1

2

A consortium of catchment scientists, including: Adam Spargo USA Mary Beth Adams (FER) John Campbell (HBR) Alan Covich (LUQ) David Clow (LVW) Clifford Dahm (SEV) Kelly Elder (FRA) Nancy Grimm (CAP) Julia Jones (AND) Stephen Sebestyen (MAR) James Vose (CWT) Mark Williams (NWT) Canada Fred Beall (TLW) Tom Clair (KEJ) John Pomeroy (MRM) Patricia Ramlal (ELA) Rita Winkler (UPC) Huaxia Yao (DOR)

Rationale

3

0 These watershed studies are unique.

0 Represent longest existing paired records of climate and hydrology.

0 Provide opportunity to explore effects of climate on water yields in headwaters of the US and Canada.

0 Collective potential of these studies is only beginning to be realized.

Coweeta Experimental Forest and LTER, North Carolina (CWT)

Motivation Why are we interested in the Budyko curve?

The Budyko Curve provides a reference condition for the water balance.

If we assume it depicts the expected partitioning of P into Q, then we can begin to account for the reasons why sites vary from the curve.

Russian climatologist 1920 –2001

Jones et al. 2011. Ecosystem processes and human influences regulate streamflow response to climate change at long-term ecological research sites . Bioscience Under Review

Breaking down the Budyko Curve

5

Water limit: AET = P. A site cannot plot above the blue line unless there is an additional input of water beyond precipitation.

Energy limit: AET = PET. A site cannot plot above the red line unless precipitation is being lost (e.g., water lost to groundwater system).

What do deviations from the Budyko curve mean?

HORIZONTAL deviations change in the climatic conditions

6

Warmer, drier

Le

ss r

un

off

VERTICAL deviations change in partitioning

between ET and Q

Question 1 Do average annual values for

reference catchments fall on the Budyko curve?

7

“LTER” Sites (including US LTER, USGS, USFS and Cdn sites)

8

0 30 sites across North America selected based on

0 Reference catchments

0 Coverage of major biomes

0 Coverage of major climate regions

“LTER” Sites (including US LTER, USGS, USFS and Cdn sites)

9

0 30 sites across North America selected based on

0 Reference catchments

0 Coverage of major biomes

0 Coverage of major climate regions

Record length - matched P, T, Q Long-term matching climate and flow records from

US clim/hydroDB website (http://www.fsl.orst.edu/climhy/) Cdn HELP website (http://www.canforhydro.org/)

Ca

tch

me

nt

are

a (

ha

)

Theoretical vs. observed distribution of study sites relative to the Budyko Curve

11

Accuracy of Budyko in predicting long-term (10-yr) average discharge

y = 1.055xR² = 0.5953

y = 1.1321xR² = 0.6922

0

500

1000

1500

0 500 1000 1500O

bse

rved

AET

(P

-Q

)

Predicted AET using Budyko

LUQ

Line of best fit

Line with LUQ removed

13

Hurricane Hugo hit the site in September 1989 reducing above ground biomass by 50%.

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.0 0.2 0.4 0.6 0.8

AET

/P

PET/P

1988 - 1989 (pre huricane)

1990 - 1995 (1- 5 yrs post Hugo)

1996 - 2005 (6 -10 yrs post Hugo)

Luquillo Experimental Forest, LTER site, and USGS WEBB site, Puerto Rico (LUQ)

y = 1.055xR² = 0.5953

y = 1.1321xR² = 0.6922

0

500

1000

1500

0 500 1000 1500

Ob

serv

ed A

ET (

P -

Q)

Predicted AET using Budyko

LUQ

Line of best fit

Line with LUQ removed

Maximum deviation of the longterm average Evaporative index from the Budyko Curve for each of

the 30 study sites

Whilst the Evaporative Index for the 30 sites generally follows the Budyko curve there remains a lot of deviation.

Why does this occur?

-0.30

-0.20

-0.10

0.00

0.10

0.20

0.30

0.40

HB

RA

ND

FER

TLW

MA

RN

TLO

LYC

AS

ELA

DO

RSB

CK

EJM

RM

MA

YK

BS

PIE

TEN

CW

TU

PC

GC

EN

WT

CA

RK

NZ

BN

ZB

ESSE

VLU

QFR

ALV

WC

AP

AR

C

Lon

gte

rm A

vera

ge A

ET/P

de

viat

ion

th

e

Bu

dyk

o

14

0 ARTIFACT?

0 ET=P-Q

0 Timing of P vs. ET over the year

0 Inadequate measures of P, ET, Q

0 Missing measures of other components of the water balance (i.e., groundwater gains or losses?)

0 REAL?

0 Natural disturbance legacy effects

0 Vegetation, topography, and soils may modify water balance

0 Ecosystem may acclimate/adapt to climate change

Whilst the Evaporative Index for the 30 sites generally follows the Budyko curve there remains a lot of deviation.

