2015 CUAHSI Spring Cyberseminar Series - Adrian Harpold

Water Vapor Fluxes from Snow Covered Landscapes: The Importance of Biotic

and Abiotic-Mediated Processes

Adrian A. Harpold Natural Resources and Environmental Science

University of Nevada, Reno CUAHSI Cyberseminar 4/17/2015

Ecohydrological Paradoxes and Tradeoffs of Water Vapor Fluxes in Snowy Systems

How can snowmelt be both an efficient irrigator of vegetation and streamflow generation?

Will warming temperatures generate more/less vapor loss?

What are the tradeoffs between canopy interception and snowpack sublimation?

What are the tradeoffs between abiotic and biotic vapor losses on overall water budgets?

What can we learn from natural experiments (i.e. forest disturbance) to answer these questions?

Acknowledgements

Paul Brooks, University of Utah

Joel Biederman, USDA ARS

Patrick Broxton, University of Arizona

John Knowles, University of Colorado

Theo Barnhart, University of Colorado

Noah Molotch, University of Colorado, JPL

1981-2010 Mean Precipitation

Topography, Water, and Carbon Co-vary

Sierra Nevada

S. Rockies Appalachians


5

1981-2010 Min Temperature


Sierra Nevada



6

1981-2010 Min Temperature


Kellendorfer et al., 2011, RSE

Sierra Nevada


Dry Year P=73 cm

ET Q

The Water Balance of Snow-Dominated Systems

Water balance is dynamic Example: Upper Truckee, CA

Simplified equation P=ET+Q

Wet Year P=155 cm

ET Q

From USFS

30-Year Average

ET Q

Interception

Snow sublimation

Soil evap Transpiration

Storage

Streamflow

Dry Year P=73 cm

ET Q

The Water Balance of Snow-Dominated Systems

Water balance is dynamic Example: Upper Truckee, CA

More resolved equation:

How do we partition P into various stores and fluxes?

P=Esoil+Vinterception+Vsublimation+T+Q+S

Snowmelt Timing Westerling et al., 2006

Biotic Water Demand is Changing: Effects of Forest Fire

Year 2000 Fire Regime

Schmidt et al., 2002

Beetle Outbreaks (Meddens et al., 2012)

Large departure from historical means Small departure from historical means

Snowpack Dynamics Are Changing: Changes in AccumulaIon

Change in snow to rain Nov to March 1949-2004

Knowles, 2006, Journal of Climate

Change SWE 1950-2000

Mote et al., 2009

More rain, less

snow

Smaller April 1 snowpacks

Snowpack Dynamics Are Changing: Changes in AblaIon

Trends over the last 30 years (1980-2010) Shorter snowmelts (SM50=Ime 50% melt occurs) Increased sublimaIon (SWE: Winter P raIo)

Harpold et al., 2012, WRR

Paradox/Tradeoff 1: How can snowmelt be both an efficient

irrigator of vegetation and streamflow generation?

Snow is a More Efficient Streamflow Generator Than Rain

Snowy watersheds show a range of ET/P Snowy watersheds generate more

streamflow when normalized to climate

Berghuijs et al., 2014, Nature CC

Long-term average for one watershed (red=snowy, green=rainy)

Over generates streamflow

Under generates streamflow

Snow is a More Efficient Streamflow Generator Than Rain. Why?

259 sites, 2100+ station years

Snowmelt is responsible for peak soil moisture (PSM) response across varying stations

Consequently, soil field capacity most likely to be exceeded during snowmelt

Harpold and Molotch, in prep

1:1 relationship between peak soil moisture and snow

disappearance

Snow is a More Efficient Streamflow Generator Than Rain. Why?

