Ambient Measurement Programs in the United States Rich Scheffe, United States Environmental Protection Agency Good Day…. The purpose of this presentation is to provide you with an overview of some exciting work being done by EPA and representatives from various State, Local and Tribal organizations. The reengineering of air monitoring is going forward in no small measure in response to many of our mutual concerns, voiced during the PM 2.5 implementation. The future is upon us and we want to begin the process of getting you engaged in the effort, so that you may focus your staffs on the important work ahead. This strategy will impact your: Program’s policies and focus (New Data will give context to issues now being raised-toxics, diesel PM) Budgets (this work is expensive) Ability to provide air quality products to our interested and informed public. (Using AIRNOW and AQI and local websites) We are limited in time and want questions at the end. We will be posting more detailed information on the AIRWEB site and hope you will read find it helpful. EMEP Workshop New Orleans, Louisiana April 20, 2004

Topics Current networks…routine Anticipated changes Overview of incommensurability and artifact issues

National Level Routine Networks…S/L/T’s, EPA PM2.5: FRM, cont., spec (trends (daily), SIP, IMPROVE, SS); >1000 sites PM10 >1000 sites O3 > 1000 sites NOx/NO (NO2) > 400 SO2 > 400 CO > 400 Pb > 400 O3 precursors, PAMS >70 sites S, N deposition, CASTNET, > 50 sites

Current Network > 3,000 “sites” Many are single pollutant sites Different purposes and scales Expensive Fixed Redundant Outdated PM/SO2 VOC’s PM O3 CO SO2 Toxics 50 mi 15mi 25 mi 35 mi Real-time PAMS IMPROVE CASTNET

PM2.5 Networks SS Mass Sampling Routine Speciation 8 ~175 SIPs ~ 1050 FRMs ~ 200 cont. ~ 54 Trends ~175 SIPs ~150 IMPROVE 8

2000-2002 3-Year Average Annual Mean PM2.5 Data from AQS 7/9/03. Sites that operated anytime 2000-2002 (n=1239)

Urban and Rural PM2.5 Speciation Networks Active Sites as of 1/20/04: EPA data from AQS, IMPROVE data from VIEWS

Annual Urban PM2.5 Speciation Patterns

Annual Rural PM2.5 Speciation Patterns

Annual Rural PM2.5 Speciation Patterns Annual Urban PM2.5 Speciation Patterns Annual Rural PM2.5 Speciation Patterns

Annual Average Urban Excess

Supersites IRD Study Domain Southeastern Canada New York Pittsburgh Baltimore Fresno St. Louis Los Angeles Atlanta Houston Phase I Phase II Both Phases

Pittsburgh: Avg. Size Distribution Courtesy: Prof. Jose-Luis Jimenez et al. Dept. Chemistry & CIRES University of Colorado at Boulder

Strong Nucleation Pittsburgh, August 11, 2001 August 11, 2002 12x104 Number (#/cm3) 6x104 105 500 104 100 dN/dlogDp (cm-3) Response Particle Size (nm) 103 10 102 00:00 06:00 12:00 18:00 24:00 Time of Day

Chemistry of Growth: Particle Mass Spectra at 20-33 nm Detection of Nucleation by Particle Sizer Remember to talk about all points on the graph Zhang, Stanier, Caragaratna, et. al. Insights into the Chemistry of Nucleation Bursts and New Particle Growth Events in Pittsburgh based on Aerosol Mass Spectrometry. Environ. Sci. Technol., submitted.

