An evaluation of mass absorption cross-section for optical carbon analysis on Teflon filter media

Figures & data

Table 1. List of sites within each EPA Region chosen for analysis in the tailored MAC study.

Region	Site name	Site ID	Count of filters (n)
Region 1	Criscuolo Park, CT	090090027	127
	Burlington, VT	500070012	34
	Roxbury (Boston), MA	250250042	76
Region 2	Whiteface, NY	360310003	35
	Albany Co HD, NY	360010005	148
	Queens College, NY	360810124	184
	Rochester, NY	360551007	157
	Division St., NY	360610134	134
	IS 52, NY	360050110	24
Region 3	McMillan Reservoir, DC	110010043	160
	PHILA–AMS Laboratory, PA	421010004	140
	Arendtsville, PA	420010001	83
	Essex, MD	240053001	150
Region 4	Spring Hill Elementary School, TN	470931020	70
	Millbrook, NC	371830014	167
	North Birmingham, AL	010730023	188
	South DeKalb, GA	130890002	161
	Douglas, GA	130690002	89
Region 5	Dearborn, MI	261630033	87
	Maple Canyon, OH	390490081	98
	Gary, IN	180890022	94
	Com Ed, IL	170310076	160
Region 6	Del Norte, NM	350010023	97
	Deer Park, TX	482011039	269
	Shreveport Airport, LA	220150008	97
Region 7	Liberty, MO	290470005	178
	Jefferson Elementary, IA	191630015	191
	Bonne Terre, MO	291860005	111
	Wichita Department of Environmental Health, KS	201730010	91
Region 8	Cheyenne NCore, WY	560210100	25
	Denver Animal Shelter, CO	080310025	130
	Lindon, UT	490494001	89
	Commerce City, CO	080010006	123
	Sioux Falls School Site, SD	460990008	125
Region 9	North Los Angeles, CA	060371103	146
	Fresno–First Street, CA	060190008	181
	Phoenix Supersite, AZ	040139997	179
	Riverside–Rubidoux, CA	060658001	309
Region 10	Portland–SE Lafayette, OR	410510080	144
	Beacon Hill, WA	530330080	144
	Fairbanks State Bldg, AK	020900010	163

Figure 1. Comparison of Robotic OT21 and Benchtop OT21 IR and UV attenuation for filters previously analyzed on the Benchtop in 2010 and analyzed (a) in 2014 on the Robotic OT21 and (b) as recheck on the Robotic OT21 of a subsample (39) of these filters in 2016.

Figure 2. Comparison of Robotic OT21 and Benchtop OT21 analysis for unsampled and sampled filters illustrating the correlation for (a) raw transmission for MTL brand filters and (b) raw transmission for Whatman brand filters.

Figure 3. Raw transmission of blank filters analyzed for this study where individual lots are grouped and averaged. The count (n) is that of blank filters from the respective individual lot. The All Average Blank is that of every blank filter from the time frame of the archived filter analysis.

Figure 4. Raw IR transmission values at 880 nm as a function nominal filter weight for unsampled and exposed MTL filters in use at the time of this study.

Figure 5. Raw IR transmission reading at 880 nm for exposed filters as a function of PM_2.5 mass concentration (μg/m³).

Figure 6. Investigation of a lot-specific derivation of MAC for lots 2010-5 and 2011-5 illustrating differences between using (a) Lot-Specific I₀ and (b) All Average I₀.

Figure 7. Correlation of the calculation of BC using an estimated MAC of 10 m²/g to the EC TOR by IMPROVE_A for the (a) 2011-4 and (b) 2011-5 Lot-Specific I₀.

Figure 8. Image of the 10 EPA Regions as found on https://www.epa.gov/aboutepa/visiting-regional-office.

Table 2. MAC derivation for the 10 EPA Regions.

2010/2011 EPA Region	Slope σatt[λ] (m^2/g)	Intercept	r^2	Count (n)
Region 1	8.36	2.39	0.65	237
Region 2	9.11	1.94	0.57	682
Region 3	9.44	1.89	0.54	533
Region 4	9.24	3.05	0.50	675
Region 5	7.95	2.83	0.40	439
Region 6	7.25	3.19	0.32	463
Region 7	6.86	3.23	0.17	571
Region 8	8.49	2.34	0.52	492
Region 9	9.30	2.04	0.70	815
Region 10	8.90	1.48	0.76	451

Figure 9. Seasonal variation of the (bars) MAC and (circles) correlation across all 10 EPA Regions.

Table 3. Derived slope (MAC), y-intercept, and correlation for the seasonal variation in the relationship of light attenuation as a function of EC by IMPROVE_A TOR (μg/m³) for Region 7 and Region 10.

Season	Region 7				Region 10
	Slope σatt[λ] (m^2/g)	Intercept	r^2	Count (n)	Slope σatt[λ] (m^2/g)	Intercept	r^2	Count (n)
All	6.86	3.23	0.17	571	8.90	1.48	0.76	451
Spring	8.49	1.12	0.28	109	6.75	0.96	0.65	116
Summer	7.93	3.50	0.19	171	9.18	2.16	0.18	108
Fall	4.84	4.50	0.10	146	8.53	2.70	0.54	124
Winter	6.94	3.05	0.20	145	9.43	0.61	0.87	116

Figure 10. Average PM_2.5 chemical compositions in EPA Regions 3, 7, and 10 during summer and winter seasons as measured by the CSN. The remaining species was calculated by subtracting the sum of the species indicated from the total PM_2.5 mass concentration. Note that this category will include the multiplier for OC, soil components, and any undetermined species.

Figure 11. Seasonal evaluation of the fine soil fraction using aluminum (Al), calcium (Ca), iron (Fe), silicon (Si), and titanium (Ti) analyzed by XRF for (a) Region 1, (b) Region 2, (c) Region 3, (d) Region 4, (e) Region 5, (f) Region 6, (g) Region 7, (h) Region 8, (i) Region 9, and (j) Region 10. Note that the y-axis scales are same for all regions, except Region 6.

Figure 12. BC calculation using a constant MAC of 10 m²/g as a function of PM_2.5 mass concentration (μg/m³) using (a) All Average I₀ on archived Whatman filters and (b) filter-specific I₀ on MTL filters in use during the study.

Table 4. Region 9 relationship of PM_2.5 mass concentration (μg/m³) for the derived slope (MAC), y-intercept, and correlation.

Region 9 PM mass concentration (μg/m^3)	Average EC TOR by IMPROVE_A (μg/m^3)	Slope σatt[λ] (m^2/g)	Intercept	r^2	Count (n)
0–84.5	0.99	9.3	2.04	0.70	811
0–3.0	0.14	85.28	–6.76	0.02	3
3.1–6.2	0.39	5.83	2.78	0.15	104
6.3–10.4	0.7	7.3	3.16	0.37	225
10.41–20.69	1.1	9.13	1.96	0.65	356
20.7–84.5	1.6	8.97	4.38	0.75	123

Figure 13. Counts of individual filters within each mass concentration range for EPA Regions for all the sampled filters analyzed in the MAC study.