Plot Scale © Oregon State University Isotope Hydrology Shortcourse Prof. Jeff McDonnell Richardson Chair in Watershed Science Dept. of Forest Engineering.

Plot Scale

© Oregon State University

Isotope Hydrology Shortcourse

Prof. Jeff McDonnellRichardson Chair in Watershed Science

Dept. of Forest Engineering

Oregon State University

HydrologicModels

Isotopes in Hydrological Models

Plot Scale


Outline

Day 1 Morning: Introduction, Isotope Geochemistry Basics Afternoon: Isotope Geochemistry Basics ‘cont, Examples

Day 2 Morning: Groundwater Surface Water Interaction, Hydrograph separation

basics, time source separations, geographic source separations, practical issues

Afternoon: Processes explaining isotope evidence, groundwater ridging, transmissivity feedback, subsurface stormflow, saturation overland flow

Day 3 Morning: Mean residence time computation Afternoon: Stable isotopes in watershed models, mean residence time and

model strcutures, two-box models with isotope time series, 3-box models and use of isotope tracers as soft data

Day 4 Field Trip to Hydrohill or nearby research site

Catchment Scale

IsotopeBasics

ConcludingStatementsHydrologic

Models

Plot Scale


Example 1

HydrologicModels

Plot Scale


Svartberget Sweden

Isotopes in model validation

Plot Scale


Svartberget Sweden

Isotopes in model validation

Plot Scale


Example 2

HydrologicModels

Plot Scale


Maimai New Zealand

Isotopes in model calibration

Plot Scale


Catchment

Scale

Maimai Watershed in New Zealand

HydrologicModels

Plot Scale


Process Study Findings

Channel stormflow is usually 85-90% old water (Pearce et al., 1986 WRR)

Most subsurface stormflow is via soil pipes at the soil bedrock interface (Mosley, 1979 WRR; McDonnell, 1990 WRR).

Riparian zone, hillslopes and hollows have distinct 18-O (McDonnell et al., 1991 WRR and base cation concentration (Grady et al., in review).

Mean age of baseflow is 3 mo (Stewart and McDonnell, 1991 WRR).

…process experimental work at Maimai for past 25 yr

Catchment Scale

HydrologicModels

Plot Scale


Hillslope

Riparian Zone

Hollow

Catchment Scale

HydrologicModels

Plot Scale


A Three Box Model

H ills lope box

R iparian box

P E

Runoff

U m ax

UU m in

H ollow box

P E

P E

Seibert and McDonnell, 2002 WRR

Catchment Scale

HydrologicModels

Plot Scale


Namol/L)

0

10

20

30

40

50

60

70

80

90

100

K (

mol/L)

StreamRainRiparian ZoneSoil-RidgeSoil-Hollow

0 50 100 150 200 250

Riparian ZoneHollow

Hillslope

Geochemical End MembersCatchment

Scale

HydrologicModels

Plot Scale


McDonnell et al.,1991; WRR)

Hillslope

Hollow

Riparian

Cluster Analysis of Deuterium Concentration in Subsurface Water

Catchment Scale

HydrologicModels

Plot Scale


A Three Box Model

H ills lope box

R iparian box

P E

Runoff

U m ax

UU m in

H ollow box

P E

P E

Hillslope typediscussed earlier…

Seibert and McDonnell, 2002 WRR

Catchment Scale

HydrologicModels

Plot Scale


The hollow box

<0

<0

dSdZdZ

Storm Rainfall

Catchment Scale

HydrologicModels

Plot Scale


The time series

(c) 1987 Hydrological Data Series (20min.)