Don’t believe it

Areas with PAMS Networks PAMS areas Type #2 site Type #1, #3, #4 site

Air Toxics Monitoring Network: Pilot sites and proposed trend sites

Comments on Historical Routine Networks Adequate ground level spatial coverage PM2.5 mass, ozone But, Very limited multiple pollutant sites Act of convenience rather than design Aerosols (mass and species) too much reliance on integrated techniques providing no diurnal characterization Criteria gases Except for ozone and NO, many meaningless measurements Trace levels, source oriented/microscale siting (CO, SO2) Other gases NOy….very limited True NO2..? VOCs…mostly ozone season through PAMS Other precursors and indicators Nitric acid and ammonia….episodic/intensive programs only Peroxides, hydroxyl radical….intensive programs only Artifacts/problems….later

Changes expected from Implementing National Monitoring Strategy Based on these concerns….. The United States EPA is undertaking a major modification of the national ambient air monitoring networks

National Core Network: NCORE Goal: Move from loosely tied single-pollutant networks to coordinated, highly leveraged multi-pollutant networks with real time reporting capability PAMS SO2 O3 PM PM CO PM IMPROVE CASTNET PM O3 Toxics

Principal Data Objectives of NCore Public Information Real-time Input of Data From Across the Country Using Continuous Technologies Spatial Mapping (E.G., AIRNOW), Health Advisories Health/Exposure Assessment Support Input for Periodic NAAQS Reviews Emissions Strategy Planning (Emphasis on Initial Timeframe) What are the best emission reduction approaches? E.g., Provide for Routine Model Evaluation and Source Attribution Air Quality Trends and Program Accountability Does the monitoring confirm strategies are working? Major National Initiatives (Acid Rain, Clear Skies, NOx SIPS, FMVCP) Including HAPS (National) and Visibility Assessments Science Support Backbone for More Diagnostic Level Work (Same for Local Sips), Health Studies NAAQS Determinations and Related Regulatory Rqmts. Emphasis on More Pervasive Ozone and PM2.5 Ecosystem assessment support

Speciation Program, Air Toxics NCore Measurements Level 1. 3-10 Master Sites Comprehensive Measurements, Advance Methods Serving Science and Technology Transfer Needs Level 2: ~ 75 Multi-pollutant (MP) Sites,“Core Species” Plus Leveraging From PAMS, Speciation Program, Air Toxics L2 Level 3 L1 Level 3: Minimum Single Pollutant Sites (e.g.> 500 sites each for O3 and PM2.5 and related spatial Mapping Support Minimum “Core” Level 2 Measurements Continuous NO,NOy,SO2,CO, PM2.5, PM10/PMc,O3,Meteorology (T,RH,WS,WD); Integrated PM2.5 FRM, HNO3, NH3,

Proposed Siting Approach – Level 2 “Representative locations” 5-15 km urban scale 50 km or more…regional scale “contrast with historical search for highest concentrations…at odds with collocation” Start With “Reasonable” Coverage From Health/ Exposure Perspective Population Based (Range of Sizes) With Varying Chemical Composition. Assumes Need for Multiple Pollutants to Tease Out Confounding Factors Add in Desired Rural Coverage Accountability of Major National Programs Such As 3P, NOx SIP) “Operational” Model Evaluation Equitable Resource (and Constrained) Considerations Determine Ability of Existing Networks to Address, Modify

Proposed Siting Approach – Level 2… Health meets atmospheric sciences Suggested Rural Locations for Level 2 Sites 3 2 1 4 10 24 9 12 11 20 7 6 18 19 21 8 5 13 17 16 15 14 23 Transport, Corridor, Background and Inflow Locations

Urban & Rural PM2.5 Speciation Networks Current/Planned Supplemental Information

Overview of Measurement…Modeling incommensurabilities and other issues Spatial representation Volumetric (model) versus point representation (measurement) Breathing level measurements….most of mass often elevated Measurement Artifacts Model attempts to characterize reality relative to true ambient properties, and natural removal processes (e.g., deposition to land/water/foliage) Measurements techniques alter ambient properties (heating, dehumidification), and removal processes (changing concentration gradients) within sampler universe

Also troubling is the delineated use of measurements and modeled predictions. …..Measurements are the current tool for strict regulatory applications, and models are used as a planning tool. ………The reality is that measurements really are just estimates of surrounding reality, and in one sense no different from a predictive output from a model. ……. Both these tools need to be more effectively merged to support in unity a host of regulatory and planning applications.