Model results suggest that more rapid snowmelt generates more streamflow (less vapor losses)

1100'0"W1200'0"W

450'0"N

400'0"N

350'0"N

350Kilometers

Barnhart et al., in prep

High snowmelt rates efficiently generate streamflow

High snow fractions show range of response

Study Area

Snowmelt is an Efficient Irrigator

Niwot Ridge Ameriflux carbon measurements

Corresponding snow depth

Synchronicity between carbon uptake and snow water availability

0 50 100 150 200 2500

100

200Sn

ow d

epth

cm

2007

Ordinal Day0 50 100 150 200 250

5

0

NEE

umol/

m2 /s

Snow

Dep

th (c

m) NEE (um

ol/m2/s)

Snowmelt begins & NEE increases

Maximum annual NEE occurs at snow disappearance

Snowmelt is an Efficient Irrigator Longer growing season

lead to less net ecosystem productivity (NEP)

Snow water used for transpiration throughout the year

GPP

(g C

m-2

wee

k-1 )

Hu et al., 2010, GCB

Snow derived Rain derived

Less SWE, less NEP

Paradox/Tradeoff 2: Will warming temperatures generate more/less

biotic vapor losses?

BioIc Controls: LimitaIons

Most of the snow-dominated Western U.S. has both temperature and water limitaIons on transpiraIon

Boisvenue and Running, 2006; GCB

Biotic Controls: Changes in Temperature Limitations

Most runoff is generated at high elevations

Reduced temperature limitations can increase ET and decrease runoff

Assumes forests move up in elevation

Goulden and Bales, 2014, PNAS

Most runoff comes from high elevations

Increased ET w/ warming

Biotic Controls: Importance of Alpine and Subalpine Area

Knowles, Harpold, et al., (in review), Hydro. Proc.

Spatial distributions matter! 35% of catchment area generates 60% of streamflow Catchment water balance will depend on how these

ecosystems respond and adapt to warming temperatures

From niwot.colorado.edu

Paradox/Tradeoff 3: What are the tradeoffs between canopy interception and snowpack

sublimation?

Abiotic Controls: Tradeoffs Between Interception and Snowpack Ablation

Distribution of snow in healthy forests reflects interception and sublimation losses

Pea

k S

WE

to P

Rat

io

(SW

E/P

)

Canopy sublimation

Snowpack sublimation

Distance (m)

Canopy cover

Canopy cover

Forest Structure Influences on Abiotic Vapor Losses

Lidar shows impacts of interception and ablation across mosaic of canopy structure and topography

Canopy position matters!

Snow%Depth%(cm)%

0.4%

0.3%

0.2%

0.1%

%%%0%20%%%%%%%40%%%%%%%60%%%%%%%%80%%%%%%%100%%%%%%120%%%%%140%Pr

obab

ility%of%O

ccurrence%

Under%Canopy%Near%Canopy%Distant%Canopy%

Observed:%Solid%Line%Modeled:%Dashed%Line%

1000 m

100 m

Broxton et al., Ecohydrology, 2015 0"

100"

200"

Snow"Depth"(cm)"

Near canopy Distant NUnder canopy

Predicting Abiotic Vapor Losses: Snow Physics and Laser Mapping (SnowPALM)

Broxton et al., Ecohydrology, 2015

Topography and canopy structure parameterized at 1-m resolution

Forced by tower micrometeorology

Verified with snow depth at 1-m scale

0%

10%

20%

30%

40%

50%

Snow sublimation Interception Total vapor losses

Frac

tion

of W

inte

r P

Jemez, NM Boulder, CO

PredicIng AbioIc Vapor Losses: Site-Level Controls

Climate and forest structure lead to differing tradeoffs between interception and snow sublimation at each site

Snowpack Vapor Loss (mm)

39 65 92

Snowpack Vapor Loss (mm)

119 166 212


Boulder, CO Jemez, NM

50%

55%

60%

65%

70%

75%

1 meter 3 meter 10 meter 30 meter 100 meter

Frac

tion

of to

tal V

fr

om s

ublim

atio

n

Jemez, NM Boulder, CO

Smart Forest Management: Fine-Scale Canopy Matters For Water Partitioning Higher resolution leads to different

estimates using the same physics Characterizing canopy as under or

open is insufficient

1000 m

100 m

Sublimation increases 15%

Jemez River

Boulder Creek


Paradox/Tradeoff 4: What are the tradeoffs between abiotic and biotic

vapor losses in snow dominated systems?

Tradeoffs In Abiotic and Biotic Vapor Losses: Forests and Alpine

Response of water budgets depend strongly on distribution of abiotic and biotic-mediated processes

Changes in runoff generation in alpine areas from warming could overwhelm changes in subalpine forests

!

!

