On the Relationship of Biogenic Primary and Secondary Organic Aerosol Tracer Compounds on the Aethalometer Model Parameters

The aethalometer model has shown to offer a fast, inexpensive and robust method for source apportionment. The method relies on aerosol light absorption attribution, mass balance of the total carbon and results in a fraction of unaccounted, residual carbon that has been associated to biogenic carbon due to its presumably non-light absorbing properties. This residual carbon and its relation to tracers of biogenic primary and secondary organic aerosol was investigated at a rural measurement station in Sweden. Special focus is devoted to 3-methyl-1,2,3-butanetricarboxylic acid (MBTCA), a secondgeneration oxidation compound in biogenic secondary organic aerosols. The results show that the residual carbon and the biogenic tracers show a high degree of correlation and that the tracers were highly seasonally dependent with largest carbon contributions during summer. MBTCA showed positive correlation with the aethalometer model derived absorption coefficients from fossil fuel carbonaceous aerosol, stressing the suspicion that biogenic aerosol might be falsely apportioned to fossil fuel carbon in the aethalometer model. MBTCA showed an increasing degree of correlation with higher aethalometer absorption coefficient wavelengths. However, spectrophotometric analysis revealed that the ambient concentrations of MBTCA are most likely to low to give a significant response in the aethalometer. These results support the application of MBTCA as a molecular tracer for biogenic secondary organic aerosol and indicates that a large fraction of the aethalometer model residual carbon is of biogenic origin. Future studies should investigate the light absorbing properties of precursor monoterpenes such as α-pinene, their oxidation products and eventual influence on the aethalometer model.

show a high degree of correlation and that the tracers were highly seasonally dependent with

A C C E P T E D M A N U S C R I P T
(30-500 Tg y -1 ) of monoterpenes such as α-pinene, β-pinene, Δ 3 -carene and limonene (Raisanen

A C C E P T E D M A N U S C R I P T
and suggested to use the m/z 141 as a signature for MBTCA in AMS measurements. Hence, 88 MBTCA has great potential acting as a high time resolution tracer for BSOA emissions.

A C C E P T E D M A N U S C R I P T
Thermo-optical analysis 139 Organic carbon (OC), elemental carbon (EC) and total carbon (TC) was derived from thermo- Light absorption measurements 156 The light absorption of ambient aerosols was measured with an aethalometer (AE33, Magee

A C C E P T E D M A N U S C R I P T
The aethalometer output parameter is attenuation coefficients which, after automatic artefact In these equations, the AAE is the absorption Å ngström exponent that describes the spectral 182 absorption dependence of the source specific aerosols. These parameters need to be accurately

A C C E P T E D M A N U S C R I P T
and 500 mmol L -1 was produced by dissolving weighed MBTCA powder in ultrapure water 258 (MilliQ). Spectrophotometric analyses were carried out using a portable spectrophotometer 259 (USB-650, Red Tide Spectrometer, OceanOptics) using a wavelength range of 200-850 nm. 2 260 mL of ultrapure water was placed in a quartz cuvette and used as background measurement.

261
The cuvette was then rinsed with ultrapure water before adding any MBTCA solution. Output 262 data from the spectrophotometer was absorbance (A). Absorbance was then transformed to and also increased AAE (1.58 and 1.52 compared to winter mean of 1.39, Fig. 1(b)). The

A C C E P T E D M A N U S C R I P T
transition from the 2013 to the 2015-2016 measurement period did not result in any abrupt and 308 significant change in the carbonaceous aerosol concentration.

309
The absorption Å ngström exponent (AAE) experience low levels during the spring and

324
Temporal variations of measured biogenic tracers are displayed in Fig. 1(c) and in Table 1.

A C C E P T E D M A N U S C R I P T
The temporal behavior of the aethalometer model output parameters is showed in Fig. 2(a).

359
The carbon concentration reflects the carbonaceous content, as displayed in Fig. 1(a)  in the range of 2 -4 times higher than TC, which obstruct any endeavor to accomplish a sound 377 source apportionment. It is indeed a non-trivial task to explain the highly elevated C1 parameter 378 found in this study. By studying the linear regression that was used to obtain the C1 parameter 379 (Fig. S1), it is evident that there are no clear outliers that could explain the elevated slope of the 380 regression line. Hence, from a statistical point of view, it seems like the slope is correct, and 381 that the issue is instead systematic. Another bias might occur when organic aerosol coats a soot 382 core and hence increases the MAC value of the aerosol (Bond and Bergstrom, 2006). This  is consistent with the temporal concentration variation for arabitol, sucrose and fructose as 463 displayed in Fig. 1(c) and Table 1. In this study, we demonstrated that four common biogenic tracers showed positive correlation 551 with the aethalometer model residual carbon, apportioned as biogenic carbon, CMBio. We also 552 illustrated the need and performance to modify and optimize the C parameters, in this case the 553 C1 parameter. It should be noted that such a modification demands a thorough comparison 554 between the aethalometer model output parameters and a tracer-based source apportionment 555 that includes radiocarbon, levoglucosan and biogenic tracer measurement. In this study we used 556 radiocarbon and levoglucosan source apportionment data generated from a previous study 557 conducted at the same measurement station. To conclude, several years of measurements may 558 be needed in order to establish stable C parameters for a particular measurement station, an 559 aethalometer instrument and a thermo-optical instrument.

A C C E P T E D M A N U S C R I P T
Surprisingly, we also found that MBTCA displayed a positive correlation with the derived 561 absorption coefficients from fossil fuel carbonaceous aerosol, stressing the suspicion that