Mercury and Organic Carbon Dynamics in an Upland Landscape during Snowmelt Jamie Shanley 1 Paul Schuster 2 Mike Reddy 2 Dave Roth 2 Howard Taylor 2 George.

Mercury and Organic Carbon Dynamics in an Upland Landscape during Snowmelt

Jamie Shanley1

Paul Schuster2

Mike Reddy2

Dave Roth2

Howard Taylor2

George Aiken2

1U.S. Geological Survey Montpelier, Vermont

2 U.S. Geological Survey, Boulder, Colorado

Objectives

Develop and test methodologies for intensive spatial and temporal event sampling of Hg

Investigate the dynamics of Hg transport during snowmelt across a gradient of land cover and watershed size

Evaluate linkages between dissolved and particulate Hg and dissolved and particulate carbon

Previous Hg research in Vermont Wet and dry deposition measurements in Underhill since 1993 - world’s longest

continuous record. High Hg in leaffall (Rae, 1998) corroborates high Hg in dry deposition (up to

5x wet) (Scherbatskoy et al., 1998) Hg deposition retained (>90%) in terrestrial watershed; export episodic,

associated with POC (Scherbatskoy et al., 1998; Shanley et al,. 1999) High Hg in O-horizon soil solution, associated w/DOC (Donlon, 1999) Hg export in agricultural landscape similar to forest (Rinehart, 2000) Routine measurements by State since 1970s show high Hg in fish Current focused study of sediment, water and fish Hg in VT lakes (Kamman et al.,

1998; 2000)

Sleepers River Research Watershed

USDA Agricultural Research Service 1957-1975 Established 17 stream gages, 13 meteorological stations

National Weather Service 1968-1986 Snowmelt modeling - energy balance approach

Cold Regions Lab (CRREL) 1979-present Snowmelt modeling, sensor testing

US Geological Survey 1991-present Water, Energy, and Biogeochemical Budgets (WEBB)

Sleepers River Research WatershedBasin characteristics

Elevation: 200 to 690 m, slopes gentle to very steep Precipitation: 1100 mm per year, ~30% as snow Climate: Humid continental, Mean annual T: 6°C Gaged watersheds from 3.4 ha to 111 km2

Land cover - overall 2/3 forested, 1/3 open land Mixed / Northern hardwoods (Sugar maple, beech, ash, spruce/fir) Dairy farming (pasture, hayfields, corn fields)

Economics: Dairy farming, maple sugaring, logging

Sleepers River Research Watershed

Basin characteristics

Bedrock: quartz-mica phyllite with beds of calcareous granulite

Surficial: 1-3 m of dense basal till from local bedrock Soils: Inceptisols, spodosols, and histosols w/wetlands Stream chemistry: well-buffered calcium bicarbonate /

sulfate waters Hydrology: Flashy streams, snowmelt dominated

Study watersheds (with % of area forested)

W-5 11,125 ha (67%)

W-3 837 ha (80%) W-2 59 ha (27%) W-9 41 ha (100%) W-9A 17 ha (100%)

W-9B 13 ha (100%) W-9C 7 ha (100%) W-9BX 5.1 ha (100%) W-9BY 3.4 ha (100%) W-9BXA ~2 ha (100%)

.

W-9

W-2

W-3

W-5

N

Stream gaging station

Meteorological and snow survey station Stream

Catchment boundary

St.Johnsbury

Danville

N. Danville

PassumpsicRiver

Sleepers River

FieldLaboratory

Montpelier

VERMONT

44°30'

44°25'

72°05'72°10'

2 4 6 8 Kilometers0

Burlington

450

60

0

Elevation contour (m)

300

W-9 watershed

.

100 m0

Contour interval 100 ft

Hg sampling site

Wetland

W-9BXA

W-9

W-9BY

W-9C

W-9BX

W-9BW-9A

Field procedures Streamwater

Grab sample in teflon holding bottle for Hg - collected wearing particle-free gloves.

Grab sample in pre-baked amber glass bottle for C fractions. Grab sample in PE bottle for major ions Field sonde for Temp., pH, DO, Sp. Cond., ORP

Snow Lexan scoop into PE bag while wearing particle-free gloves; 7

profiles at peak snowpack

Groundwater and seeps Peristaltic pump with teflon tubing - 6 samples near peak melt

On-site sample processingField laboratory - within 4 hours - complete duplicate processing of

all samples Mercury

Dissolved:Vacuum filtered through 0.4-µm cellulose acetate membrane into acid-washed 125-mL glass bottle. Acidified with K2Cr2O7.

