+ All Categories
Home > Documents > Physical and chemical factors controlling mercury and methylmercury concentrations in stream water...

Physical and chemical factors controlling mercury and methylmercury concentrations in stream water...

Date post: 13-Dec-2015
Category:
Upload: michael-luke-lloyd
View: 215 times
Download: 2 times
Share this document with a friend
Popular Tags:
25
Physical and chemical factors controlling mercury and methylmercury concentrations in stream water Mark E. Brigham and Dennis A. Wentz 5 th National Monitoring Conference San José, California May 7-11, 2006 U.S. Department of the Interior U.S. Geological Survey
Transcript

Physical and chemical factors controlling mercury and methylmercury concentrations in stream water

Mark E. Brigham and Dennis A. Wentz

5th National Monitoring ConferenceSan José, CaliforniaMay 7-11, 2006

U.S. Department of the InteriorU.S. Geological Survey

Willamette Basin

Georgia-Florida Coastal Plain

Western Lake Michigan

Drainages

Reference stream

Urban stream

USGS NAWQA mercury study areas

Aqueous methylmercury (MeHg) is a major control on mercury bioaccumulation.

Mean Hg in

forage fish

(μg/g wet wt.)

N ≈ 24 at each site

(2 species x 12 individuals)

R2 = 0.8418

0.00

0.05

0.10

0.15

0.20

0.0 0.1 0.2 0.3 0.4 0.5

Mean aqueous MeHg (ng/L)

N ≈ 35 at each site

What controls aqueous MeHg (and THg) concentrations in

streams?

• Weight-of-evidence approach to assess: – Atmospheric inputs– Watershed processes (methylation

and subsequent delivery to stream)– Methylation in channel sediments

Simplified mass balance

Watershed soils: storage / runoff

methylationdemethylation

fluvial transport

Wet deposition

Channel sediments: storage / resuspension

methylation demethylation

Evasion (Hg°)Dry

deposition

resuspension

Wet Hg & MeHg deposition: Mercury Deposition Network (MDN) sites

Load:

∑ (weekly [Hg] x precip volume),

expressed as μg/m2/yr

Hg in precipitation Popple River, WI site (WI09—Mercury Deposition Network)

Oct

‘02

Jan

’03

Jan

’04

Jan

’05

Methylmercury (MeHg) and total mercury (THg) in stream water

• ~35 samples per site from 2003-05

• Key measure of food-web exposure

• Key component of mass balance

Mercury in stream water: sample processing

0.7 μm QFF

Whole waterMeHgTHg

ParticulatePMeHgPTHg

FilteredFMeHgFTHg

===

+++

Fluvial mercury loads & yields

Fluvial load: • Regress load vs. flow for sampled dates.• Predict to unsampled dates using daily

flows Reference: Runkel et al., 2004, USGS Techniques &

Methods, Book 4, Ch. A5; LOADEST S-Plus program by D. Lorenz, USGS

• Yield = load / watershed area, μg/m2/yr• Examine yield as % of wet depositional

loads to ecosystem…

MeHg deposition unrelated to MeHg yield

0

100

200

300

400

500

600

700

800

900

1000

OR-Urb

OR-Ref

-L

WI-R

ef-H

WI-R

ef-L

WI-U

rb

FL-Ref

-H

FL-Ref

-L

FL-UrbF

luvi

al y

ield

as

% o

f w

et d

ep l

oad

200

3-04

*

THg yield: 4.4–48% of wet deposition

MeHg yield: 22–926 % of wet deposition (excludes site where MeHg < MDL*)

*

Florida

Wisconsin

Oregon

THg yield vs precip Hg deposition, 2003-2004

1:10 line

WI09Pike

OR10Lookout WI22

Oak

FL32LWekiva

GA09 StMary

FL05Santa Fe

OR01Beaverton

WI32Evergreen

0

1

2

3

4

5

6

7

0 5 10 15 20Precip THg load, micrograms/m2/yr, 2004

Flu

via

l T

Hg

yie

ld, u

g/m

2/y

rF

luvi

al T

Hg

yiel

d, μ

g/m

2 /yr

Wet THg deposition, μg/m2/yr, 2003-04

Urban

Reference

Summary of partial mass balance

• Wet MeHg deposition could account for MeHg in most streams– low [MeHg] streams.

