2008 Progress Report: An Instrument for Real Time Speciation of Water Soluble Tracers in Atmospheric Particulate Matter

EPA Grant Number: R832835
Title: An Instrument for Real Time Speciation of Water Soluble Tracers in Atmospheric Particulate Matter
Investigators: Weber, Rodney J. , Schauer, James J. , Shafer, Martin M.
Institution: Georgia Institute of Technology , University of Wisconsin Madison
EPA Project Officer: Chung, Serena
Project Period: May 1, 2006 through April 30, 2010
Project Period Covered by this Report: May 1, 2008 through April 30,2009
Project Amount: $492,354
RFA: Continuous Measurement Methods for Particulate Matter Composition (2005) RFA Text |  Recipients Lists
Research Category: Particulate Matter , Air , Air Quality and Air Toxics


Terrestrial carbon (C) pools play an important role in uptake, deposition, sequestration, and emission of atmospheric mercury (Hg). Biomass and soil C pools are highly sensitive to climate and land use changes with potentially serious consequences for the fate of large Hg pools, including past atmospheric Hg pollution, associated within these C pools. Our overall objective is to assess how global change during the next 100 years is likely to affect Hg cycling processes (i.e., atmospheric Hg uptake, sequestration, and emission) associated with vegetation and soil C pools. This will be achieved by developing a first systematic inventory of total Hg pools associated with and sequestered in U.S. forest ecosystems, assessing how global change will affect plant-derived atmospheric Hg inputs to ecosystems via changes in plant productivity, plant senescence, and litterfall; assessing fate processes of Hg sequestered in terrestrial C pools during decomposition processes; and modeling how global change impact on above processes may feedback on the future atmospheric Hg burden.

The work is implemented during a 4-year period through the following interlinked tasks:

    Task 1: We will systematically quantify Hg concentrations associated with vegetation, litter, and organic soil horizons in forested ecosystems across the United States.

    Task 2: We will assess the fate of Hg during C mineralization processes using controlled laboratory incubation and field studies to evaluate to what degree decomposition of organic matter leads to emission and re-emission of Hg to the atmosphere, increased mobilization within terrestrial ecosystems, or long-term sequestration by incorporation and accumulation of Hg in the remaining C matter fraction.

    Task 3: We will statistically evaluate and model the dependence of atmospherically derived plant and organic Hg pools and fluxes on climatic conditions.

    Task 4: We will integrate data on measured plant, litter, and soil Hg pools and fluxes—along with the hierarchy of climatic factors—into an existing terrestrial C model (i.e., the TECO model; Luo and Reynolds 1999). We will use model results to explore possible mitigation measures for atmospheric Hg and land management options to stabilize Hg pools associated with terrestrial C pools.

Progress Summary:

Year 2 of the study has mainly focused on Task 1 and 2, which will build the foundation for the modeling and scaling Tasks 3 and 4.  We have also started Task 3 and Task 4, including cooperation with the modeling group of Prof. Y. Luo to discuss detailed modeling approaches to be conducted in year 3 and 4.

Task 1. Systematic quantification of Hg concentrations associated with vegetation, litter, and organic soil horizons and carbon pools.

A main focus of the first 2 years of the study was in developing a systematic database on current Hg stocks associated with terrestrial C pools across U.S. forest ecosystems. This includes vegetation and soil sampling in forested ecosystems across the Western United States and analyzing samples for Hg mass concentrations, Hg/C ratios, N concentrations, and texture (for soils). This activity addresses project goals (1) to quantify total Hg pools associated with terrestrial C pools; (2) to estimate atmospheric Hg inputs and sequestration through leaf litterfall and plant senescence; (3) to evaluate total Hg pools sequestered across forests of the United States; and (4) to determine resilience and turnover times of Hg sequestered in vegetation and soil pools. Specific activities include the following:

  1. After development of sampling and analytical protocols in year 1 and implementation of a new total mercury analyzer, we have now finished field sampling of 10 sites across U.S. forests. The sites which have been sampled are: Jeffrey Pine forest in Nevada, Jeffrey Pine forest in California, Blue Oak savanna in California, Douglas Fir forest in Washington, Red Alder forest in Washington, Spruce/Hemlock forest in Maine, mixed Maple/Beech forest in New Hampshire, mixed deciduous forest in Tennessee, and a Slash/Longleaf Pine forest in Florida.
  2. We We have finished sample analysis, including drying, grounding, and analyzing for Hg, C, N, soil texture (soils only), for a total of eight sites, with two more sites being currently processed for analyses. In these eight sites, we also have used available biomass and C pool inventories of the respective sites in cooperation with local site investigators to scale up measured Hg, C, and N concentrations to scale up stocks to the total ecosystem levels.
  3. Our Our results show total ecosystem Hg storage and sequestration ranging between from 40 g ha-1 (site in Nevada) up to 375 g ha-1 (site in Maine), indicating highly variable accumulation and sequestration pools of Hg.  In all sites, most of the Hg was stored in the soils (> 95%) with the remaining stocks mainly associated with the litter horizon, tree foliage, and tree bark. The presence of C and N most strongly controls Hg accumulation and sequestration in U.S. forest ecosystems, for example, explaining a full 63% of the variability in soil Hg levels across all sites and across different soil depth. In addition, there are some relationships of Hg concentrations and pools to climatic variables (see Task 3).  These results point to a pronounced and dominant role of ecosystem processes—including C cycling—in determining ecosystem Hg accumulation. Specifically, plant-derived inputs of atmospheric Hg may be the dominant factor for Hg deposition in forested ecosystem.

Task 2: Assessing the fate of Hg during C mineralization processes using controlled laboratory incubation and field studies.

Because terrestrial carbon pools are highly dynamic and subject to both oxidative and anaerobic mineralization processes, we aim to quantify fate processes of Hg associated with C pools during decomposition processes. In the first 2 years of this study, we have been using multiple approaches to study fate of Hg during C mineralization processes, including (i) stochiometric comparisons of Hg and C pools in variously decomposed substrates (i.e., green leaves, variously decomposed litter, and different soil horizons) to infer about losses and/or accumulation of Hg during decomposition; (ii) controlled laboratory flux studies to synchronously measure Hg degassing and soil respiration rates from soils; (iii) long-term (> 1 year) controlled laboratory incubation studies of litter samples to assess C and Hg losses during mineralization using a mass balance approach; and (iv) continuous, in situ measurement of soil air Hg and CO2 concentration depth profiles to infer linkages between CO2 and Hg0 production in the soil.

  1. Results from stochiometric comparisons of Hg concentrations and Hg/C ratios show a consistent increase in concentrations and ratios in the following order: Green Needles/Leaves < Dry Needles/Leaves < Oi litter < Oe litter < Oa litter.  Stochiometric relations show negative correlations between Hg and C (r2 = 0.58) and N and C (r2 = 0.64) in decomposing litter, but a positive correlation between litter Hg and N (r2 = 0.70).  These inverse relations indicate a possible preferential retention of Hg over C during decomposition—similar to known patterns of N—indicating that that a majority of Hg may be retained in soils and organic layers during decomposition processes.
  2. Results from controlled laboratory flux measurements in terrestrial soils show a statistically significant linear correlation (r2 = 0.49)  between Hg and CO2 soil emissions, an effect unlikely caused by temperature, radiation, different Hg contents, nor soil moisture. Stochiometric comparisons of Hg/C ratios in emissions and in the underlying soil substrates suggest that only about 3% of Hg sequestered in the soil was subject to atmospheric evasion losses. Experimental manipulations were conducted to induce changes in CO2 respiration rates and observe Hg flux response, including inducement of anaerobic conditions by changing air supply to soil samples from a mixture of N2/O2 (80% and 20%, respectively) to pure N2. Unexpectedly, Hg emission fluxes from soil samples almost quadrupled after O2 deprivation while oxidative mineralization (i.e., CO2 emissions) was greatly reduced, a response which was lacking when the same experiment was repeated with sterilized soils. This study indicates that only small fractions of Hg sequestered in organic pools may be subject to atmospheric evasion losses to the atmosphere during C mineralization. Nontheless, these natural emissions may be a very significant source of Hg to the atmosphere on a global scale. This study further provides first experimental evidence that Hg volatilization and possibly Hg2+ reduction to the volatile elemental Hg in is related to O2 availability in the terrestrial soils. Both aerobic and anaerobic Hg emissions may involve direct microbial reduction of Hg2+ to Hg0 in soils, or be caused by indirect effects such as reducing byproducts, soil chemistry, or redox potentials.
  3. Ongoing long-term litter incubation studies from litter collected at four field sites have been started in Fall 2008. Litter samples are allowed to decompose under controlled laboratory conditions (dark, constant temperature) in 1-liter glass jars covered with 0.2 um filters to allow gaseous air exchange (e.g., CO2, Hg0) while minimizing additional atmospheric Hg deposition processes.  Initial results from sequential “harvesting” of litter jars after 3 months and 6 months indicate a loss of litter and C mass of ~ 10% in the first half year of the incubation.  No Hg losses were visible after 3 months, but slight decrease—albeit not yet statistically significant—in Hg stocks (1-2 %) started to show after six month. These results indicate minor—if any—Hg losses to the atmosphere during decomposition. Test of Hg solubility show decreasing Hg solubility in Hg with continued decomposition stage also indicating that runoff losses as soluble Hg may be small during decomposition. The study will continued for a total of 18 months to further evaluate these patterns.
  4. An automated measurement system was developed to measure in situ concentration gradients of Hg0 and CO2 in soil air at multiple depths and locations.  Measurement system was initially deployed and tested in the field in summer 2009. First results are currently being analyzed.

