State Comments on Frozen Data - 2010

For Matrix A01B1S:

Facility ID 10-001-00015 showing as SM facility, it isn't, so manually corrected in AFS (this would reduce the count from 84 to 83)

Facility ID 10-777-00233 did not have a SM air program set up in DEN as it should have, we have since set it up and done an upload to AFS (this would put the count back up to 84)

For Matrix A01C2S:

Facility ID 10-003-00324 had a NESHAP air program in AFS, however state's database did not have a NESHAP air program, so not sure how AFS had one. Manually shut down the NESHAP air program in AFS.

For Matrices A01F1S & A01F2S:

A NOV for Facility ID 10-001-00117 was not counted (both for the informal enforcement action and the source). It is in AFS, not sure why it's not in the OTIS data verification spreadsheet.

For Matrix A01J0S:

Delaware's database is incorrectly combining penalty and cost recovery and reporting the combined total as penalty to AFS. Therefore, the following are the actual penalties assessed in FFY 2010:
Facility ID 10-003-00365 -- $20,000 penalty (instead of $23,000)
Facility ID 10-003-00016--$1,945,000 (there was no cost recovery)
Facility ID 10-003-00087--$9,900 penalty (instead of $11,385)
Facility ID 10-003-00705--$12,775 penalty (instead of $14,691)
Grand Total of Assessed Penalties: $1,987,675 (instead of $1,994,076)

For Matrix A05E0S:

The unknown compliance status for Facility ID 10-003-00516 was caused by the un-archiving of the facility.

Do not understand how Facility ID 10-001-00002 had unknown compliance status for January and February of 2010. State does not think it should have had that status however, due to state database's lack of ability to track compliance status by historical date(s), no way of reconciling.

For Matrix A05G0S:

Counted should be increased by 2 and Not Counted reduced by 2. Review of 2010 Self-Certifications for Facility ID 10-003-00063 & 10-003-00021 were completed on 10/4/10.

The following caveats should be posted:

1. Some gasoline station partial and full compliance evaluations have not yet been entered into the system, though they were performed in FY2010.
2. Some Area Source MACT sources do not have appropriate air programs and subparts identified, primarily gasoline stations and perchloroethylene dry cleaners.

CAA Subpart Designations: Percent NESHAP facilities with FCEs conducted after 10/1/2005: This metric does not appear to be properly calculating the correct percentage, because it is including facilities that have shutdown operation (i.e. status = 'X'). When considering only operating sources in the universe of sources, 99% of the facilities have the correct data.

Historical Non-Compliance Counts (1 FY): Sources on this list include those under the responsibility of EPA Region IV, which has lead enforcement over the CFC sources and their compliance status.

Number of Sources with Unknown Compliance Status (Current): The # of facilities with an unknown compliance status will change once EPA runs their Unknown Compliance Status Generator in the Air Facility Subsystem for the month of February (2011). Only a few sources will remain in this metric after the utility has run.

In addition to the general caveat that Region 5 is providing to address data issues that involve data entered by EPA in the State agency's compliance status field instead of EPA's field, there are a few ongoing issues we deal with that affect how Indiana data is represented in AFS:

1. We cannot upload all stack tests conducted because AFS does not accept more than one test per facility per date per pollutant; and,
2. The length of time an Indiana source appears to be out of compliance is not accurately reflected in OTIS/ECHO as a result of the periods between data pulls and AFS updates. Frequently the reported period of non-compliance is inconsistent with the actual non-compliance (usually showing up as longer) and also reflected as occurring in a later period than actually occurred.

U.S. EPA Region 5 has reviewed the CAA data for all six States, specifically, the entries reported to AFS for the air program pollutant compliance status field at the source level. There are data issues that still remain which involve inaccurate data entered by EPA in the State agency's compliance status field instead of EPA's field. Region 5 will continue working with the States and HQs to clean-up and correct all inaccurate data entered for all six States.

Maine DEP has posted information regarding data completeness and data interpretation at the following website: http://www.maine.gov/dep/enforcement/echo.html.

