Input-Output Budgets of Major Ions, Trace Elements and Mercury for a Forested Watershed in Western Maryland

Program Description

Wet deposition of trace elements is a function of precipitation volume and trace element concentrations. Precipitation volume varies across the state. The spatial differences in trace metal wet deposition fluxes observed in CBADS indicate that the present coastal region data sets may not be appropriate for interior regions of the watershed, where rain fall trace element concentrations and amounts may vary with topography and proximity to source regions in the mid-west. The goal of this portion of the study was to increases our understanding of the retention and movement of major ions and trace elements in forested watersheds. To accomplish this goal, atmospheric inputs (precipitation and throughfall) and stream water export of major ions, trace metals and mercury from a completely forested watershed in western Maryland was measured from 1996 through May 1997.

The three primary goals of this study were:

• Determine spatial variability in major ions and metals deposition across the watershed, using existing data for the Piedmont Region and Coastal Plain available from the literature, and new data collected during the field portion of the current study;
• Estimate total wet deposition of major ions and select trace elements to the Bay and its tributary regions using data sets developed for each region; and
  
• Determine the relationship between precipitation inputs to a forested watershed in western Maryland and annual stream water export of these same trace elements.

Precipitation/Air Sampling and Analysis

Wet deposition samples were collected at Piney Dam, elevation 780 m, which is located in northeastern Garrett County, Maryland. Throughfall and stream water samples were collected in a 255-hectare watershed of an unnamed tributary to Herrington Creek located in southwestern Garrett County, near Oakland, Maryland. This site is approximately 55 kilometers southeast of Piney Dam.

Wet deposition samples were collected using an automated wet-only MIC-B collector (MIC Corporation, Richmond Hill, Ontario, Canada). The commercially available MIC-B was modified to sample precipitation simultaneously for major ions, selected trace metals, and mercury using suitable sampling equipment. This modification has been rigorously tested and shown to have no artifacts associated with the collection of major ions, mercury, and trace metals. Precipitation was collected on a daily basis from June 1, 1996 to June 4, 1997 using ultra-clean sampling techniques. Throughfall samples for major ions, trace metals, and mercury were collected at three sites in the Herrington Creek watershed. During the snow-free season (April- October), throughfall samples for major ions were collected using 20-cm funnels attached to one liter amber colored HDPE bottles. Funnels were mounted one meter above the forest floor on permanent posts.

Throughfall samples for mercury were collected using 10-cm diameter polycarbonate funnels attached to one liter FEP Teflon bottles that were supported by one-meter tall cylindrical PVC tubes. During the snow season, (November-April), throughfall samples for major ions and mercury were collected in rigorously cleaned HDPE bags (10 mil thickness) supported by 50-cm diameter plastic cans that were attached to the 1 meter tall permanent posts. Trace metal samples were collected in a 30.5-cm diameter HDPE bucket, which was nestled in a second bucket that was attached to the top of a cedar post at an approximate height of one meter. To compare these two types of throughfall collectors, both were deployed simultaneously, side-by-side, for three weeks at the start of the snow season. Throughfall samples were collected weekly on Tuesday mornings from May 28, 1996 to June 4, 1997.

Stream water samples were collected on a weekly basis coinciding with throughfall sample collection. Grab samples of stream water for major ion analysis were collected using standard EPA protocols. Stream water was collected using manual grab samples for trace element analysis. Storm-flow sampling for trace elements was carried out using an ISCO Model 3700 automated sequential sampler (ISCO Environmental Division, Lincoln, NE), that was manually started at the onset of each major runoff event along with another sampler for major ions.

To estimate wet deposition rates to Herrington Creek Watershed, volume-weighted concentrations from Piney Dam were multiplied by the precipitation volume measured at Herrington Creek. This approach accounted for the regional variations in precipitation amounts between the two study sites but it assumes that the concentrations at Piney Dam were similar to those reported for other sites in t he region. To estimate throughfall deposition to Herrington Creek, volume-weighted average throughfall concentrations were multiplied by the average weekly throughfall amounts measured at Herrington Creek. These weekly rates were summed over seasons and year to give seasonal and annual throughfall deposition rates, respectively. Stream water export of major ions was calculated on a weekly basis and summed over the year to estimate the annual watershed export rates. Weekly major ion export was the product of the average major ion concentrations for two consecutive sampling periods and the average to the continuous discharge divided by the watershed area.

