Use of Advanced Probing Tools in One Atmosphere

Use of Advanced Probing Tools in One. Atmosphere Air Quality Models for Model Evaluation, Culpability Assessment and Control Strategy Design Presented at Community Modeling and Analysis System (CMAS) October 2006 Conference Chapel Hill, North Carolina Ralph Morris, Bonyoung Koo, Gary Wilson & Greg Yarwood Presents: /slides/ Presents: slides/ ENVIRON International Corporation 101 Rowland Way, Novato, CA. 94945 (rmorris@environcorp. com) 1

Introduction • Photochemical grid models (e. g. CMAQ and CAMx) are used to design emission control strategies to demonstrate attainment of the 8 -hour ozone and PM 2. 5 standards • Such models are quite complex with full-science representation of transport and diffusion, gas-, aqueousand aerosol-phase chemistry, cloud processes, gas/particle deposition etc. • It is difficult and computationally intensive to diagnose why the model obtained its solution, what corrective action is needed to improve model performance and which sources contributed to the high concentrations Presents: /slides/ 2

Probing Tools • Both CMAQ and CAMx incorporate “Probing Tools” that extract additional information: – Process Analysis (PA) • Identify the chemical processes in the simulation • Perform mass flux closure analysis – Decoupled Direct Method (DDM) • Sensitivity to emissions categories and regions, BCs, etc. – Source Apportionment (SA) • Identify source regions/categories that contribute to ozone, PM, etc. Presents: /slides/ 3

What is a Probing Tool? Probe verb, probed, prob‧ing, noun 1. to search into or examine thoroughly; question closely 2. to examine or explore with a probe 3. to examine or explore with or as if with a probe 4. a slender surgical instrument for exploring the depth or direction of a wound, sinus, or the like 5. an investigation, esp. by a legislative committee, of suspected illegal activity. Presents: /slides/ 4

Alien Probing Presents: /slides/ 5

Process Analysis (PA) • Pioneered out at research group at UNC Chapel Hill – Chemical PA; Mass Flux (IPRs, IRRs) • Extensively used for ozone – E. g. , identification of VOC- vs. NOx-limited ozone formation regimes in Houston (Jeffries et, al) – py. PA and py. PASS visualization tools (Vizuete et al. ; http: //www. unc. edu/~vizuete/research. htm) • Recently extended to PM modules in CAMx (CRC Project A-51 b, Tonnesen et al. , 2006) – Inorganic Aerosol, Organic Aerosol and Aqueous-Phase Chemistry (http: //www. crcao. com) Presents: /slides/ 6

Decoupled Direct Method (DDM) • Sensitivity of modeled concentrations to input parameters – Primarily used for emissions – Georgia Institute of Technology (GIT) implemented 3 -D DDM in CMAQ (URM) • Has assisted in the identification of control strategies (e. g. , GIT SAMI analysis) – DDM recently extended to PM modules in CAMx (CRC Project A-51 a, Koo et al. , 2006; http: //www. crcao. com) • Used for inverse modeling to identify potential areas of Presents: /slides/ missing emissions (e. g. , Houston ship channel VOC emissions) 7

Source Apportionment (SA) • CMAQ – Tagged Species Source Apportionment (TSSA; UCR) • Implemented for SO 4 and NO 3 families – Particle and Precursors Tagging Methodology (PPTM; SAI) • CAMx – Ozone Source Apportionment Technology (OSAT) • Used extensively for source culpability and optimal control strategy identification (e. g. , NOx SIP Call; CAIR; Ozone SIPs) – PM Source Apportionment Technology (PSAT) Presents: /slides/ • Just beginning to be used, examples follow 8

PM Source Apportionment for WRAP (Western United States for Visibility) • WRAP Applied CAMx/PSAT and CMAQ/TSSA PM Source Apportionment to WUSA – Identify source categories and states that contribute to visibility impairment at western US Class I areas – PSAT results available on WRAP RMC and TSS websites: • http: //pah. cert. ucr. edu/aqm/308/cmaq. shtml • http: //vista. cira. colostate. edu/tss/Default. aspx? code=1 Presents: /slides/ 9

WRAP PSAT/TSSA PM SA Source Groups 16 Source Regions 6 Source Categories = 98 Source Groups MV_ Mobile Sources PT_ Point Sources AR_ Area Sources ANF_ Anthro Fires NTWF_ Nat Fires NWF_ Non-WRAP fires AE_ Area+ Sources Presents: /slides/ 10

2002 Ranked SO 4 Contributions for W 20% Days Presents: /slides/ 1. AR Offshore 2. PT Offshore 3. BCON 4. PT CA 5. MV CA 6. AR CA 7. AR MX 8. PT MX 11