Why does this occur?

15

0 ARTIFACT?

0 ET=P-Q

0 Timing of P vs. ET over the year

0 Inadequate measures of P, ET, Q

0 Missing measures of other components of the water balance (i.e., groundwater gains or losses?)

0 REAL?

0 Vegetation, topography, and soils may modify water balance

0 Ecosystem may acclimate/adapt to climate change

ET underestimated

Whilst the Evaporative Index for the 30 sites generally follows the Budyko curve there remains a lot of deviation.

Why does this occur?

16

0 ARTIFACT?

0 ET=P-Q

0 Timing of P vs. ET over the year

0 Inadequate measures of P, ET, Q

0 Missing measures of other components of the water balance (i.e., groundwater gains or losses?)

0 REAL?

0 Vegetation, topography, and soils may modify water balance

0 Ecosystem may acclimate/adapt to climate change

Gro

un

dw

ate

r lo

ss

un

de

rest

ima

ted

? Can we use the Budyko curve to identify where we have confidence in closing the water budget? What are our gaps in knowledge?

Question 2 Do year-to-year deviations in the evaporative index

provide insight into the responsivity (resistance) and elasticity (resilience) of catchment water yields

to changing climatic conditions?

17

Spider plots showing year-to-year deviations from long term average

18

Spider plots showing year-to-year deviations from long term average

19

Water Yield Metrics RESPONSIVITY is the degree to which runoff (Q) is synchronized with precipitation (P), and is measured from the deviation in the Evaporative Index (i.e., Δy-axis).

20 Carey et al. 2010. Inter-comparison of hydro-climatic regimes across northern catchments: synchronicity, resistance and resilience. Hydrol. Process. (2010)

Eva

po

rati

ve in

dex

(A

ET

/P)

Dryness Index (PET/P) Dryness Index (PET/P)

HIGH responsivity (or resistance); water yields are expected as P is transferred to Q (synchronous)

LOW responsivity (or resistance); water yields are higher or lower than expected (not synchronous)

Metrics defined for reference

conditions

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9M

axim

um

Inte

ran

nu

al D

evia

tio

n in

AET

/P

No

rmal

ise

d t

o t

he

Bu

dyk

o C

urv

e

Rank ordering of amplitude of year-to-year deviation in Evaporative index (normalized to the Budyko Curve)

21

LOWER responsivity

0.0

0.2

0.4

0.6

0.8

1.0

1.2

0.00 0.20 0.40 0.60 0.80 1.000.0

0.2

0.4

0.6

0.8

1.0

1.2

0.0 0.2 0.4 0.6 0.8 1.00.0

0.2

0.4

0.6

0.8

1.0

1.2

0.0 0.2 0.4 0.6 0.8 1.0

0.0

0.2

0.4

0.6

0.8

1.0

1.2

0.0 0.2 0.4 0.6 0.8 1.0

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9M

axim

um

Inte

ran

nu

al D

evia

tio

n in

AET

/P

No

rmal

ise

d t

o t

he

Bu

dyk

o C

urv

e

Rank ordering of amplitude of year-to-year deviation in Evaporative index (normalized to the Budyko Curve)

22

LOWER responsivity

22 Fraser Andrews Niwot Coweeta

R² = 0.4509

0

100

200

300

400

500

600

700

0 200 400 600 800 1000

An

nu

al

Dis

ch

arg

e (

mm

)

Annual Precipitation (mm)

0.0

0.2

0.4

0.6

0.8

1.0

1.2

0.00 0.20 0.40 0.60 0.80 1.00

R² = 0.9726

0

500

1000

1500

2000

2500

0 500 1000 1500 2000 2500 3000 3500

An

nu

al

Dis

ch

arg

e (

mm

)

Annual Precipitation (mm)

R² = 0.2956

0

200

400

600

800

1000

1200

0 500 1000 1500 2000

An

nu

al

Dis

ch

arg

e (

mm

)

Annual Precipitation (mm)

Andrews

Niwot

0.0

0.2

0.4

0.6

0.8

1.0

1.2

0.0 0.2 0.4 0.6 0.8 1.0

0.0

0.2

0.4

0.6

0.8

1.0

1.2

0.0 0.2 0.4 0.6 0.8 1.0

R² = 0.7645

0

200

400

600

800

1000

1200

1400

1600

0 500 1000 1500 2000 2500

An

nu

al

Dis

ch

arg

e (

mm

)

Annual Precipitation (mm)

Fraser

23

Cowetta

0.0

0.2

0.4

0.6

0.8

1.0

1.2

0.0 0.2 0.4 0.6 0.8 1.0

Associations between year-to-year deviations (inset) and P vs. Q coupling)

Responsivity=0.12

Responsivity=0.64 Responsivity=0.41

Responsivity=0.37

0.0

0.2

0.4

0.6

0.8

1.0

0.0 0.2 0.4 0.6 0.8 1.0

AET

/P

PET/P

Budyko

Can we identify sites where ecosystems are undergoing fundamental changes in response to climatic conditions?

