Climatic Impacts on Watershed Biogeochemistry:

Nitrogen, Sulfur, and Mercury Dynamics

Myron J. Mitchell

Wednesday, October  14, 2009

Some Major Climatic Effects in Northeastern U.S.A.

• Summer– Higher temperatures– More droughts– More extreme events

• Winter– Warmer temperatures– Shorter period of snowpack– More rain on snow events

Climatic Impacts on Watershed Biogeochemistry:

Nitrogen, Sulfur, and Mercury Dynamics

• Nitrogen (nitrate)‐‐Winter

• Sulfur (sulfate)‐‐Summer

• Mercury‐‐Summer

Climatic Impacts on Watershed Biogeochemistry:Nitrogen, Sulfur, and Mercury Dynamics

• Nitrogen (NO3‐)

– Importance of nitrate to chronic and episodic acidification

– Explanation for long‐term (interannual) changes in NO3

‐ concentrations in watersheds still not complete

– Nitrogen important limiting nutrient

– Understanding  of winter processes in affecting NO3

‐ dynamics incomplete

0

100

200

300

400

0

100

200

300

400

500

0

100

200

300

400

YEAR82-83

83-8484-85

85-8686-87

87-8888-89

89-9090-91

91-9292-93

0

100

200

300

400

BBWM

HBEF

HF

CS

GROWING

DORMANT

mol

NO

3- ha-1

For snow dominated watersheds >87% NO3

‐ lost during dormant season:

High discharge & microbial generation of nitrate dominates

Mitchell, M.J., C.T. Driscoll, P. Murdoch, G.E. Likens, J.S. Kahl and L. Pardo. 1996a. Climatic control of nitrate loss from forested watersheds in the northeast United States.Environmental Science and Technology 30: 2609‐2612.

NO3-

12/95 1/96

2/96 3/96

4/96 5/96

NO

3- con

cent

ratio

n ( μ

mol

L-1

)

0

10

20

30

40

50

60

70

Dai

ly a

vera

ge a

ir te

mpe

ratu

re (o C

)-30

-20

-10

0

10

20

Run

off (

mm

day

-1)

0

10

20

30

Dai

ly a

vera

ge a

ir te

mpe

ratu

re (o C

)

-20

-10

0

10

20

30Sn

owpa

ck d

epth

(cm

)0

50

100

150

200

Temperature

Runoff

Snowpacka

c

Responses of snowpack depths, runoff amounts, and nitrate concentrations in drainage water from the Archer Creek catchment to temperature changes from December 1995 to May 1996. Adirondack Mountains, New York State)

Park, J., M.J. Mitchell, P.J. McHale, S.F. Christopher and T.P. Myers. 2003. Interactive effects of changing climate and atmospheric deposition on N and S biogeochemistry in a forested watershed of the Adirondack Mountains, New York State. Global Change Biology 9: 1602‐1619.

a. January - December R2 = 0.30, P = 0.018

3.5 4.0 4.5 5.0 5.5 6.0 6.5

Annu

al N

O3- e

xpor

t (m

ol h

a-1 y

r-1)

0

50

100

150

200

250

b. January - March R2 = 0.72, P < 0.001

-10 -8 -6 -4 -2

NO

3- exp

ort f

rom

Jan

- M

ar (m

ol h

a-1pe

riod-1

)

0

20

40

60

80

Relationships between average air temperatures andNO3

‐ export in Arbutus Lake outlet for either  the entire year or the months preceding spring snowmelt from 1984 to 2001. Winter air temperature major driver of NO3

‐ export. Our current research is trying to evaluate what is causing this relationship.Is this a linear relationship or is there some tipping point?

Average daily air temperature °C

Stream

Low Ca2+

Low NO3-

NO3-

Ca Rich Site (S14) Ca Poorer Site (S15)

Stream

High Ca2+

High NO3-

LitterHigh Ca

Litter Low C:N

Ca2+

Ca-rich parent material Ca-poorer parent material

Litter High C:N

Low Ca

Litter

3-NO

Ca2+

UptakeUptake

Geology affects vegetationinfluence on N dynamics

Christopher, S.F., B.D. Page, J.L. Campbell and M.J. Mitchell. 2006. Contrasting biogeochemistry in two adjacent catchments: The contributions of soil Ca and forest vegetation in affecting spatial and temporal patterns of NO3‐ in surface waters. Global Change Biology 12: 364‐381.