• Caveat—Missing key components of mass balance– watershed retention– demethylation– dry deposition

• Must invoke watershed methylation to explain high [MeHg] streams.

Aqueous total Hg and methylmercury correlate strongly to dissolved organic carbon (DOC):

• among all sites (shown here)• within a site (most sites)

Log 1

0 [

FM

eHg]

(ng

/L)

Log 1

0 [

FT

Hg]

(ng

/L)

Log10 [DOC] (mg/L) Log10 [DOC] (mg/L)Log10 [DOC] (mg/L)

Runoff-mobilized Hg-DOC complexes controls: -- THg in most streams -- MeHg in half the study streams.

Evidence for watershed inputs of MeHg

Evidence against in-channel methylation as dominant source

Santa Fe River, Florida

Log10 [Q] (cfs)

Log 1

0 [

FT

Hg]

(ng

/L)

Log 1

0 [

FM

eHg]

(ng

/L)

Negative relation between MeHg and flow?

Evidence for in-channel methylation?

Or, high [MeHg] in wetlands during low-flow periods?

St Mary’s River, Florida

Log10 [Q] (cfs)

Log 1

0 [

FT

Hg]

(ng

/L)

Log 1

0 [

FM

eHg]

(ng

/L)

Aqueous methylmercury strongly linked to wetland density (mean methylmercury; all study sites)

R2 = 0.9224

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

0.45

0 10 20 30 40 50 60

Wetland density (% of total land cover in basin)

FMH

g (

ng/L

)

Log 1

0 T

Hg

conc

entr

atio

n (n

g/L)

DOC and Suspended Sediment—a potential screening tool for total mercury…

R2=0.62

Log10 DOC (mg/L)

Log10 Susp Sed

(mg/L)

…and methylmercury.Lo

g 10 M

eHg

conc

entr

atio

n (n

g/L)

Log10 DOC (mg/L)

Log10 Susp Sed

(mg/L)

What have we learned about THg and MeHg in streams?

Summary

Precipitation and watershed influences

• Precipitation inputs– main source of THg to ecosystem– Could account for all MeHg in

some streams• Watershed inputs

– major vector for MeHg and THg delivery to streams, particularly in wetland-rich basins

Summary

Concentration relationships

• DOC and suspended sediment– Control THg & MeHg in streams

(MeHg picture is noisier)– key explanatory variables– perhaps a useful screening tool– Erosion control—useful to reduce

particulate Hg, and hence THg

Summary

Role of channel sediments

• MeHg source? – At most, a minor source of MeHg to

stream water– Low MeHg at low flow (evidence

against substantial inputs from sediments)…

– …except at one site (either sediment methylation or seasonally high MeHg from wetlands)

• MeHg sink? – Fast demethylation rates in sand, a

dominant substrate in some streams

Implications for monitoring THg & MeHg in streams

• Sample size (N)—depends on objectives…– BAF’s: Perhaps as few as N ≈ 6, well

spaced seasonally (see: Paller and others, 2004, Archives of Environ. Contam. & Toxicology)

– Concentration relationships & fluvial loads: N ≥ 35, well spaced seasonally and hydrologically

Acknowledgements

USGS: Dennis Wentz, Barb Scudder, Lia Chasar, Amanda Bell, Michelle Lutz, Dave Krabbenhoft, Mark Marvin-DiPasquale, George Aiken, Robin Stewart, Carol Kendall, Bill Orem, Rod DeWeese, Jeff Isely, and many others…

USGS: NAWQA and several other USGS programs

MDN site support: USGS, Wisconsin DNR, Oregen DEQ, Forest Service, US Fish & Wildlife Service, St. John’s River Water Management District (FL)

Menomonie Indian Tribe of Wisconsin


Recommended