Responsivity=0.19

Caspar (California, USA)

0.0

0.2

0.4

0.6

0.8

1.0

1990 1992 1994 1996 1998 2000 2002 2004

Ru

no

ff R

ati

o

0

100

200

300

400

500

600

700

800

0

500

1000

1500

2000

2500

1990 1993 1996 1999 2002

PET

(m

m)

Pre

cip

itat

ion

, Dis

char

ge (

mm

) Precipitation Discharge PET

-0.1

0.0

0.1

0.2

0.3

0.4

0.5

0.0 0.1 0.2 0.3 0.4 0.5 0.6

AET

/P

PET/P

1974 - 1979

1980 - 1989

1990 - 1995

Budyko

Carnation Creek (British Columbia, Canada)

560

600

640

680

720

760

800

0

500

1000

1500

2000

2500

3000

3500

4000

4500

1974 1977 1980 1983 1986 1989 1992 1995

PET

(m

m)

Pre

cip

itat

ion

, Dis

char

ge (m

m)

Precipitation

Discharge

PET

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1974 1977 1980 1983 1986 1989 1992 1995

Ru

no

ff R

atio

Responsivity=0.42

Can we identify sites where ecosystems are undergoing fundamental changes in response to climatic conditions?

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1994 1997 2000 2003 2006 2009

Run

off R

atio

-0.4

-0.2

0.0

0.2

0.4

0.6

0.8

0.0 0.2 0.4 0.6 0.8

AET

/P

PET/P

1994 - 19992000 - 20072008 - 2009Budyko

0

50

100

150

200

250

300

0

200

400

600

800

1000

1200

1400

1994 1997 2000 2003 2006 2009

PET

(m

m)

Pre

cip

itat

ion

, Dis

char

ge (m

m)

Precipitation Discharge PET

Loch Vale (Colorado, USA)

!

Responsivity=0.72

Can we identify sites where ecosystems are undergoing fundamental changes in response to climatic conditions?

Question 3 Does elasticity lead to shorter recovery times (return to

pre-disturbance water yield) following disturbance?

27

28

Water Yield Metrics ELASTICITY is the degree to which a catchment can return to normal functioning following perturbations, and is measured as the ratio of deviations in dryness index to evaporative index (i.e., Δx-axis/Δy-axis).

Dryness Index (PET/P)

LOW elasticity (<1) is the result of small horizontal range relative to vertical range

Dryness Index (PET/P)

HIGH elasticity (>1) is the result of large horizontal range relative to vertical range

Eva

po

rati

ve in

dex

(A

ET

/P)

Eva

po

rati

ve in

dex

(A

ET

/P)

Carey et al. 2010. Inter-comparison of hydro-climatic regimes across northern catchments: synchronicity, resistance and resilience. Hydrol. Process. (2010)

Responsivity does not imply elasticity

29

y = -2.2806x + 2.0876 R² = 0.2701

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

0.0 0.2 0.4 0.6 0.8 1.0

Ela

stic

ity

Responsivity

y = -2.2806x + 2.0876 R² = 0.2701

0.0

2.0

4.0

6.0

8.0

10.0

12.0

14.0

0.0 0.2 0.4 0.6 0.8 1.0

Ela

stic

ity

Responsivity

0.1

1.0

10.0

Max

imu

m In

tera

nn

ual

Dev

iati

on

in A

ET/P

N

orm

alis

ed

to

th

e B

ud

yko

Cu

rve

Responsivity

Elasticity

10 1 0.1

Max

imu

m i

nte

ran

nu

al

dev

iati

on

in

AE

T/P

10 1 0.1 M

axim

um

in

tera

nn

ual

d

evia

tio

n i

n P

ET

/P /

AE

T/P

Study Area HJ Andrews Hubbard Marcel

Location Oregon New Hampshire Minnesota

Treated Watershed WS01 WS2 S6

Control Watershed WS02 WS3 S2

Cut (year/percent) 1963 / 100% 1964 / 100% 1980 / 87%

Data period pre cut 5 years 7 years 13 years

Data period post cut 47 years 43 years 27 years

Vegetation Type Coniferous Deciduous Deciduous

Demonstration of elasticity vs. recovery following disturbance using paired catchment studies

30

Site selection criteria: 1. Pre and post disturbance data available

2. Similar disturbance (100% cut)

Marcell (MAR)

