Fill and Sign the Safa Agent Registration Form

School Air Toxics Monitoring Initiative Overview and L w n Lessons L n Learned n Barbara Driscoll Office of Air Quality Planning & Standards, EPA y g , Emission Inventory Conference, Sept 28, 2010 School Air Toxics: Initial Charge • Assess potentially elevated ambient air toxics levels at some of our nation’s schools • Schools selected: – Results of 2002 NATA – Results from 2008 USA Today Study (using 2005 TRI) – Recommendations from EPA regional offices, State and Local Agencies – 2 tribal schools • Criteria School selection: – Near large industrial sources – Urb n areas – n r interstates or airports Urban r s near int rst t s r irp rts – Mix of large and small sources Project Design • Monitor for key pollutants for at least 60 days collecting minimum of 10 valid samples • Collect meteorological data for at least 6 months if possible • Analysis: evaluate air toxics levels at each site for short- and long-term exposures short long term • Determine next steps based on sample results School Selection: Problems Identified • Several databases w/ school information • Risk calculation differences between NATA and RSEI model used by USA Today – NATA emphasis on cancer risk – RSEI higher weighting of non-cancer risk – Result – different key pollutants identified from each model • 2002 NEI versus 2005 TRI data – Concerns with accuracy of some information w u y f m f m • State and local agencies identified – Schools – renamed, closed, moved or scheduled for demolition – Facilities closed or emission estimates inaccurate – Sources not included in inventories During Monitoring Period at a School Individual Sample Review • Individual sample results reviewed in light of individual sample screening levels – Sample screening levels help us gauge potential for p pollutant levels in air to raise health concerns for short-term exposures h • Findings above sample screening levels were considered more closely, with regard to – S Sample QA/QC l – Other results for that pollutant at that school (e.g., pattern of concentrations) – Information regarding potential sources of pollutant at school and variability – Information regarding circumstances associated with health effects, and type of health effects 150 Cadmium PM10 (LC) ng/m3 0.07 0.07 Ashland City Nickel PM10 (LC) School (470215501) Antimony PM10 (LC) Cobalt PM10 (LC) Sample g Screening a Level 0.02 0.04 0.12 0.75 0.13 0.09 0.08 0.13 0.15 30 3 0.41 0.15 1.01 0.23 0.57 0.68 0.46 2.29 0.49 2.26 0.25 0.13 200 3 0.33 0.51 0.74 0.41 0.31 0.47 0.44 0.69 0.55 0.49 ng/m ng/m 0.1 0.4 04 2 6 /6 /2 0 0 9 0.76 9.56 0 76 9 56 0.41 0.17 0.51 0.61 0.47 0.65 0.84 0 41 0 17 0 51 0 61 0 47 0 65 0 84 5 /3 1 /2 0 0 9 ng/m3 0 35 0.35 5 /2 8 /2 0 0 9 Arsenic PM10 (LC) 5 /2 5 /2 0 0 9 580b 5 /1 9 /2 0 0 9 ng/m3 1.55 0.43 1.04 1.43 1.86 1.34 1.13 0.71 0.67 3.87 1.88 1.13 5 /1 3 /2 0 0 9 Chromium PM10 (LC) 5 /1 1 /2 0 0 9 500 5 /7 /2 0 0 9 2 2.22 6.61 5.88 16.8 3.51 4.06 8.27 0.99 56.4 7.96 3.14 2 5 /1 /2 0 0 9 7 0.6 06 4 /2 5 /2 0 0 9 ng/m3 Parameter 4 /1 9 /2 0 0 9 Units Manganese PM10 (LC) School Name 4 /1 3 /2 0 0 9 Example: Individual Sample Review ng/m3 0.17 0.03 0.08 0.04 0.18 0.06 0.05 0.1 0.7 1.7 2,000 0.02 1.05 0.07 0.04 100 Mercury PM10 (LC) ng/m3 0.44 0.61 0.41 1.25 0.25 0.29 0.19 0.18 0.09 0.25 0.07 0.14 3000c Beryllium PM10 (LC) ng/m3 0.002 ND 2E-04 ND 0.03 0.03 0.008 0.006 0.002 0.008 0.01 0.005 ng/m3 0.67 0.64 0.5 0.13 0.27 0.9 0.69 0.37 3.67 0.5 1.04 0.85 20 20,000 Selenium PM10 (LC) : Key Pollutant Initial Monitoring: Problems Identified • Some pollutants had high values relative to values typically monitored – Determined problem with VOC monitoring equipmenttimer – Evaluated data and developed criteria w/ NACAA input to validate/invalidate VOC data • Additional evaluation pointed to problems specifically w/ method used for acrolein – I iti t d st d of canisters and m th ds used b diff Initiated study f ist s d methods s d by different t labs – Determined acrolein values could not be used for analysis in SAT reports – E l ti how to improve acrolein methods Evaluating h t i l i th d • Anemometer used: – Would get stuck – report no data – Might report exceptionally high winds on calm days – Lightening and storms might set off Analysis At End of Monitoring Period at a Sch l School • Considers several types of information, including: – Concentrations