In addition to the general caveats at the above link, please consider the following specific caveats and corrections:

1. Maine DEP conducted inspections at LQGs that are part of the RCRAInfo LQG Universe, but these LQGs do not show up in the Biennial Report LQG universe. Because EPA calculates inspection coverage by using the Biennial Report LQG universe, Maine's inspection coverage is not fully captured by EPA's metric [EPA metric 5.B.0 = Inspection coverage for LQGs (1 FY)]. EPA's metric for percent coverage of the LQG Universe underestimates the actual coverage of the LQG Universe by Maine inspection activities.
2. The data issue identified in item #1 above also occurs for EPA metric 5.C.0 [Inspection coverage for LQGs (5 FYs)]. Again, during the five year review period, Maine's inspection coverage for LQGs included LQGs that are part of the RCRAInfo LQG Universe, but which do not show up in the Biennial Report LQG universe. As a result, EPA's metric for inspection coverage for the five fiscal year review period understates Maine's inspection coverage for LQGs.
3. Maine conducts inspections at facilities that do not have EPA ID#s and are not required to notify EPA or Maine of their hazardous waste activities (i.e. Maine-defined Small Quantity Generators, i.e. federal CESQGs, and other non-notifier facilities identified through citizen complaints). The inspections activities, enforcement actions, and resulting monetary penalties assessed or collected as a result of such enforcement activities by Maine are not captured or reflected in the EPA metrics. As a result, EPA's metrics understate Maine's full range of hazardous waste compliance, inspection, and enforcement efforts.
4. Under Metric 10.A.0, EPA reports that 33.3 % of SNCs had formal action/referral taken place within 360 days (1 FY). This metric is incorrect. The actual percentage is 100%. The EPA data metric does not count or include formal actions taken in two case-specific examples listed by EPA as "not counted". In fact, formal actions were concluded with both Northeast Coating Technologies and eWaste Recycling Solutions within the 360 days metric measure. If these two cases are included in the calculation, then 100% of the SNCs identified in that metric had formal actions taken within 360 days.

The State of Minnesota has completed data verification for the 2010 SRF AQ data and has made appropriate corrections.

Please see the attachments and comments below. The first attachment contains the synthetic minor inspections conducted by MPCA the past 5 years, 102 facilities, the second attachment contains the actual Minnesota SM80s, 41 facilities. The 3rd attachment lists the 20 SM Universe corrections that were made to bring the AFS and Minnesota SM Universe counts into agreement. The 4th attachment lists the HPV Pathways added during FFY2010 showing the 6 "missing" discovery dates.

The 6 unrecognized (by AFS) discovery actions were replaced with actions recognized by AFS. Similarly unrecognized HPV discovery actions from FFY2009 (attachment not included) were also replaced with actions recognized by AFS.

John seems to have figured out why him and I have different numbers when looking at the test reports reviewed in the EPA database. He uses our stack test screens in Delta to populate the EPA database. So, if we have completed the review and sent the letter out but have yet to get the screen filled in, which sometimes does take a while since this seems to carry less priority than getting letters out in a timely manner, then John is not aware that these reports have been reviewed as well.

We're not sure how to address this situation or if in fact we even need to. How often do discussions like this come up with EPA? If we knew in advance a dump of the data in their database was going to be completed, we could prioritize and see that we get the stack test screens populated prior to this. Or, it may be enough to just send you a list of the other facilities that we are aware of having reports reviewed, similar to what was done this time. You would know that any difference between the EPA database and our list was due to the data entry lag time. We could also copy John on our NOCs however, this seems redundant for him to populate the EPA database since we are already needing to populate Delta with the same information on a routine basis.

U.S. EPA Region 5 has reviewed the CAA data for all six States, specifically, the entries reported to AFS for the air program pollutant compliance status field at the source level. There are data issues that still remain which involve inaccurate data entered by EPA in the State agency's compliance status field instead of EPA's field. Region 5 will continue working with the States and HQs to clean-up and correct all inaccurate data entered for all six States.