0

2

4

6

8

01/09/87 16/09/87 01/10/87 16/10/87 01/11/87 16/11/87 01/12/87 16/12/87 01/01/88

Date

0

5

10

15

20

25

30

35

40

45

Discharge

Rainfall

Catchment Scale

HydrologicModels

Plot Scale

© Oregon State University….but does it work for the right reasons???28-Sep 8-O ct 18-O ct 28-O ct 7-N ov 17-N ov 27-N ov

0

2

4

6

Q [

mm

/h]

0

1

2

3

Gro

undw

ater

leve

l [m

]

O bserved QSim ulated Q

Hills lopeHollowR iparian

Model efficiency 0.93Model efficiency 0.93Catchment

Scale

HydrologicModels

Plot Scale


But does this agree with our conceptual picture of the how the watershed works based on our isotope information????

HydrologicModels

Plot Scale


Model performance with hard data Model performance with hard data calibrationcalibration

0

0.2

0.4

0.6

0.8

1

Goo

dnes

s m

easu

re

A1, A2 and A3A1 and A2

Q, hard GW & new water

Q and new water

Q and soft GW A1 Q

Runoff efficiencyGW hardGW soft

Parameter valuesNew water

Increasing soft data

Catchment Scale

HydrologicModels

Plot Scale


Soft Data: Qualitative knowledge from the experimentalist that cannot be used directly as exact numbers (e.g. % new water, soil depth, reservoir volume, macropore flow, etc

Bypass flowand mixing

Pipeflow of oldwater

Rainfall/snow

How can we use the process knowledge

Soils

Catchment Scale

HydrologicModels

Plot Scale


Type of soft data

New water contribution to peak runoff

Range of groundwater levels, min/max, fraction of saturated soil

Frequency of groundwater levels above a certain level

Parameter values Fraction of riparian, hillslope, hollow Porosity of riparian, hillslope, hollow Soil depth of riparian, hillslope, hollow Threshold level in hollow zone

Catchment Scale

HydrologicModels

Plot Scale


Dialog between the experimentalist and modeler

Experimentalist ModellerEvaluation rules

Values for evaluation rules (ai)

a1

a2 a3

a4

0

1

Model value or parameter

“Degree of acceptability”

Catchment Scale

HydrologicModels

Plot Scale


Different ways of evaluating Different ways of evaluating model acceptabilitymodel acceptability

Acceptability according to: Example Measure

A1 Fit between simulated and Runoff

Efficiency observed dataA2 Agreement with process New water

Percentage of (qualitative) knowledge contribution peak flowA3 Reasonability of parameter Spatial extension

Fraction of values according to of riparian zone catchment area experimentalist 321321

321 nnnnwithAAAA n nnn

Combined objective function:

Seibert and McDonnell, 2002 AGU Monograph

Catchment Scale

HydrologicModels

Plot Scale


Soft data discussions

4

4334

4

32

2112

1

1

0

1

0

)(

axif

axaifaa

xa

axaif

axaifaa

ax

axif

x

a1

a2 a3

a4

Fuzzy Rules- new water at peak- reservoir volumes, Ksat etc- range of gw levels- hollow threshold level

0.03

0.06 0.12

0.15

(30/9/87 event, McDonnell et al. 91; WRR)

Catchment Scale

HydrologicModels

Plot Scale


Improvement of model performance with Improvement of model performance with soft datasoft data

0

0.2

0.4

0.6

0.8

1

Goo

dnes

s m

easu

re

R unoff effic iencyG W hardG W softParam eter va luesN ew w ater


e.g. from Seibert and McDonnell, 2002 WRRe.g. from Seibert and McDonnell, 2002 WRR

Catchment Scale

HydrologicModels

Plot Scale

© Oregon State University….but does it work for the right reasons???

28-Sep 8-O ct 18-O ct 28-O ct 7-Nov 17-N ov 27-N ov

0

2

4

6

Q [

mm

/h]

0

1

2

3

Gro

undw

ater

leve

l [m

]


H ills lopeH ollowR iparian

Model efficiency 0.93Model efficiency 0.93Catchment

Scale

HydrologicModels

Plot Scale


Model efficiency 0.92


0

2

4

6

Q [

mm

/h]

0

1

2

3

Gro

undw

ater

leve

l [m

]



…….maybe not?.maybe not?