Date

Stre

am 1

4 D

isch

arge

(cfs

)

0

2

4

6

8

10

12

Stre

am 1

5 D

isch

arge

(cfs

)

0.0

0.2

0.4

0.6

0.8

1.0

Sno

w D

epth

(m)

0.0

0.2

0.4

0.6

0.8

1.0

Pre

cipi

tatio

n (m

m)

0

10

20

30

40

50

Stream 15

Stream14S

oil T

empe

ratu

re (o

C)

0

2

4

6

8

10

12 5 cm 15 cm 30 cm 50 cm

Dec Jan Feb Mar Apr

Air

Tem

pera

ture

(o C)

-20

-10

0

10

20

30

Snow

Precipitation

Winter Meteorology and Hydrology atTwo small headwater catchments (S14 and S15) in the Adirondack Mountains of New York

Lisa Kurian, M.S. Thesis SUNY‐ESF (2009) Unpublished data

Rain on snow event

Note changesin soil temperature

Stream 14-Nitrate

2007-2008Nov Dec Jan Feb Mar Apr

NO

3- μm

ol L

-1

0

5

10

15

20

Snow

Dep

th (m

)

0.0

0.5

1.0

1.5

2.0

2.5

3.0Stre

am D

isch

arge

(cfs

)

0.0

0.2

0.4

0.6

0.8

Stream 15-Nitrate

2007-2008Nov Dec Jan Feb Mar Apr

NO

3- μm

ol L

-1

0

5

10

15

20

Snow

Dep

th (m

)

0.0

0.5

1.0

1.5

2.0

2.5

3.0Stre

am D

isch

arge

(cfs

)

0.0

0.2

0.4

0.6

0.8

Stream 14-Calcium

2007-2008Nov Dec Jan Feb Mar Apr

Ca2+

μeq

L-1

0

200

400

600

800

1000

1200

1400Sn

ow D

epth

(m)

0.0

0.5

1.0

1.5

2.0

2.5

3.0Stre

am D

isch

arge

(cfs

)

0.0

0.2

0.4

0.6

0.8

Stream 15-Calcium

2007-2008Nov Dec Jan Feb Mar Apr

Ca2+

μeq

L-1

0

200

400

600

800

1000

1200

1400

Snow

Dep

th (m

)

0.0

0.5

1.0

1.5

2.0

2.5

3.0Stre

am D

isch

arge

(cfs

)

0.0

0.2

0.4

0.6

0.8

Climatic Impacts on Watershed Biogeochemistry:Nitrogen, Sulfur, and Mercury Dynamics

• Sulfur– Importance of sulfur in watersheds linked to chronic and episodic acidification

– Sulfur emissions and SO42‐ concentrations in 

surface waters decreasing in the northeastern U.S.

– Watershed drying and wetting has a major impact on sulfur oxidation and reduction processes (summer events)

Sulfur Reduction/Oxidation

• Dissimilatory Reduction: Sulfate to sulfide (consumes acidity, stored/immobilized as FeS2and organic sulfides)

• Oxidation: sulfide to sulfate (mobilization, generates acidity)

0 Days Discharge

O42-][S

Drier conditions result in higher SO42- concentrations

in Ontario, Canada (Eimers and Dillon, 2002) (Harp Lake). Are droughts a type of tipping point?

Late Summer/Fall Storms at Archer Creek

Storm # (Study) (LTM = Long-Term Avg.)

LTM 1 2 3 4 1 1 2 3

pH

4.5

5.0

5.5

6.0

6.5

7.0

7.5

Mitche

ll et a

l. (20

06)

Inamda

r et a

l. (20

04)

McHale

et al

. (200

2)Lowest pH

Long-Term Avg.

4.7

High Sulfate Concentrations at pH minima

pH4.6 4.8 5.0 5.2 5.4 5.6 5.8 6.0 6.2 6.4

Tota

l Al (μm

ol L

-1)

0

2

4

6

8

10

12

14

16

y= -8.89x+56.0r ² = 0.668

Low pH &

High [Al] =

Sulfur isotopic composition of ecosystem components

• Stable isotope of an element defined by the number of neutrons in the nucleus which can vary

• Isotopic composition Controlled by two factors– isotopic composition of sources (in combination

with mixing ratios)– isotopic discrimination during sulfur

transformations. – Most important-- dissimilatory sulfate reduction

(2H+ + SO42- + 4H2 H2S + 4H2O)

S2-

34S32S

↓δ34S-Sulfide

↑δ34S-Sulfate≅ δ34S-Sulfate

δ34S- Original Sulfate SourceDissimilatory sulfate reduction

Oxidation

SO42-SO4

2-

SO42-

How do the relationships between δ34S values and [SO4

2-] compare among watersheds?

Example of three watersheds in 2002

Mitchell, M.J., S.W. Bailey, J.B. Shanley and B. Mayer. 2008. Evaluating storm events for three watersheds in the northeastern United States: a combined hydrological, chemical and isotopic approach. Hydrological Processes 22:4023-4034.

