Integrating geochemical tracers with physics-based ... · NSF Hydrology Program and CZEN ... 326 ha...

Integrating geochemical tracers with

physics-based modeling to understand

Rio Icacos storm response

Funding from

NSF Hydrology

Program and CZEN

Andy Kurtz Boston University

Festo Lugolobi

Guido Salvucci

326 ha monolithogic

catchment

50 Ma quartz diorite

4200mm annual rainfall

22°C mean annual temp

Critical Zone Exploration Network

Water, Energy, and Biogeochemical Budgets

Rio Icacos Study AreaResearch Approach:

-Understand tracer behavior

in soils, porewaters

-Link tracer to stream chemistry,

infer solute sources and flowpaths

-Test inferences against simulations

(physics-based hydrology model)

Geochemical and Isotopic Tracers of Solutes

Ge substitutes for Si in

minerals, behaves like Si

in solution

Silicon tracer:

Ge/Si ratios in water

reflect fractionation by

weathering and

biological cycling

0

0.5

1

1.5

2

2.5

0.000 0.010 0.020 0.030 0.040

Ge

/Si (

µm

ol/

mo

l)

1/Si

Streamwater

Bedrock Springs

Overland flow

Surface ponding

Streamwater Ge/Si ratios reflect mixing

Streamwater Ge/Si ratios reflect mixing

0

0.5

1

1.5

2

2.5

0.000 0.010 0.020 0.030 0.040

Ge

/Si (

µm

ol/

mo

l)

1/Si

Streamwater

Bedrock Springs

Overland flow

Surface ponding

Streamwater falling

Streamwater Ge/Si ratios reflect mixing

0

1

2

3

4

5

6

0.000 0.010 0.020 0.030 0.040

Ge

/Si (

µm

ol/

mo

l)

1/Si

Streamwater

Bedrock Springs

Overland flow

Surface ponding

Streamwater falling

Soil water

Example Event: 6-24-2006 StormPrecip. = 3.3 cm in~2 hours

0

1

2

3

4

5

6

7

8

9

100

1

2

3

4

5

6

7

8

9

10

8:00 10:00 12:00 14:00 16:00

Pre

cip

(cm

/hr)

Dis

char

ge (

m3

/se

c)

Time

Example Event: 6-24-2006 Storm

0

0.2

0.4

0.6

0.8

1

1.2

1.4

0 1 2 3 4 5 6 7

Ge

/Si (

µm

ol/

mo

l)

Discharge (m3/sec)

Example Event: 6-24-2006 Storm

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

8:00 10:00 12:00 14:00 16:00

frac

tio

n

Time

Error bars reflect propagation of analytical and endmember uncertainty

Rising

Three component mixing

“surface water”

“ground water”

“soil matrix water”

Physics-based 3-D integrated

subsurface-surface hydrologic model

-Surface

-Porous Media

-Macropore

FLOW and TRANSPORT

Hydrologic Model InHM VanderKwaak and Loague

Parameterized with

hydrometric and physical

data from site

(soil characteristic tables,

hydraulic conductivity,

porosity)

Finite element grid from DEM

Soil, saprolite, bedrock layers

(days)

(m3/s

ec)

Model Performance: July-August 2007

“flashiness” requires dominance of overland flow

TDR data:

hillslope

site

Model

simulation:

hillslope

site

(sec)

(sec)

Model Performance: July-August 2007

Synthetic hydrograph separationModel Tracers6-24-2006 Storm

June Event: Surface flow velocity (m/s) at max Q

June Event: Surface “Event” tracer at max Q

June Event: Porous Media “Pre-Event” tracer at max Q

ConclusionsGeochemistry

• Streamwater Ge/Si data indicate three components to stormflow

• Shallow/overland flow component dominates most of hydrograph

• A soil matrix component becomes significant at the end of hydrograph recession

Model

• Model indicates catchment “flashiness” controlled by (Hortonian) overland flow rather than macropores

• Application of model tracers indicate dominance of “event water” delivered by overland flow early in event, and displaced “pre-event water” during hydrograph recession.