of air toxics monitored at school f – Wind direction and speed measurements taken at the school – Information on nearby sources of air toxics y • Addresses key questions, such as – Was sampling conducted during time with potential to see f y ( ) p ( ) evidence of key source(s)/pollutant(s) • Were samples taken on days when winds indicate potential for suspected source(s) to be contributing to air concentrations at the school? • Was source(s) operating on sampling days? • Any indication that monitoring period conditions are not similar to conditions expected over longer-term? Analysis At End of Monitoring Period y E f g at a School – cont’d • Addresses k questions, such as dd key h – Do monitored concentrations of key pollutants (or others) indicate levels of concern for health impacts related to (short- or) long-term exposures? • Concentrations of key pollutants • Concentrations of other pollutants monitored • Concentrations of multiple pollutants (key or other) Analysis A l i At End of Monitoring P i d E d f M it i Period at a School 56.00 Manganese Noncancer-based Long-term Comparison Level C Concentration (ng/m3) n 48.00 40.00 32.00 24.00 16.00 Upper 95% Confidence Level pp on the Mean Average 8.00 0.00 Lower 95% Confidence Level on the Mean Projected Longer-term Average End of Monitoring Period at each School Technical Report – Describes analysis for individual school – Includes key findings and recommendations for next steps, such as: steps • Monitoring does not need to be extended • Extend monitoring to better characterize pollutant y concentrations n the community • Evaluate emission reduction options Non-technical Summary – Presented on EPA web site (www epa gov/schoolair) (www.epa.gov/schoolair) – Findings and analysis from technical report summarized in non-technical language • Technical report itself also available from web site Project Status • Initial I iti l monitoring complete at all 65 schools it i l t t ll h l • Some schools slated for additional monitoring • Additional monitoring ranges from additional screening analysis and additional monitors to high-end, state-of-the-art continuous metals monitoring p p p 7 final reports completed; additional reports to be finalized throughout Fall 2010 Final project summary report to be completed by summer 2011 All reports and data will continue to be posted on the Schools ( p g ) website (www.epa.gov/schoolair) • • • – – – – 63 schools in 22 states 2 tribal schools Final data release September 1, 2010 Over 73 000 d O 73,000 data points processed; 1 47 million values f i d 1.47 illi l form associated i d meteorological stations added to AQS data system – Screening analysis indicated levels of concern to continue monitoring in the community – Information about nearby sources being below normal operating capacity; continue schools monitoring – Acrolein measurement concerns; continue schools monitoring ( (timeframe TBD) ) SAT: Lessons Learned • Need better source specific information – Under CAA can not require states to collect air toxics information… but state data is critical! – As MACT rules revised requiring sources to submit emission information which may be used to improve inventories • TRI d t – N d b tt i f data Need better informed i d t d industry – TRI’s primary use is community-right-to-know – Data used for other purposes – sometimes regulatory – Better education of industry about other potential uses • If using models to inform, try to use most recent emission inventory – Over 60% of sources had significant emissions reductions from 2002-2008 • 24-36% of sources had increase in emissions from 20022008 – At l least 12 f iliti officially requested changes i t facilities ffi i ll t d h in 2002-2005 TRI data as a result of this project SAT: Lessons Learned –cont’d • Consistent application of monitoring methods and better methods – Working to improve method for acrolein • The easiest place to monitor isn’t always the best – Schools are representative of a population but may not p p p y best characterize the community • Need good met data – met collection methods • Need buy-in from partners y p • Helpful to pilot the concept and work out the bugs before implementation • Even what appears simple – will take longer than expected Questions? • B b Barbara D Driscoll ll – driscoll.barbara@epa.gov • School Air Toxics: – http://www.epa.gov/schoolair/ p p g