Task 3 and Task 4: Statistically evaluation of pools and fluxes in respect to climatic conditions and other variables, and integration of data into a terrestrial C model.

We started to screen experimental results obtained in Task 1 in regards to climatic variables and other factors. Statistical tests show that latitude and annual precipitation were related to observed pools sizes and concentrations across the current eight sites analyzed, but that Hg levels were mainly unrelated soil texture, longitude, nor temperature. These variables will be used in the future to scale up observed Hg stocks to the level of the United States after all sites have been sampled and analyzed.  One of the strongest associations was an observed between Hg and C and N levels, which in soils C and N levels explained a large part of the variability (r2 = 0.63) of observed Hg levels. Reasons for this effect include (1) high sorption of Hg inputs (e.g., from atmospheric deposition) to organic matter pools; (2) a major role of plant-derived atmospheric Hg inputs (e.g., leaf litterfall, plant senescence) in forest ecosystems which lead to corresponding C, N, and Hg pools based on ecosystem productivity of the sites; and (3) link between biogeochemical cycling of Hg and C with possible losses of C during decomposition processes accompanied by corresponding losses of Hg (e.g., via degassing, mobilization, or runoff).

These experimentally observed relationships and patterns will be further explored using mathematical and process-based modelling. Initial work on this front includes inclusion of mercury-carbon ratio introduced to the Terrestrial ECOsystem model by Prof. Luo’s group. First dynamics of mercury in the ecosystem was obtained according to the proportional dynamics of carbon, and first ttempts were made to separate carbon and mercury dynamics in order to obtain mercury carbon ratios through inverse modeling.

Future Activities:

The activities of the third year will focus on continued sample collection in additional field sites and finishing sample analysis for Hg, C, N, and soil texture. We intend to systematically quantify whole-ecosystem Hg stocks of a total of 14 to 15 sites (i.e., additional 4-5 sites). Further experimental emphasis will be placed on continued fate quantification of Hg sequestered in litter and soils by continuing experiments that have started in year 2 of the study. Year 3 will mark the start of scaling component to scale up mercury pools, deposition loads, and fate processes to the level of U.S. forests and begin modeling of modeling component.

Journal Articles on this Report : 1 Displayed | Download in RIS Format

Other project views: All 13 publications 4 publications in selected types All 4 journal articles
Type Citation Project Document Sources
Journal Article Rastogi N, Oakes MM, Schauer JJ, Shafer MM, Majestic BJ, Weber RJ. New technique for online measurement of water-soluble Fe(II) in atmospheric aerosols. Environmental Science & Technology 2009;43(7):2425-2430. R832835 (2008)
R832835 (Final)
R833378 (2008)
  • Abstract from PubMed
  • Full-text: ResearchGate-Abstract & Full Text PDF
  • Abstract: ACS-Abstract
  • Other: ACS-Full Text PDF
  • Supplemental Keywords:

    RFA, Scientific Discipline, Air, particulate matter, air toxics, Environmental Chemistry, Environmental Monitoring, Atmospheric Sciences, ambient air quality, real time measurement, particulate organic carbon, atmospheric particulate matter, inorganic aerosols, ambient measurement methods, airborne particulate matter, Speciation Trends Sampler, air sampling, airborne aerosols, aerosol composition, PM, trace metals, particulate organic matter, particulate matter components, chemical speciation sampling

    Relevant Websites:

    Daniel Obrist exit EPA

    Progress and Final Reports:

    Original Abstract
  • 2006 Progress Report
  • 2007 Progress Report
  • Final Report