2002 vs. 2018 SO 4 SA Agua Tibia, CA 2002 Presents: /slides/ 2018 12

2002 vs. 2018 NO 3 SA Agua Tibia, CA 2018 2002 Presents: /slides/ 13

PM Source Apportionment for CENRAP • Regional Haze Rule (RHR) requires States to demonstrate Reasonable Progress toward clean visibility conditions in 2064 – Modeled 2018 visibility projections compared with Uniform Rate of Progress (URP) goal obtained by linear Glide Path from current (2000 -2004) Worst 20% conditions to 2064 Natural Conditions – Modeled URP goal can be important component of a State’s Reasonable Progress demonstration in their December 2007 RHR SIP Presents: /slides/ 14

Wichita Mtn, OK – Meets URP Goal Big Bend, TX – Does Not Meet UP Goal Presents: /slides/ Why WIMO, OK meets UP goal but BIBE, TX does not? Use SA to identify source regions and categories that contribute to visibility on W 20% days 15

Geographic Source Apportionment; 27 Regions * WIMO * BIBE 22 Separate States; rest of West and East US; Canada; Mexico Gulf. Mex ; IC; & BC Presents: /slides/ 16

Contribution to Visibility Impairment (Mm-1) for W 20% Days at Wichita Mtn, OK in 2018 70 ~20% visibility extinction at WIMO SOA_B & SOA_A All Sources due to uncontrollable sources: Mex, Can, BCs (Global Transport) BC and SOA_B Mexico VISTAS + MANE-VU CENRAP Presents: /slides/ 17

Contribution to Visibility Impairment (Mm-1) for W 20% Days at Big Bend, TX in 2018 SOA_B & SOA_A All Sources BCs (Global Transport) Mexico ~60% of visibility extinction on average of Worst 20% Days due to international transport CENRAP States (Texas largest) Presents: /slides/ 18

Total Visibility Bext Wichita Mtns Oklahoma Presents: /slides/ 19

SO 4 Visibility Bext Presents: /slides/ 20

NO 3 Visibility Bext Presents: /slides/ 21

Which Regional AQ Model is Used the Most for Regulatory Decision Making • CMAQ? CAMx? • UAM? URM? CIT? STEM? REMSAD? – None of the above – CALPUFF! – CALPUFF SO 4 & NOx “chemistry” – SO 4: – NO 3: Presents: /slides/ CALPUFF recommended long-range transport model 22

Photochemical Grid Models (PGMs) Not Used due to Perception they are Too Costly, Difficult and Fail to Treat Plumes • Development of PGM modeling databases costly – RPOs/EPA developed canned 36/12 km databases for 2001&2002 • Application of PGMs requires expertise – With databases available applications easier and available of Linux boxes make them easy and cheaper to apply • PGM resolution limited by grid cell size – New Plume-in-Grid modules treat near-source chemistry and plume dynamics (Pin. G & APT in CMAQ; Pi. G in CAMx) – two-way grid nesting can treat near- to far-field plume resolution • Need 2 runs to get a single source impact – PM Source Apportionment allows separate tracking of PM impacts from individual sources within one run (TSSA & PPTM in CMAQ; PSAT in CAMx) Presents: /slides/ 23

PGM resolution limited by grid cell size: Plume-in-Grid (Pi. G) and Two-Way Grid Nesting • Plume-in-Grid (Pin. G/APT/Pi. G) – Treats near-source plume chemistry and plume dynamics using a Lagrangian puff module – When puff size is commensurate with grid cell resolution, mass in puff is released for further tracking • Two-Way Grid Nesting – Flexi-nesting allows specification of higher resolution grids without need for other inputs for better resolution of point source plumes Presents: /slides/ 24

Texas BART Point Source Screening Analysis • Use of CAMx PGM to perform group BART screening exemption modeling – PM Source Apportionment Technology (PSAT) to obtained source-specific visibility impact at Class I areas from BART Sources – Two-way 36/12 km grid nesting to better resolve fair-field impacts – Use of Plume-in-Grid (Pi. G) with PSAT to resolve near-source plume chemistry and plume dynamics • Provides full-science chemistry from plume- to regional-scale within 3 -D PGM framework Presents: /slides/ 25

CENRAP 36 km Domain with Texas 12 km Flexi-Nest Grid Add 12 km grid using flexi-nest Presents: /slides/ 26

Texas 12 km Flexi. Nest Domain IMPROVE monitors (circles) Potential BARTeligible Sources (triangles) Treat emissions from BART sources with PSAT and Pi. G Presents: /slides/ 27

Conclusions: Hybrid PGMs for BART/PSD • Current generation of hybrid PGMs with grid nesting, Plume-in-Grid and PM Source Apportionment able to treat single source PM, ozone and visibility impacts at all scales – Use of full chemistry – Accounts for three-dimensional winds/dispersion • Available databases = Ease of use • Better science will result in more accurate and reliable assessments • Texas BART modeling demonstration: http: //www. tceq. state. tx. us/implementation/air/sip/bart/haze. html Presents: /slides/ 28