From niwot.colorado.edu

Knowles, Harpold, et al., (in review), Hydro. Proc.

Abiotic: more efficient streamflow generator, less sensitive to climate

Biotic: less efficient streamflow generator, more sensitive to climate

What can we learn from natural experiments to answer understand paradoxes and tradeoffs in vapor

losses from snow-dominated systems?

Forest Disturbance in the Southern Rockies: A Natural Experiment

Can we use forest disturbance to learn about: Tradeoffs between

interception and snowpack sublimation?

Tradeoffs between abiotic and biotic vapor losses?

Denver, CO

MPB impacts Chimney Park, WY

Heavy fire impacts Las Conchas, NM

Impacted study catchments

Effects of Disturbance: Expectations from Mountain Pine Beetle (MPB)

(weeks)

Gray Attack

Interception

(months) (years)

Hyd

rolo

gic

Par

titio

ning

Ano

mal

y E

nerg

y A

nom

aly Radiationsw

Wind

Courtesy: J. Biederman

Transpiration

Hypothesis 1: Larger Snowpacks

Hypothesis 2: Increased Streamflow

CanEvap Soil Evap

Total ET

0

200

400

600

CLM4 Noah CLM4 Noah

Sensi&vity to Vegeta&on Change Chimney Park, WY

Impacted LAI=1

Healthy LAI=4

Water (cm)

Courtesy of D. Gochis, NCAR

LAI=1

LAI=4

Eects of Disturbance: Model Fidelity Model experiment using two land-surface models

CLM Noah

Different total partitioning of vapor losses between models

0

5

10

15

20

New

sno

w e

vent

(cm

)

Healthy Post-fire

Eects of Disturbance: Tradeos Between IntercepIon and Snowpack SublimaIon

Individual snow event shows evidence of interception changes following disturbance

0

10

20

30

40

Healthy Pre-Fire

Wat

er (c

m)

Winter P Maximum SWE

SWE:P = 68% SWE:P = 67%

0

5

10

15

20

25

Healthy Post-fire

Wat

er (c

m)

Winter P Maximum SWE


0"

10"

20"

30"

40"

50"

Healthy" MPB"Die4O"

Water"(cm)"

April"2011"Snow"Survey"

SWE:P"="74%""

SWE:P"="69%""

Winter precipita&on (cm)

0"

10"

20"

30"

40"

50"

Healthy"MPB"Die4O

"

Water"(cm)"

April"2011"Snow"Survey"

SWE:P"="74%

""

SWE:P"="69%

""

Peak SWE (cm)

Surprisingly, peak snowpacks did not change after disturbance

SWE:P = 56% SWE:P = 62%

2012 POST-FIRE (survey) 2011 POST MPB (survey) Biederman et al., 2015, Ecohydro. Harpold et al., 2015, Ecohydro.

MPB below/above 12% 20% Healthy below/above 9% 11%

Site

Wind Speed (m/s)

Rsw (W/m2)

NWT Above Canopy Sub-canopy

CP Above Canopy

Sub-canopy

4.1 (2.4) 0.38 (0.15)

3.6 (1.3) 0.43 (0.30)

131 (65) 14 (9)

126 (65) 25 (15)

Change from sublimation in canopy (Healthy) to sublimation of the snowpack (Disturbed)

Energy to snowpack increased following disturbance

Biederman, Harpold, et al., WRR

0%

10%

20%

30%

40%

50%

2010 2011 2012

Win

ter V

apor

Flu

x/Pr

ecip

itatio

n (c

m/c

m)

Healthy MPB die-off


Larger snowpack sublimation post-disturbance

Snowpack sublimation compensates for lower interception (total vapor losses still 30-45%)

Canopy sublimation


Healthy Forest

Canopy sublimation

Post-Fire Forest



Eects of Disturbance: Tradeos Between AbioIc and BioIc Vapor Losses

~50% of water budget to summer vapor loss

Similar cumulative vapor losses in post-disturbance forest

Biederman, Harpold, et al., 2014, WRR

0%

10%

20%

30%

40%

50%

60%

70%

80%

2010 2011 2012

Sum

mer

Vap

or F

lux/

Prec

ipita

tion

(cm

/cm

)