Metric 3.A.1 - Percent HPVs Entered ≥ 60 Days After Designation, Timely Entry (1 FY)

This metric evaluates the timely entry of HPVs in AFS by determining how long it takes to enter the Day Zero date in AFS. Because of the difficulty in AFS changing a HPV day zero to a non-HPV day zero, and vice versa, we do not enter a day zero until we're certain which it is. Therefore, we often do not make the determination until we've reviewed the company's response to our NOV. This means our actual determination of HPV status is after the day zero. Consequently, we cannot guarantee day zeros are entered within 60 days; however, we do believe that the majority of the time day zeros are entered into AFS within 60 days of our determination of HPV/non-HPV. Additionally, issues with the batch upload process and the Universal Interface have contributed to data entry delays.

Metric 3.B.1 - Percent Compliance Monitoring related MDR actions reported ≥ 60 Days After Designation, Timely Entry (1 FY)

This metric evaluates the timely entry of actions such as a Stack Tests, ACCs or FCEs in AFS by determining how long it takes to enter the actual action into AFS. For a Stack Test, the data entry is measured from the date the stack test was conducted. Permits typically allow a company 45 to 60 days to submit a stack test report. Our data entry on the report does not occur until our review of the report has been completed. This process typically creates the potential for the 60 day entry time to be exceeded. Stack test for the FY 10 review comprised of 38% of the universe in this metric. Additionally, issues with the batch upload process and the Universal Interface have contributed to data entry delays.

Metric 3.B.2 - Percent Enforcement related MDR actions reported ≥ 60 Days After Designation, Timely Entry (1 FY)

This metric evaluates the timely entry of NOVs and discovery activities in AFS by determining how long it takes to enter the action date in AFS. As discussed in 3.A.1, our business process cannot guarantee HPV day zeros are entered into AFS within 60 days. NOV's and most discovery activities are not entered into AFS until the appropriate day zero is created to facilitate linking of these actions to the day zero. We believe data entry for the other enforcement related MDR's is generally timely. Additionally, issues with the batch upload process and the Universal Interface have contributed to data entry delays.

Metric 10.A - Percent HPVs not meeting timeline goals (2 FY)

This data metric evaluates only the Air HPV enforcement timeline goals and does not account for more than one media being evaluated. A significant number of MDEQ's enforcement actions are multi-media and our business practice is to address all violations, regardless of media, under one enforcement action. Due to the complexity of having multimedia violations included in one enforcement action and each media having their respective timeline(s), a particular media's timeline goal may be exceeded in resolving the enforcement.

Metric R05B0S - Inspection Coverage for LQGs (1FY) - The universe that SRF is using is the number of LQGs that filed a report for the BRS reporting cycle 2009. Mississippi's LQG universe has changed significantly since March 1, 2010, and does not accurately reflect the State's LQG universe as of September 30, 2010. There are 145 active LQGs on February 17, 2011. Metric R01A2S is a more accurate representation of Mississippi's LQG universe for FY10, and should be the universe to measure inspection coverage (Metric R05B0S & R05C0S) for the LQGs.

The SRF counted universe for FY 2010 LQG inspection coverage is 39. The State performed a total of 52 LQG inspections during FY 2010. The discrepancy is due to a change in generator status at 13 facilities who are no longer LQGs.

Based on the number of facilities that have LQG inspections for FY 2010, the inspection coverage should be 52/145, or 35.9%.

Metric R05D0S - Inspection Coverage for SQGs (5FY) - The SRF counted universe for FY 2006-2010 SQG inspections coverage is 133. The State performed a total of 254 SQG inspections at SQGs for FY 2006-2010. The discrepancy is due to a change in generator status at facilities that had inspections during this period.

Based on the number of facilities that have SQG inspections for FY 2006-2010, the inspection coverage should be 254/389, or 65.3%.

NY does not use the federal generator status field in RCRAInfo to determine a site's generator status, i.e. LQG, SQG, and this field is inaccurate for NY sites. We use an internal database of manifest data instead.

NY also does not use the active site field in RCRAInfo.

This results in the calculation of RCRA State Review Framework Metrics being inaccurate, especially # 5 b, c, and d.