Major ion concentrations in water samples were analyzed according to EPA Laboratory methods appropriate for monitoring surface water quality in acid deposition studies. Closed (for stream water) and open pH (for wet deposition and throughfall) were measured with an Orion (Model 611) pH meter. Stream water ANC was measured using the acidimetric Gran titration technique with electrometic pH detection. Specific conductance was measured with a pipette cell (Mode G01, Rosemount Analytical) and a YSI (Yellow Springs Incorporated) Model 32 meter with temperature compensation to 25^o C. All major ions, except Ca⁺² and Mg⁺² in precipitation and throughfall, were measured using a Dionex DX-500 ion chromatograph, equipped with electronic conductivity suppressor. Calcium and Mg⁺² were analyzed by ion-coupled plasma atomic emission spectrophotometry. Atmospheric deposition (wet only and throughfall) and stream samples were analyzed for Al, Cd, Cu, Cr, Fe, Mn, Ni, Pb, and Zn using a Perkin and Elmer 3300 AA Spectrophotometer, equipped with a 600HGA graphite furnace (GFAAS). The metalloids arsenic and selenium were analyzed using selective hydride generation, preconcentration by cryogenic trapping and fractional volatilization. A hydrogen/air flame was then employed to reduce the As and Se hydrides to their atomic state, which was then quantified using atomic absorption spectrometry. Mercury samples were analyzed in a cleanroom for total and speciated mercury using standard methods.

Quality Assurance (QA)

Quality control and quality assurance analysis was performed using blank and replicate samples, and blind duplicate samples. In addition, ion and conductivity balances were calculated and checked (measured vs. calculated) for all water samples. Essential elements of the QA program for trace elements included:

• rigorous application of "ultra-clean" sampling and handling techniques,
  
• conducting and evaluating procedural blanks on a regular basis,
  
• use of externally certified analytical reference solutions,
• collecting and independently analyzing multiple sample splits for wet-only precipitation, throughfall, and stream intensives,
  
• for canopy throughfall, side-by-side sampling using different collector types, during selected overlap periods, and
  
• conducting replicate analysis using redundant analytical techniques, when possible.

Study Results

Wet deposition was an important source of major ions, trace elements, and mercury (Go to Program Data). Among the major ions, H⁺, SO₄^-2, NO₃^-, and NH₄⁺ had the highest annual wet deposition rates, which are similar to those reported for other high deposition sites in the northeastern United States. For trace metals, Al, Fe, and Zn had the highest wet deposition rates; Cr, Cu, Mn, Ni, Pb and Se had intermediate rates; and Cd and As had the lowest rates. Annual wet deposition of total mercury was 14.9 ug/m²/yr and wet deposition of methylmercury was less than 1% of the total mercury deposition.

The forest canopy had a major effect on most major ions (K⁺, Ca⁺², Mg⁺², SO₄^-2, and NO₃^-) and one trace metal (Mn); some effect on a few trace elements (Fe, Ni, Zn, Cd, Cr, Cu, Se); and little to no effect on a few major ions (Na⁺, NH₄⁺), trace elements (Al, As, Pb) and total mercury. On an annual basis, the forest canopy consumed 20% of the free acidity in incident precipitation, had no net effect on Na⁺ and NH₄⁺ deposition, and was a strong net source of K⁺, Ca⁺², Mg⁺², SO₄^-, NO₃^-, and Mn. The enhanced (1.5 to 60 times greater than wet deposition) throughfall of these ions was due to canopy exchange reactions and leaching of dry deposited gases and particles. For trace metals, the canopy was a small net source of Fe, Cu, Zn, and Se; throughfall deposition rates were 30 to 50% greater than wet deposition rates and were consistent with expected dry deposition rates. In contrast, the canopy was a net sink for Ni, Cd, and Cr. For mercury, annual throughfall deposition rates for total mercury were about 30% greater than wet deposition rates. It was suggested that this was due to the washoff of dry deposited material.

On an annual basis, Herrington Creek watershed retained essentially all of the throughfall inputs of H⁺ and NH₄⁺, about 35% of the throughfall inputs of K⁺ and NO₃^-, and was a net source of SO₄^-2, Cl^-, Ca⁺², Mg⁺², and Na⁺. Export of these ions was two to five times greater than the throughfall inputs. For trace metals, atmospheric inputs equal stream water outputs for Al and Mn, inputs are two to three times lower than outputs for Zn, Ni, and Cd, and the watershed is a net sink for Fe, As, Cu, Pb, Se, and Cr. From among the elements (Fe, As, Cu, Pb, Se, Cr), Pb, As and Se are most strongly retrained in the Herrington Creek watershed, representing a 50 to 90% retention of atmospheric input. For Fe, Cu, and Cr, only about 25% of the atmospheric input is retained Herrington Creek watershed. For mercury, about 80% of the atmospheric input was retained by the watershed.

Related Publications/Reports

Input-Output Budgets of Major Ions, Trace Elements and Mercury for a Forested Watershed in Western Maryland. July 2000. Mark Castro, Joseph Scudlark, and Robert Mason. PPAD-AD-1

For more information, e-mail Dr. John Sherwell at the PPRP, or call him at 410-260-8660

Go to I/O Budget of Major Ions, Trace Elements and Mercury Data