Qualify or Validate Data after an Exceedance of Critical Criteria Checks”. The technical memo can be found at https://19january2021snapshot.epa.gov/amtic/policy-memoranda-and-technical-guidance.