1. Due to program commitments and manpower limitations, ODEQ does not code general & minor facility enforcement actions (metrics 1E.3, 1E.4, 1F.3, & 1F.4), inspections (metrics 5B.1 & 5B.2), or SEVs (metric 7A.2) into ICIS-NPDES.
2. Metric 1A.2 is a combined (State & EPA Region 6) universe. Oklahoma Water Quality Division can verify that 125 of the 182 facilities listed are state non-major general permits.
3. ODEQ's fiscal year (July-June) is different than EPA's fiscal year (October-September), which has caused the discrepancy of the major inspection coverage (metric 5.A) to report fewer inspections than the actual value.
4. Metrics 7.B & 7.C (facilities with unresolved compliance and permit schedules) reflect the schedules for ODEQ's major facilities only.

In addition, the metrics1.F, 1.G, 5.A, 5.B, & 6.B values have been updated in the list below to reflect the correct values listed in the revised column. An itemized spreadsheet of the penalties & enforcement data has been included to support the revised values stated.

"Metric entitled 'CAA Subpart Designations: Percent NESHAP facilities with FCEs conducted after 10/1/2005 is in error': This metric indicates that 98 NESHAP facilities in AFS were missing subpart data. However, these 98 facilities were not subject to non-MACT NESHAP and had non-MACT NESHAP air programs and program pollutants marked in error."

"Metric entitled 'CAA Subpart Designations: Percent NSPS facilities with FCEs conducted after 10/1/2005': This metric indicates that 4 NSPS facilities in AFS were missing subpart data. However, these facilities were not subject to NSPS and had NSPS air programs and program pollutants marked in error."

"Metric entitled CAA Subpart Designations: Percent MACT facilities with FCEs conducted after 10/1/2005': This metric indicates 2 MACT facilities in AFS were missing subpart data. These facilities were not subject to MACT and had been designated as such based upon NAICS codes, which had been entered incorrectly. NAICS codes have been corrected, and MACT air programs and program pollutants have been closed."

Metric 5.A (Inspection coverage for operating TSDFs (2 FYs)): TCEQ inspection coverage for TSDFs in FY09 was reduced due to TCEQ state-wide assistance with environmental response efforts related to waste-management issues (ex. debris management) resulting from Hurricane Dolly (landfall July 2008) and to even a greater extent, Hurricane Ike (landfall September 2008). Based on the logic in the EPA's National Program Manager Guidance, TCEQ inspection coverage for TSDFs in TCEQ's FY10 (9/1/2009 to 8/31/2010) was 58.8%, exceeding the 50% annual coverage requirement for operating TSDFs.

Metric 5.C (Inspection coverage for LQGs (5 FYs)): Under the RCRA Core LQG Pilot Project (alternative approach), TCEQ was allowed to use a 2:1 SQG/CESQG CEI to LQG CEI to count towards their universe coverage for LQGs, allowing them to redirect resources to facilities other than LQGs. Under this alternative approach, the TCEQ meets the 100% coverage requirement for LQGs over a 5 year period.

The number of multi-facility orders varies between the state database and RCRAInfo. The state database counts multi-facility orders as a single order.

A translation problem between EPA and the state exists for handler data, so the data metrics do not reflect the correct number of active sites in Washington.

Metric 1A2 (Active LQG Universe) does not accurately reflect this universe in Washington. Ecology uses an Ecology developed database named TurboWaste to track annual reports of hazardous waste generation and management. TurboWaste contains the correct record of which sites are active and which are inactive. Ecology has had difficulty translating the handler records from TurboWaste to RCRAInfo, especially regarding active and inactive site status. We continue working towards a resolution of the problem.

Ecology's BRS submittal as shown in metric 1A5 is correct and provides accurate information for the related LQG metric calculations. Inspection rates for LQGs are calculated by the SRF using BRS data and are therefore correct.

The calculation affected by the universe count issue is Metric 5D, the 5 year inspection coverage for active SQGs. As Metric 5D is "Informational Only", it is not a great concern, although the correction would be to Ecology's benefit. Currently, OTIS lists 1,328 SQGs in Washington with a calculated SQG inspection coverage of 21%. Ecology's TurboWaste system currently lists 602 SQGs, which would result in a calculated SQG inspection coverage of 46%.