Hubbard (HUB)

HJ Andrews (HJA)

Study Sites HJA HUB MAR

Treated Watershed

WS01 WS2 S6

Vertical Variation (V)

0.08 0.18 0.37

Horizontal Variation (H)

0.03 0.24 0.54

Ratio of Horizontal to Vertical (H/V)

0.38 1.36 1.46 0.0

0.2

0.4

0.6

0.8

1.0

0.0 0.2 0.4 0.6 0.8 1.0

Evap

ora

tive

Ind

ex (

AET

/P

Dryness Index (PET/P)

0.0

0.2

0.4

0.6

0.8

1.0

0.0 0.2 0.4 0.6 0.8 1.0

Evap

ora

tive

Ind

ex (

AET

/P

Dryness Index (PET/P)

0.0

0.2

0.4

0.6

0.8

1.0

0.0 0.2 0.4 0.6 0.8 1.0

Evap

ora

tive

Ind

ex (

AET

/P

Dryness Index (PET/P)

HJ Andrews – WS01

Hubbard – WS2

Marcell – S6

31

Elasticity metrics for study sites

Study Sites HJA HUB MAR

Treated Watershed

WS01 WS2 S6

Vertical Variation (V)

0.08 0.18 0.37

Horizontal Variation (H)

0.03 0.24 0.54

Ratio of Horizontal to Vertical (H/V)

0.38 1.36 1.46 0.0

0.2

0.4

0.6

0.8

1.0

0.0 0.2 0.4 0.6 0.8 1.0

Evap

ora

tive

Ind

ex (

AET

/P

Dryness Index (PET/P)

0.0

0.2

0.4

0.6

0.8

1.0

0.0 0.2 0.4 0.6 0.8 1.0

Evap

ora

tive

Ind

ex (

AET

/P

Dryness Index (PET/P)

0.0

0.2

0.4

0.6

0.8

1.0

0.0 0.2 0.4 0.6 0.8 1.0

Evap

ora

tive

Ind

ex (

AET

/P

Dryness Index (PET/P)

HJ Andrews – WS01

Hubbard – WS2

Marcell – S6

32

Elasticity metrics for study sites

If elasticity is linked to time required to return to pre-disturbance water yields, then we expect HJ Andrews to have a long recovery.

33

Elasticity metrics for study sites

Lower elasticity results in longer recovery times in water yields

following disturbance.

Study Site HJA HUB MAR

Treated Watershed

WS01 WS2 S6

Vertical Variation (V)

0.08 0.18 0.37

Horizontal Variation (H)

0.03 0.24 0.54

Ratio of Horizontal to Vertical (H/V)

0.38 1.36 1.46

-200

-100

0

100

200

300

400

500

600

195

3

1956

1959

1962

1965

1968

1971

1974

1977

1980

1983

1986

1989

1992

1995

1998

2001

2004

2007

Ob

serv

ed -

pre

dic

ted

dis

char

ge (

mm

)

HJ Andrews – WS01

Elasticity=0.38 Recovery=50+ yr

-200

-100

0

100

200

300

400

500

600

195

8

1961

1964

1967

1970

1973

1976

1979

1982

1985

1988

199

1

1994

1997

200

0

2003

2006

Ob

serv

ed -

pre

dic

ted

dis

char

ge (

mm

)

Hubbard – WS2

Elasticity=1.36 Recovery=10 yr

-200

-100

0

100

200

300

400

500

600

1965

1968

1971

197

4

1977

1980

1983

1986

1989

1992

1995

1998

2001

2004

Ob

serv

ed -

pre

dic

ted

dis

char

ge (

mm

)

Marcell – S6

Elasticity=1.46 Recovery=1 yr

Summary Budyko curve described partitioning of P into ET and Q Deviations (average) provide insight into

0 Inaccurate or incomplete representation of water balance components (HJA, MAR)

0 Natural disturbances and their legacies (LUQ)

Deviations (year to year) provide insight into responsivity (resistance) and elasticity (resilience) of water yields to global change Future work will focus on:

0 Incorporating uncertainty estimates in water balances 0 Discriminating climate signal from natural or anthropogenic

disturbance effects 0 Exploring future scenarios and how they may result in changes in

water yields 0 Considering downstream consequences to water supplies

34

Acknowledgements

0 “LTER Synthesis Workshops” funded by the LTER Network Office

0 NSF LTER grants to participating USA sites

0 USFS and USGS for initial establishment and continued support of watershed studies at many of the study sites

0 NCE-SFM funded project on HydroEcological Landscapes and Processes (HELP) and the participating Canadian sites

Future Changes in Water

Fire Snow Outbreaks

Logging

Roads, landslides

Exurban expansion