Catchment Scale

HydrologicModels

Plot Scale


Model efficiency 0.93


0

2

4

6

Q [

mm

/h]

0

1

2

Gro

undw

ater

leve

l [m

]



…….maybe not???!!.maybe not???!!

Catchment Scale

HydrologicModels

Plot Scale


Improvement of model performance with Improvement of model performance with soft datasoft data

0

0.2

0.4

0.6

0.8

1

Goo

dnes

s m

easu

re

R unoff effic iencyG W hardG W softParam eter va luesN ew w ater


e.g. from Seibert and McDonnell, 2002 WRRe.g. from Seibert and McDonnell, 2002 WRR

Catchment Scale

HydrologicModels

Plot Scale


Example 3

Catchment Scale

HydrologicModels

Plot Scale


Brugga Basin, Germany

Isotopes in model structure

Plot Scale


Beyond a 1km2 research watershed

Obtain as much map info as possible

Do synoptic survey of stream flow (if possible, temp, pH, EC etc)

Gauge trib junctions

Measure mean age of water

Dominant runoff generation processes Example Rietholzbach catchment

Translation into model elements

Catchment Scale

HydrologicModels

Plot Scale


TopographyCatchment

Scale

HydrologicModels

Plot Scale


Soils and Geology

S o il

R e g o s o l

S a u r e B r a u n e r d e

B r a u n e r d e

K a lk b r a u n e r d e

V e r b r a u n t e r u . B u n t e r G l e y

F a h le r g le y

Geology

Moräne der W ürm-Vergletscherung

Würm, Niederterasse

OSM, Tösswald Schichten

OSM, Öhninger Schichten

Schw emmkegel

Catchment Scale

HydrologicModels

Plot Scale


Digital Elevation Model

Topo Index

Topography Slope

Curvature

HydrologicModels

Plot Scale


HortonianOverland

Flow

SaturationOverland

Flow

Sub-surface

Flow

Process IdentificationCatchment

Scale

HydrologicModels

Plot Scale


Dominant Runoff Processes

Hortonrarely saturatedsometimes saturatedFrequently saturatedOften saturatedAlways saturatedSubsurface flowDrained areasNo runoff

Catchment Scale

HydrologicModels

Plot Scale


MRT and watershed modeling

.... following Uhlenbrook et al. (2002) WRR

Discharge

Catchment Scale

HydrologicModels

Plot Scale


Conceptualization of Runoff Processes


Catchment Scale

HydrologicModels

Plot Scale


Runoff Generation

flat hilltop, weathering profile base layer, solifluction main layer, solifluction course top layer boulder fieldaccumulation zonesaturated area


Catchment Scale

HydrologicModels

Plot Scale


Summary

Day 1 Morning: Introduction, Isotope Geochemistry Basics Afternoon: Isotope Geochemistry Basics ‘cont, Examples

Day 2 Morning: Groundwater Surface Water Interaction, Hydrograph separation

basics, time source separations, geographic source separations, practical issues

Afternoon: Processes explaining isotope evidence, groundwater ridging, transmissivity feedback, subsurface stormflow, saturation overland flow

Day 3 Morning: Mean residence time computation Afternoon: Stable isotopes in watershed models, mean residence time and

model strcutures, two-box models with isotope time series, 3-box models and use of isotope tracers as soft data

Day 4 Field Trip to Hydrohill or nearby research site

Catchment Scale

IsotopeBasics

ConcludingStatementsHydrologic

Models

Plot Scale © Oregon State University Isotope Hydrology Shortcourse Prof. Jeff McDonnell Richardson Chair in Watershed Science Dept. of Forest Engineering.

Documents

box models

statements hydrologic

watershed models

hydrological models

new zealand hydrologic

model validation slide

wrr mcdonnell

model calibration slide