Particulate: Filter membrane placed in 60-mL glass bottle, filled with DI water, acidified with K2Cr2O7

Carbon DOC: Vacuum filtered through 0.7-µm GFF (glass fiber) filter into pre-baked amber

bottles POC: I set of duplicate filter membranes wrapped in Al foil for solid C analysis;

additional set saved for LOI determination.

Major ions Cations: syringe-filtered through 0.4-µm cellulose acetate membrane into acid-washed

125-mL PE bottle. Acidified to pH 2 with HNO3

Anions: syringe-filtered through 0.4-µm cellulose acetate membrane into clean 125-mL PE bottle. Kept chilled.

Laboratory AnalysisAll analyses except POC, PON at USGS, Boulder, CO Mercury

Dissolved:Cold vapor atomic fluorescence spectrometry (CVAFS) on filtered aliquot. Particulate: CVAFS on acidified solution containing filter membrane .

Carbon DOC (and TOC): Ultraviolet persulfate oxidation with infrared detection POC (and PON) at Chesapeake Biological Lab, University of Maryland Loss-on-ignition (LOI) to estimate organic C by combustion at 550 C.. SUVA - Absorbance at 254 µm

Major ions Cations: ICP Anions: Ion chromatography Alkalinity: Gran titration.

Concentrations, ng/L Duplicates median min max mean % diff.

Hg, dissolved 0.9 <0.3 4.1 5.8Hg, particulate 2.3 0.6 16.3 12

Hg concentration ranges and results of duplicate analyses

Based on 66 duplicate analyses

Hg vs. alkalinity, all sites, Sleepers River snowmelt 2000

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0

Hg,

dis

solv

ed,

ng/L

Alkalinity, meq/L

r2 = 0.43, p<.0001

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

0.0 0.2 0.4 0.6 0.8 1.0 1.2

DO

C,

mg/

L

Discharge, mm/hr

r2 = 0.41, p<.0001

DOC vs. discharge, all sites, Sleepers River snowmelt 2000

0

2

4

6

8

10

12

14

16

0 5 10 15 20 25 30

Hg,

par

ticul

ate,

ng/

L

POC, mg/L

r2 = 0.84, p<.0001

Particulate Hg vs. POC, all sites, Sleepers River snowmelt 2000

Hg vs. discharge, all sites, Sleepers River snowmelt 2000

0.0

0.5

1.0

1.5

2.0

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4

Hg,

dis

solv

ed,

ng/L

Discharge, mm/hr

r2 = 0.46, p<.0001

0.0

0.5

1.0

1.5

2.0

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0

Hg,

dis

solv

ed,

ng/L

DOC, mg/L

r2 = 0.50, p<.0001

Dissolved Hg vs. DOC, all sites, Sleepers River snowmelt 2000

Specific discharge and DOC at W-9B (forested, 13 ha)

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

0.0

0.5

1.0

1.5

2.0

2.5

3.0

80 85 90 95 100 105 110 115

DO

C,

mg/

L

Discharge, C

FS

Day of year, 2000

0.0

1.0

2.0

3.0

4.0

5.0

0.0

0.5

1.0

1.5

2.0

2.5

3.0

80 85 90 95 100 105 110 115

Hg,

ng/

L

Discharge, C

FS

Day of year, 2000

12.3

Particulate

Dissolved

Dissolved and particulate Hg at W-9B

0.0

1.0

2.0

3.0

4.0

5.0

0.0

0.5

1.0

1.5

2.0

2.5

3.0

80 85 90 95 100 105 110 115

Hg,

dis

solv

ed,

ng/L

,D

OC

, m

g/L

or

Discharge, C

FS

Day of year, 2000

Dissolved Hg and DOC at W-9B

0.0

1.0

2.0

3.0

4.0

5.0

0.0

0.5

1.0

1.5

2.0

2.5

3.0

80 85 90 95 100 105 110 115

Hg,

dis

solv

ed,

ng/L

,D

OC

, m

g/L

or

Discharge, C

FS

Day of year, 2000

Dissolved Hg and DOC at W-9B

0

1

2

3

4

5

6

7

60 70 80 90 100 110 120

DO

C/H

gD x

10

6

Day of year, 2000

W-9B

All other sites

DOC vs. Hg flushing, all sites

Conclusions We developed a methodology for accurate and reproducible

sampling of Hg at high temporal frequency using ultraclean sample handling protocols

Dissolved and particulate Hg concentrations closely tracked the respective dissolved and particulate C fractions, despite 6 order-of-magnitude difference in concentrations

Hg and C may be biogeochemically linked, or they simply have a common source -- shallow soil -- that is flushed during snowmelt. Needs further research.