Model serves as a useful test of geochemical inferences, some consistency, but may be failing to capture some important processes.

Oxygen Isotopes - trace ‘new’ vs. ‘old’ water

Rain (6-24-06 event) = -3.3

Baseflow = -2.7

Falling LimbRising Limb

Isotope Hydrograph Separation

0

10

20

30

40

50

60

70

80

90

100

0 2 4 6 8 10

Time (hours)

% C

on

trib

uti

on

Event water

Pre-event waterFalling Limb

Solute

sources

from a soil

perspectiv

e

Geochemical and Isotopic Tracers of Solutes

Both of these tracers have paleoceanographic applications as well

Geochemical and Isotopic Tracers of Solutes

87Sr/86Sr ratios reflect

age and Rb/Sr ratio of

rocks and minerals

Cation tracer:

Sr isotope ratios in water

reflect source of cations

(mineral weathering, ion

exchange, weathering)

Ge/Si Geochemistry "pseudo-isotopic" behavior

Ge

C

Sn

Pb

Si

6

14

32

50

82

Group

IVB

W.W. Norton

Ge substitutes for Si in the silicate lattice

(Goldschmidt's “camouflage”)

1 to 5 atoms of

Ge per million

silicon atoms

Ge/Si Geochemistry "pseudo-isotopic" behavior

Ge follows Si through marine & terrestrial cycling

pH

Model of Ge/Si fractionation during

incongruent silicate weathering

. .

0

1

2

3

4

Unweathered

Rock (~1.4)

Secondary

Clays (~3.5)

Solute

(~0.35)

Incongruent

Weathering

of Primary

Silicates

(low-intensity)

Congruent

Dissolution of

Secondary Clays

(high intensity)

Streamwater

Ge/Si Results

from

Two-Component

MixingG e / S i x 1 0

- 6

Solute

(~3.5)

Str

eam

s

after Murnane and Stallard, 1990; Froelich et al., 1992

Ge/Si fractionation in synthetic allophane

Ge/Si = 1.0µmol/mol

NeoformedallophaneGe/Si = 1.3µmol/mol

52 ppm Si,55 ppm Al

24 ppm Si,1.6 ppm Al

Ge/Si = 0.5µmol/mol

5 days @ 90°C

NaOH titration,

Soil solids

Perspective

0

100

200

300

400

500

600

700

800

0 2 4 6 8

CaO (wt%)

de

pth

(cm

)

0

100

200

300

400

500

600

700

800

5 10 15 20 25

Al2O3 (wt%)

de

pth

(cm

)

0

100

200

300

400

500

600

700

800

60 65 70 75 80 85

SiO2 (wt%)

de

pth

(cm

)

Luquillo Ridgetop Saprolite Core Profiles

Weathering profiles aren’t like sediment cores…

Conveyor belt goes UP

(landscape eroding 25-50 m/my)

Losing Plagioclase

Losing Kaolinite

Hillslope Soil Profile

40 cm

PR Field photos90 cm

On Ridgetops, this

saprolite-bedrock

contact is at 800cm

depth

Solid-phase Ge/Si ratios (µmol/mol)

Regolith Minerals

Kaolinite = 5.9

Altered Biotite = 5.3

Quartz = 0.5

Primary Minerals

Hornblende = 6.6

Biotite = 5.5

Plagioclase = 1.5

Quartz = 0.5

Soil

= 2.5

Saprolite

= 2.8 to 3.3

Quartz

Diorite

= 2.0

Opal Phytoliths: 0.1 to 0.5

“Immobile Element” Mass Balance Framework

j,w Cj,w

Cj, pCi, p

Ci,w

1

Fractional net loss (or gain) of a mobile element j (e.g. Si)

calculated relative to immobile element i (Nb, Zr)

Positive tau = gain of mobile element

Negative tau = loss of mobile element

Solid-phase Ge/Si ratios (µmol/mol)

