Western
Mountain
Initiative
The climatic water
balance in an
ecological context
Nathan L. Stephenson
U.S. Department of the Interior
U.S. Geological Survey
Thesis: Consideration of the climatic water balance
improves our ability to understand and forecast the
effects of climatic changes on the biological world.
(1) The superiority of the climatic water balance
in ecological studies.
(2) In a changing climate, increasing water availability
will NOT counteract the effects of increasing
evaporative demand. … And why soil moisture
may be less important than you think!
(3) Don’t forget AET!
Thesis: Consideration of the climatic water balance
improves our ability to understand and forecast the
effects of climatic changes on the biological world.
(1) The superiority of the climatic water balance
in ecological studies.
(2) In a changing climate, increasing water availability
will NOT counteract the effects of increasing
evaporative demand. … And why soil moisture
may be less important than you think!
(3) Don’t forget AET!
Mean annual temperature (C)
Mean annual precipitation (cm)
Whittaker 1975
Traditional view: At broad scales, vegetation type is a function
of some measure of annual energy supply, annual water supply,
or their ratios or differences.
Mean annual temperature (C)
Temperature
Potential evapotranspiration (PET)
Net radiation
Growing degree-days
Precipitation (P)
P/PET (or PET/P)
PET-P (or P-PET)
Mean annual precipitation (cm)
Whittaker 1975
Traditional view: At broad scales, vegetation type is a function
of some measure of annual energy supply, annual water supply,
or their ratios or differences.
The Big Problem: The traditional view does not consider
the seasonal interactions of energy and water. E.g., with
nearly identical mean annual temperature and precipitation:
Deciduous broadleaved forest
Evergreen needleleaved forest
Portland, Oregon
Erie, Pennsylvania
180
120
100
80
20
10
0
0
0
20
40
20
0
Precipitation
Temperature
10
30
60
160
140
120
100
80
60
30
40
20
J F M A M J J A S O N D
J F M A M J J A S O N D
Month
Month
Temperature (C)
140
Precipitation
Temperature
Temperature (C)
160
Precipitation (mm)
Precipitation (mm)
180
(1) By explicitly incorporating the seasonal interactions of
energy and water, the climatic water balance is
biologically more meaningful than traditional climatic metrics.
150
mm Water
Potential
evapotranspiration
Water supply
(rain plus snow
melt water)
100
50
0
1
J
F2
3
M
4
A
5
M
J6
7J
Month
8
A
9
S
10
O 11
N 12
D
Actual evapotranspiration (AET) reflects the simultaneous
availability of biologically usable energy and water, and is
therefore an index of site potential for productivity.
150
mm Water
Potential
evapotranspiration
Water supply
(rain plus snow
melt water)
100
S
D
50
AET
0
1
J
F2
3
M
4
A
5
M
J6
7J
Month
8
A
9
S
10
O 11
N 12
D
Climatic water deficit (D) is evaporative demand that is not met
by available water, and is thus an index of potential effects of
drought stress on plants. (Don’t confuse with soil water deficit!)
150
mm Water
Potential
evapotranspiration
Water supply
(rain plus snow
melt water)
100
S
D
50
AET
0
1
J
F2
3
M
4
A
5
M
J6
7J
Month
8
A
9
S
10
O 11
N 12
D
Surplus is an index of biologically unusable water –
excess water that leaves a site, through runoff or subsurface
flow, without being evaporated or transpired.
mm Water
150
100
S
D
50
AET
0
1
J
F2
3
M
4
A
5
M
J6
7J
Month
8
A
9
S
10
O 11
N 12
D
(2) The climatic water balance (AET and D) is better
correlated with the distribution of vegetation types.
Mean annual precipitation (mm)
Example: North American coniferous and deciduous forest distributions.
2500
2000
1500
1000
500
Deciduous forest
Coniferous forest
0
0
5
10
15
20
Mean annual temperature (C)
(2) The climatic water balance (AET and D) is better
correlated with the distribution of vegetation types.
2500
2000
1500
1000
500
Deciduous forest
Coniferous forest
0
0
5
10
15
20
Mean annual temperature (C)
Annual actual evapotranspiration (mm)
Mean annual precipitation (mm)
Example: North American coniferous and deciduous forest distributions.
1400
1200
1000
800
600
400
200
Deciduous forest
Coniferous forest
0
0
100
200
300
400
Annual climatic water deficit (mm)
Stephenson 1998, J. Biogeogr.
(3) The climatic water balance exhibits these properties
across more than ten orders of magnitude of spatial scale.
North America
Stephenson 1990, Am. Nat.
Stephenson 1998, J. Biogeogr.
(3) The climatic water balance exhibits these properties
across more than ten orders of magnitude of spatial scale.
Sequoia Nat’l Park
Stephenson 1998, J. Biogeogr.
Summary -- the superiority of the climatic water balance
in ecological studies:
• More biologically meaningful.
• Better correlated with biological processes and
vegetation and species distributions.
• Exhibits these properties across >10 orders of
magnitude of spatial scale.
Thesis: Consideration of the climatic water balance
improves our ability to understand and forecast the
effects of climatic changes on the biological world.
(1) The superiority of the climatic water balance
in ecological studies.
(2) In a changing climate, increasing water availability
will NOT counteract the effects of increasing
evaporative demand. … And why soil moisture
may be less important than you think!
(3) Don’t forget AET!
