Thesis Defense: Loredana Suciu, Ph.D. Candidate
January 17th, 2020, 9:00 a.m.
Wild and prescribed fires are important sources of a broad suite of organic compounds collectively termed pyrogenic carbon (PyC). Most PyC compounds have additional sources beyond fire, adding uncertainty to their use as tracers. However, members of the anhydrosugar family of isomeric compounds – levoglucosan, galactosan and mannosan – are generated exclusively by the pyrolysis and combustion of cellulose and hemicellulose. Although anhydrosugars are some of the only unique organic markers for fire, their use as tracers in atmospheric, marine, and terrestrial systems is challenging because there is no clear theoretical framework to deal with their reactivity and phase partitioning. The atmospheric science community has made the first approximation that they are unreactive on timescales of interest. This assumption is problematic. The terrestrial and marine science communities have not yet seen wide use of anhydrosugars as tracers because our understanding of their biogeochemistry and transport through the Earth system is poorly constrained.
Chapter 2 of this thesis reviews evidence for anhydrosugar production, degradation and detection in various environments and use this information to develop a framework for uses of anhydrosugars in research on PyC and organic matter in the Earth system. Anhydrosugars are chemically reactive in all phases (gaseous, aqueous and particulate), molecularly diffusive in semisolid matter, semivolatile, water-soluble, and biodegradable. Their chemical composition also suggests that they sorb to soil mineral surfaces. Together, these characteristics mean that anhydrosugars are not conservative tracers. While these traits have historically been perceived as drawbacks, they may present opportunities for new research avenues, including tracking organic matter transport and degradation in multiple environments.
Chapter 3 of this thesis provides insights from modeling the atmospheric fate of the most abundant anhydrosugar, levoglucosan (LEV), using a zero-dimensional (0-D) framework that includes a realistic representation of levoglucosan chemistry. The mechanisms studied include heterogeneous chemistry (gas-aerosol surface reactions), homogeneous gas-phase chemistry (gas-gas reactions) and gas-particle partitioning (evaporation-condensation processes). The existing chemical mechanisms were developed to include the chemical degradation of levoglucosan and its intermediary degradation products in both phases (gas and particle). Simulations were initiated with (1) a particle-phase LEV concentration measured in a small prescribed-fire plume) (2) a gas-phase LEV concentration estimated from its vapor pressure and (3) concentrations of other species in the system needed for the full chemical mechanism, which were estimated using a three-dimensional (3-D) chemical transport model (CTM). Multiple sensitivity analyses of seven-day scenarios show that the degradation time scale of LEV varied from a few minutes to more than seven days, suggesting that LEV can be transported regionally, but its concentration is significantly reduced. Several first- or second- generation products (five in the gas phase and seven in the aerosol phase) resulted from the chemical degradation of LEV; most of the products include one or two carbonyl groups, one product contains a nitrate group, and a few products show the cleavage of C-C bonds. The relative importance of the products varies, some dominating over short time and others becoming important over longer time. These findings may have important atmospheric implications, such as the formation of secondary organic aerosols (SOA). Estimated SOA yields (10-80%) reveal that conversion of LEV to secondary products is significant and occurs rapidly in the studied scenarios. Validation of the LEV degradation by measurements from chamber experiments shows that the 0-D modeling approach in this study agrees well with the data (± 30% average absolute error) in a slower heterogeneous chemistry scenario and at low mass accommodation coefficients. Evaluation of the model predictions against measurements from a fire plume, however, shows that the model is less accurate (average absolute errors > ±30%). In both cases, extending the validation of LEV degradation beyond 3-5 hours using more data from chambers and fire plumes is an important next step to confirm the longer degradation time scales predicted by the model and assess the atmospheric implications.
From a tracing perspective, results in this study suggest that the interpretation of post-depositional LEV concentrations (such as distinguishing between vegetation sources or inferring vegetation/climate shifts in ice cores) must consider the fraction degraded in the atmosphere. However, a 3-D CTM perspective in addition to the 0-D perspective of this study is necessary to improve estimates of the atmospheric degradation of LEV and its isomers.
Another application of anhydrosugars would be to trace regional air transported to highly polluted urban areas, such as the Houston area. Vegetation fires occurring outside this region contribute emissions of ozone (O3) precursors, such as volatile organic compounds and nitrogen oxides (NOx). However, in the Houston area, there are multiple sources of such emissions (industrial activity, vehicle exhaust, etc.), and mixing with those sources challenges the quantification of regional contributions to locally measured concentrations. This is important because air pollution control measures impact the industrial activity in the area. While anhydrosugars have not been used in this study to help constrain regional background O3 and NOx, they open an unexplored pathway for future studies that can build on the additional work presented in Chapter 4 of this thesis, such as the estimation of regional background O3 and NOx using statistical analysis of O3, NOx and meteorology measured in the Houston-Galveston-Brazoria (HGB) region. This study used four different approaches based on principal component analysis (PCA). Three of these approaches consist of independent PCA on both O3 and NOx for both 1-h and 8-h levels to compare the results with previous studies and to highlight the effect of both temporal and spatial scales. In the fourth approach, O3, NOx and meteorology were co-varied.
Results show that the estimation of regional background O3 has less inherent uncertainty when it was constrained by NOx and meteorology, yielding a statistically significant temporal trend of -0.68 ± 0.27 ppb y-1. Likewise, the estimation of regional background NOx trend constrained by O3 and meteorology was -0.04 ± 0.02 ppb y-1 (upper bound) and -0.03 ± 0.01 ppb y-1 (lower bound). The best estimates of 17-y average of season-scale background O3 and NOx were 46.72 ± 2.08 ppb and 6.80 ± 0.13 ppb (upper bound) or 4.45 ± 0.08 ppb (lower bound), respectively. Average background O3 is consistent with previous studies and between the approaches used in this study, although the approaches based on 8-h averages likely overestimate background O3 compared to the hourly median approach by 7-9 ppb. Similarly, the upper bound of average background NOx is consistent between approaches in this study but overestimated compared to the hourly approach by 1 ppb, on average. The study likely overestimates the upper bound background NOx due to instrument overdetection of NOx and the 8-h averaging of NOx and meteorology coinciding with maximum daily eight hours average O3.
Regional background O3 and NOx in the HGB region both have declined over the past two decades. This decline became steadier after 2007, overlapping with the effects of controlling precursor emissions and a prevailing southeasterly-southerly flow.