Soil

= 2.5

Saprolite

= 2.8 to 3.3

Quartz

Diorite

= 2.0

Weathering Ge/Si Fractionations

Plagioclase ----> Kaolinite

60% of Si

40% of Si Ge/Sisoln

= 0.4

Biotite ----> Kaolinite and

Kaolinite ----> Solution

33% of(remaining) Si

Ge/Sisoln

= 3.7

Si=-0.4

Si=-0.6

Soil Water

Perspective

Ceramic-cup

pressure-

vacuum

water

samplers

(lysimeters)

Ridgetop Site (LG-1)

Nested Lysimeters

in deep saprolite

15cm to >800cm

Rio Icacos Study Area

Soil and Saprolite Porewaters Ridgetop Site LG-1

0

100

200

300

400

500

600

700

800

0 100 200

Si (µmol/L)

dep

th(cm

)

0

100

200

300

400

500

600

700

800

0 1 2 3 4 5

Ge/Si (µmol/mol)

(WEBB project suction lysimeters)

Predictions

Ge/Sisoln= 0.4

Ge/Sisoln= 3.7

Not Seeing 0.4 here…

“Landslide Water”

Si = 340 µmol/L

Ge/Si = 0.3 µmol/mol

Stream

Perspective

Rio Icacos at Flood

Stage, Nov „06

USGS

Stream

Gauge

ISCO STREAM

SAMPLER

Silica

0

50

100

150

200

250

300

350

0 50 100 150 200 250

Discharge (ft3/sec)

Si (µ

mo

l/L

)

BaseflowRising LimbFalling Limb

Baseflow

Ge/Si & [Si] chemical hydrograph separation

fg Rt Rs

Rg Rs

2) Use Ge/Si (“R”) to partition Si flux into

“groundwater” and soilwater components

3) Multiply Si fluxes by endmember Si concentrations to

determine water flux of components

1) Define “groundwater” and “soilwater” endmembers (Si

concentration and Ge/Si) based on data

4) Close water balance by adding “Si free water”

Hydrograph Separation Based on Ge/Si and [Si]

0

10

20

30

40

50

60

70

80

90

100

0 2 4 6 8 10

Time (hours)

% C

on

trib

uti

on

Soil water

Si-free water

Groundwater

Falling LimbRising

Hydrograph Separations Based on Ge/Si and [Si]

0

10

20

30

40

50

60

70

80

90

100

0 2 4 6 8 10

Time (hours)

% C

on

trib

uti

on

Soil water

Si-free water

Groundwater

Falling Limb

Flowpath Control over Solutes

"Storm Flow" Conditions: Subsurface

stormflow dominates: Soils contribute to

solute load

Dilute solutions with high Ge/Si (& 87Sr/86Sr)

Carried by “New Water”

"Base Flow" Conditions: Groundwater

discharge maintains streamflow. Solute load

reflects incongruent weathering of primary

silicates

Concentrated solutions with low Ge/Si, (& 87Sr/86Sr)

Carried by “Old Water”

Physics-based 3-D

integrated subsurface-

surface hydrologic model

InHMVanderKwaak and Loague

Hydrologic Model

Porosity

Physics-based 3-D

integrated subsurface-

surface hydrologic model

InHMVanderKwaak and Loague

Hydrologic Model

Total Hydraulic Head

Physics-based 3-D

integrated subsurface-

surface hydrologic model

InHMVanderKwaak and Loague

Hydrologic Model

Saturation

Physics-based 3-D

integrated subsurface-

surface hydrologic model

InHMVanderKwaak and Loague

Hydrologic Model

Porous media flow velocity (m/s)

Physics-based 3-D

integrated subsurface-

surface hydrologic model

InHMVanderKwaak and Loague

Hydrologic Model

Model Slice - Flow Velocity

Physics-based 3-D

integrated subsurface-

surface hydrologic model

InHMVanderKwaak and Loague

Hydrologic Model

Model Slice - Flow Velocity

Hydrologic Model

Synthetic storm hydrograph

Hydraulic Conductivity determined by Guelph Permeameter

Stormflow zone?