Just how big is that piece of the puzzle?
Quantifying the flux of mercury in forest litter at Acadia National Park

Kit Sheehan

Project Status as of 5/20/2003

Abstract

Our limited understanding of mercury (Hg) cycling presents us with paradoxes in nature for Hg exposure and risk. In some natural ecosystems in Maine, Hg accumulates in one catchment yet we can find adjacent catchments with similar exposure but low Hg content. Maine has a statewide fish consumption advisory, in part, because researchers can not accurately predict which lakes have high concentrations of Hg in fish. Recent research by Johnson (2002) has quantified wet-only precipitation, throughfall, and stream water fluxes of Hg by monitoring two watersheds within Acadia National Park, where Hg is considered to have adverse ecological effects. Exploratory measurements have identified litterfall as a large transport vector of Hg to the forest floor, compared to more frequently monitored vectors such as precipitation or throughfall. A study has been initiated to better define this Hg flux in ecosystems. Preliminary results, from September to December 2002, confirm previous findings that reported relatively high fluxes of Hg in litterfall with mean total Hg litter fluxes (± SE) of 2.08 μg m^-2 mo^-1 (± 0.28) in Hadlock Brook watershed and 2.56 μg m^-2 mo^-1 (± 0.42) in Cadillac Brook watershed.   Although there is a weak relationship (P< 0.05) between mean total Hg litter flux and elevation and no relationship between mean total Hg litter flux and aspect, results indicate that vegetation classes exert important controls. The mean total Hg litter flux at study sites predominated by scrub vegetation was significantly lower (P< 0.05) than sites characterized by softwoods, hardwoods, or softwood/hardwood mix. The mean total Hg concentration in the litter was higher (P< 0.05) at softwood (0.076 μg g^-1± 0.008) sites than either hardwood (0.042 μg g^-1± 0.001) or scrub (0.043 μg g^-1± 0.010) sites. However, rates of litter deposition for the study period caused similar mean Hg litter fluxes between softwood and hardwood sites.

Objectives

• Quantify the annual flux of total Hg in forest litterfall in two study watersheds, Acadia National Park, ME, USA

• Identify the landscape controls on the flux of Hg in forest litterfall

• Develop a predictive model for the flux of total Hg in litterfall in Acadia National Park (ANP)

• Determine the species-specific leaching potential of Hg in litterfall

Data Collection and Analysis

• Collected litterfall monthly in acid washed 0.15m2 polyethylene tubs

• 20 collectors in Cadillac Brook watershed (~32 hectares)

• 19 collectors in Hadlock Brook watershed (~47 hectares)

• Oven dried litter samples at 25°C

• Litter samples digested in a Mars 5

• Samples analyzed for total Hg on Hewlett-Packard Flow Injection Mercury System model 400 (FIMS) with a model AS90 auto-sampler at ECL

• Watersheds aerially divided into four vegetation types

  • S: vegetation predominantly of low stature; scrub

  • SW: vegetative cover dominated by softwood species

  • HW: vegetative cover dominated by softwood species

  • W/HW: vegetative cover composed of a mix of hardwood and softwood species

Results

Table 1. Total Hg flux in major input vectors in the two study watersheds in Acadia National Park

¹ Johnson 2002

² National Atmospheric Deposition Program; Mercury Deposition Network, McFarland Hill, ME

• The monthly total Hg litter input in these watersheds is of similar magnitude as the annual input of more routinely monitored vectors, such as throughfall and wet only precipitation (Table 1). These results are consistent with other studies of total Hg annual inputs, which generally show litterfall > throughfall > open precipitation (see Grigal, 2002).

• Mass of litterfall (± SE) was greater (p<0.05) at HW sites (15.69g ± 2.58) and SW/HW sites (9.78g ± 1.30) than at SW sites (5.53g ± 0.89) or S sites (4.84g ± 1.23) (Figure 1a). 

• Total Hg concentration (± SE) was greater (p<0.05) at softwood sites (0.076 μg g-1 ± 0.008) than at hardwood (0.042 μg g-1 ± 0.001) or scrub sites (0.043 μg g-1 ± 0.010) (Figure 1b).

• Monthly total Hg flux in litterfall (± SE) was smaller (p<0.05) at S sites (0.80 μ g m-2 ± 0.17) than at HW sites (3.20 μ g m-2 ± 0.73), SW sites (2.07 μ g m-2 ± 0.35) and SW/HW sites (2.91 μ g m-2 ± 0.29) (Figure 1c).

Future Directions

• Estimate annual total Hg flux in litter based upon year-round litter collection

• Investigate seasonal variation in litterfall

• Evaluate the influence of landscape characteristics, such as elevation, aspect, canopy height, and percent canopy cover, on the flux of Hg in litter in ANP

• Utilize identified landscape controls to develop a predictive model for Hg litter deposition in ANP

• Determine if litter C:N ratio is related to Hg leaching potential through laboratory leaching studies on litter, representing the spectrum of cover types in ANP

• Refine the Hg budgets previously developed by Johnson (2002), in the study watersheds

References

Grigal D. F. (2002)  Inputs and outputs of mercury from terrestrial watersheds: a review. Environ. Rev. 10, 1-39.

Johnson K. B. (2002) Fire and its effects on mercury and methylmercury dynamics for two watersheds in Acadia National Park, Maine. M. S. Thesis, University of Maine, Orono, ME.

Mercury Deposition Network. 2001. Maine Mercury Concentration Average.