Sleepers River Cone Pond

Archer Creek

Sleepers River Cone Pond

Archer Creek

Watershed AttributesWatershed Archer Creek Sleepers River Cone Pond

Location Adirondack Mts, NY NE Vermont White Mts, NH

Landuse History/ Vegetation

Second Growth Hardwood

Second Growth Hardwood

First Growth Conifer (some hardwoods); fire in early 1800’s)

Atm. Dep. (03-04) kg/yr

SO4 (13)NO3 (12)

SO4 (14)NO3 (12)

SO4 (14)NO3 (12)

Size (ha) 135 (W-9) 49 53Geology Granitic gneiss

bedrock influenced by Grenville Formation (Ca, Mg) [Inter. base status]

Calcareous granulite interbedded with sulfidic micaceous phyllites & biotite schists [High base status]

Sillimanite-grade metapelites [Low base status]

Hydrology in Late Summer/Early Fall 2002

Date8/1/2002 9/1/2002 10/1/2002

Dis

char

ge m

m d

ay-1

0

1

2

3

4

Prec

ipita

tion

mm

day

-1

0

20

40

60

80

100

120

1401

2

3

4Archer Creek

Date8/1/2002 9/1/2002 10/1/2002

Dis

char

ge m

m d

ay-1

0

1

2

3

4

Prec

ipita

tion

mm

day

-1

0

20

40

60

80

100

120

140

Sleepers River

Date8/1/2002 9/1/2002 10/1/2002

Dis

char

ge m

m d

ay-1

0

1

2

3

4

Prec

ipita

tion

mm

day

-10

20

40

60

80

100

120

140

Cone Pond

Biogeochemical responses vary substantially among watersheds: effect of landscape on climate response

SO4 ( eq L )100 150 200 250 300 350 400

2- μ -1

3410

8

6oo/oS

4

δ 2

0

-2

Archer CreekSleepers RiverCone Pond

SO42- (μeq L-1)

100 150 200 250 300 350 400

δ34 S

o / oo

-2

0

2

4

6

8

10Archer CreekSleepers RiverCone Pond

Low δ34S & high SO42- from oxidized sulfide from

wetlands

SO42- (μeq L-1)

100 150 200 250 300 350 400

δ34

S o / oo

-2

0

2

4

6

8

10Archer CreekSleepers RiverCone Pond

High sulfate and δ34S values from S minerals

SO42- (μeq L-1)

100 150 200 250 300 350 400

δ34 S

o / oo

-2

0

2

4

6

8

10Archer CreekSleepers RiverCone Pond

Low variation: dominance of atmospheric SO42-

Pattern: oxidation of sulfide.

Lower wetland contributions atCone Pond versus Archer Creek

Derived from: Selvendiran, P., C.T. Driscoll, M.R. Montesdeoca, H‐D Choi and T. M. Holsen. 2009. Mercury dynamics and transport in two Adirondack lakes. Limnology and Oceanography 54:413‐427

DOC (mg C L-1)

0 2 4 6 8 10 12 14 16

FTH

g (n

g L-

1 )

0

2

4

6 HBEF r2 = 0.67Sleepers River r2 = 0.78Beaver Meadow r2 = 0.86Lake Inlet r2 = 0.92

Dittman, J.D., J.B. Shanley, C.T. Driscoll, G.R. Aiken, A.T. Chalmers, and J.E. Towse (2009), Ultraviolet absorbance as a proxy for total dissolved mercury in streams. Environ. Pollut., 157, 953‐1956 doi:10.1016/j.envpol.2009.01.031

Bushey, J.T., C.T. Driscoll, M.J. Mitchell, P. Selvendiran, and M.R. Montesdeoca. 2008a. Mercury transport in response to storm events from a northern forest landscape. Hydrological Processes DOI:10.1002/hyp.7091:14 pp.

9/17 9/18 9/19

TSS

(mg

L-1)

0

5

10

15

20

11/10 11/11 11/12

THg

(ng

L-1)

02468

10M

eHg

(ng

L-1)

0.00.10.20.30.40.5

6/30 7/1 7/2

DO

C (m

g L-1

)

0

4

8

12

Ups

trea

mD

isch

arge

(mm

day

-1)

0

2

4

1632 (upstream)(b) (c)(a)

Climatic Impacts on Watershed Biogeochemistry:

Nitrogen, Sulfur, and Mercury Dynamics • What we know

– Events show marked responses to watershed biogeochemistry of NO3

- ,SO42- ,

THg and MeHg)– It is clear that these solute responses vary

as a function of changing atmospheric inputs (N, S, Hg) and climatic conditions of watersheds (temperature, wetting, snow pack, etc.)

• What we do not know– The overall importance of these events in

affecting overall watershed response over the long-term with respect to climate change.

– Whether extreme, infrequent events will have a disproportionate effect on watershed responses compared to more gradual changes.

Climatic Impacts on Watershed Biogeochemistry:

Nitrogen, Sulfur, and Mercury Dynamics

• What we do not know (Continued)– The consequences of these changes in the

overall context of other changes including the effects of atmospheric deposition and biotic changes including invasive species and species composition.

– Will these extreme biogeochemical responses also affect biotic responses?

Climatic Impacts on Watershed Biogeochemistry:

Nitrogen, Sulfur, and Mercury Dynamics