Mean annual temperature (C)
At continental and global scales,
most authors have treated metrics
of energy and water supplies as
orthogonal axes. That is, they
recognize that water does NOT
counteract the effects of energy.
Mean annual precipitation (cm)
Whittaker 1975
Elevation
Yet at local to regional scales,
some of those same authors
additively combine metrics of
energy and water supplies
into a single “moisture scalar,”
thus assuming that the
effects of water and energy
counteract one another.
E.g., Whittaker’s moisture
scalar combines metrics
related to evaporative demand
(slope aspect and exposure)
with metrics related to water
availability (slope position and
proximity to water).
“Topographic moisture gradient”
Whittaker & Niering 1965, Ecology
For understanding and forecasting plant distributions, we care less about
soil moisture than the dynamics that gave us that soil moisture.
“Normal”
conditions
Plant
response
Wet
Dynamics
of evap. demand
and water supply
Dry
Water level
For understanding and forecasting plant distributions, we care less about
soil moisture than the dynamics that gave us that soil moisture.
Drought
(low precipitation)
Wet
Dry
For understanding and forecasting plant distributions, we care less about
soil moisture than the dynamics that gave us that soil moisture.
Add
water
Wet
Dry
For understanding and forecasting plant distributions, we care less about
soil moisture than the dynamics that gave us that soil moisture.
Normal precip, but high
evaporative demand
Wet
Dry
For understanding and forecasting plant distributions, we care less about
soil moisture than the dynamics that gave us that soil moisture.
Add
water
Wet
Dry
Consider these scenarios in terms of annual AET and D.
The colored dots represent species’ climatic niches;
dotted diagonal are isolines of constant PET.
Annual AET (mm)
200
100
0
0
100
200
Annual Deficit (mm)
300
Because PET = AET + D, drought (low precipitation)
can only result in increases in D that are precisely offset
by identical declines in AET.
Annual AET (mm)
200
100
0
0
100
200
Annual Deficit (mm)
300
Adding the right amount of water can put the site
back to exactly where it started in climate space.
Annual AET (mm)
200
100
0
0
100
200
Annual Deficit (mm)
300
In contrast, because PET = AET + D, increasing PET must
increase AET, D, or both. That is, the effects of changing PET
are nearly orthogonal those of changing water availability …
Annual AET (mm)
200
100
0
0
100
200
Annual Deficit (mm)
300
… so that adding water CANNOT put the site
back to exactly where it started in climate space ...
Annual AET (mm)
200
100
0
0
100
200
Annual Deficit (mm)
300
… and big enough increases in PET and water availability
will move the site to another species’ climatic niche,
even though soil moisture hasn’t changed.
Annual AET (mm)
200
100
0
0
100
200
Annual Deficit (mm)
300
What are some implications for the effects of
climatic changes on mountain ecosystems?
Example: Unlike temperature changes, precipitation changes
should not be expected to cause coordinated directional shifts
in species’ elevations.
PET declines sharply with increasing elevation.
Evaporative rate (cc/day)
80
Santa Catalina Mtns, AZ,
summer of 1911
60
40
20
South-facing slopes
North-facing slopes
0
1000
1500
2000
2500
Elevation (m)
Data from Shreve 1915
A spherical porous
cup atmometer
Livingston 1935, Ecology
PET declines sharply with increasing elevation.
Lo
w
at
io
n
El
ev
h
100
H
ig
Annual AET (mm)
200
0
0
100
200
Annual Deficit (mm)
300
Thus, temperature-induced increases in PET will alter a site’s water
balance to support species currently found at lower elevations.
This is why we expect temperature-induced upslope migrations.
Annual AET (mm)
200
100
0
0
100
200
Annual Deficit (mm)
300
At a given elevation, local water availability depends
on things like soil depth, proximity to water, etc.
(e Hi
.g g
so . de h
ils ep
)
Annual AET (mm)
200
Lo
av ca
ai l w
la at
bi e
l it r
y (e L
.g o
.
w
100
so sha
ils llo
) w
0
0
100
200
Annual Deficit (mm)
300
Declining precipitation will alter a site’s water balance to support
species currently found at sites with lower water availability
but at the same elevation …
Annual AET (mm)
200
100
0
0
100
200
Annual Deficit (mm)
300
… and increasing precipitation will alter a site’s water balance to support
species currently found at sites with higher water availability but at the same
elevation. This is why we DON’T expect precip-induced elevational changes.
Annual AET (mm)
200
100
0
0
100
200
Annual Deficit (mm)
300
These expectations are empirically supported
by observations of species’ distributions along
natural climatic gradients.
Stephenson 1998, J. Biogeogr.,
Fites-Kaufman et al. 2007, TVC
Thesis: Consideration of the climatic water balance
improves our ability to understand and forecast the
effects of climatic changes on the biological world.
(1) The superiority of the climatic water balance
in ecological studies.
(2) In a changing climate, increasing water availability
will NOT counteract the effects of increasing
evaporative demand. … And why soil moisture
may be less important than you think!
(3) Don’t forget AET!
Parting thoughts:
“Moisture availability” is a slippery term of limited value,
and can only be understood in two dimensions.
E.g., does an increase in both AET and D yield a wetter
or a drier environment? Arguably, both!
In many (most?) cases it’s better to drop the concept
of “moisture availability” in favor of explicit consideration
of AET and D.
Thank you for your attention!