Healthy MPB die-off

Effects of Disturbance: Tradeoffs Between Abiotic and Biotic Vapor Losses

Evidence of kinetic fractionation indicative of evaporation ONLY in disturbed sites


0%

10%

20%

30%

40%

50%

2010 2011 2012

Run

off E

ffici

ency

(cm

/cm

)

Healthy Q* MPB die-off Q* MPB die-off Q


No evidence for increased streamflow using both measured streamflow (Q) and water balance approach (Q*=P-ET)



Tradeoffs between interception and snowpack sublimation Increased energy inputs to

under canopy snowpacks

Potential sources of growing season vapor losses: Greater soil evaporation Compensation by trees Recovering vegetation

Soil evaporation

Remaining forest

Regrowth

Tradeoffs in Abiotic and Biotic Vapor Losses: Disturbed Forests

Eight watersheds in Colorado were investigated following severe MPB disturbance

Biederman, Harpold, et al., (in review, WRR)

Tradeoffs in Abiotic and Biotic Vapor Losses: Disturbed Forests

We infer abiotic-mediated vapor losses mediate decreases in transpiration using three different methods

Biederman, Harpold, et al., (in review)

Only significant changes were towards less runoff

following disturbance

Take Home Points Snowmelt effectively infiltrates the soil profile

thus maximizing storage (for transpiration) and water subsidies (for runoff)

Tradeoffs between interception and snowpack sublimation depend strongly on climate and vegetation structure

In semi-arid climates (i.e. Rocky Mountains) abiotic-mediated vapor losses are likely compensating for changes in biotic-mediated vapor losses following disturbance

Questions and Comments

References Berghuijs, W. R., Woods, R. A., & Hrachowitz, M. (2014). A precipitation shift from snow towards rain

leads to a decrease in streamflow. Nature Climate Change, 4(7), 583-586. Hu, J. I. A., Moore, D. J., Burns, S. P., & Monson, R. K. (2010). Longer growing seasons lead to less

carbon sequestration by a subalpine forest. Global Change Biology, 16(2), 771-783. Goulden, M. L., & Bales, R. C. (2014). Mountain runoff vulnerability to increased evapotranspiration with

vegetation expansion. Proceedings of the National Academy of Sciences, 111(39), 14071-14075. Harpold, A.A. and N.P. Molotch. Timing of snowmelt differentially influences soil moisture response in

Western U.S. mountain ecosystems. Knowles, J., A.A. Harpold, et al. The relative contributions of alpine and subalpine ecosystems to the

water balance of a mountainous, headwater catchment in Colorado, USA

Broxton, P., A.A. Harpold, J. Biederman, P.D. Brooks, P.A. Troch, &N.P. Molotch. (2015) Quantifying the effects of vegetation structure on wintertime vapor losses from snow in mixed-conifer forests. Ecohydrology. doi: 10.1002/eco.1565

Harpold, A.A., J. Biederman, K. Condon, M. Merino, Y. Korganokar, T. Nan, L.L. Sloat, M. Ross, and P.D. Brooks. (2014) Changes in winter season snowpack accumulation and ablation following the Las Conchas Forest Fire. Ecohydrology. 7: 440-452. doi: 10.1002/eco.1363.

Biederman, J.A., A.A. Harpold, D. Reed, D. Gochis, B. Ewers, E. Gutmann, & P.D. Brooks. (2014) Increased evaporation following widespread tree mortality limits streamflow response. Water Resources Research. 50, 53955409, doi:10.1002/2013WR014994.

Biederman, J., P.D. Brooks, A.A. Harpold, D. Gochis, E. Gutman, D. Reed, E. Pendall, & B. Ewers. (2014) Multi-scale Observations of Snow Accumulation and Peak Snowpack Following Widespread, Insect-induced Lodgepole Pine Mortality. Ecohydrology. doi:10.1002/eco.1342.

Biederman, J., Somor, A., A.A. Harpold, et al. Streamflow response to insect-driven tree mortality in subalpine catchments.

2015 CUAHSI Spring Cyberseminar Series - Adrian Harpold

Documents

streamflow snow

water balance of snow

biotic water demand

streamflow generation

et q wet year p

efficient streamflow

snow smaller

overall water budgets