Arid Ecohydrology Ecohydrology - PowerPoint Presentation

Slide1

Arid Ecohydrology

Ecohydrology

Fall

2017Slide2

A Global Issue…

Huge extent (40% of continental area)

Generally low rainfall (seasonal or persistent dryness governs plant-soil relations)

Significant human dependence and vulnerability (drought)Slide3

A Class Unto Itself…

Deep dependence on soil moisture

Soil physics

FirePlant adaptation & competition

N availabilityRoot distributions

Regular pattern developmentFeedbacks with animals (e.g., fairy rings)Slide4

Ecosystem RUE

All terrestrial ecosystems use MAP to create NPP, and are therefore, at some level, water

dependent

Differential sensitivity to variance in MAP

NPP:MAP defines rain use efficiency (RUE)

Sensitivity effectively measures the strength of water limitation vis-à-vis nutrient or evolutionary constraints

Huxman et al. (2004) - NatureSlide5

Convergent RUE

max

?

RUEmax generally occurs at low water availabilityThere is a convergence of

RUEmax across all biomes

RUEmax is close to

RUE

mean

for arid sites

RUEmax

is really low compared to RUEmean for humid sitesSlide6

Two Predictions

1 – NPP will be affected a lot if

PPT

is driven below the historic minimum

2 – Removal of other resource limitations will allow RUE to approach RUEmaxSlide7

Ergo

Water limitation imposes a common constraint on NPP across biomes

Ecosystems have the same

RUEmax despite large differences in MAP, physiology,

phylogenetic origin and climate historyAltering resource limitation underscores the relevance of biogeochemistry on NPP (i.e., compared with species)Slide8

Inhibited Deep Drainage Over Millenia

In dry areas (<500 mm) there is a nearly ubiquitous pore-water Cl

-

peak at 5-15 m below gradeMass balance of Cl yields 10,000-15,000 years of accumulation (without major downward flux event)

Consistent with ~ 1 mm/yr deep drainage

Seyfried et al. (2005) - EcologySlide9

But…

Upward water potential gradient

Water is moving out of the soil

Ergo, no downward movement

Seyfried et al. (2005) - EcologySlide10

Strong Climatic Control

Cl inventory suggests 7,000-8,000 years of accumulation at a site with 360 mm MAP

Inventory is 1,000 years at 400 mm MAP at nearby siteSlide11

Ecohydrologic Mechanism

Walvoord

et al. (2002) propose a conceptual model to reconcile these attributes

Low water potentials at the base of the root zone were established at the end of the Pleistocene (dramatically increased dryness)

Potentials (< -1 MPa) were maintained continuously during the Holocene

Any deep wetting event would reset the chloride concentrations

Slow upward movement of

water above ca. 20 m

Drainage below 20 m of Pleistocene aged water

Simultaneous recharge and upward

flux above 20 mSlide12

Vegetation is Essential

Plants are required to impose and maintain the low potentials

Two key attributes:

Deep roots (2-3 m) of sufficient density to capture all downward percolating water

Able to maintain low water potentials at depth continuouslyThe demand for continuous maintenance has enormous implications for contemporary vegetation management

Clearing vegetation can result in huge fluxes of Cl (and nitrate!) to the deep groundwaterSlide13

Deep Roots

Root density decreases exponentially with depth, though slower in arid biomes

Max rooting depths ~ 5 m, almost all co-dominant shrubs ~ 1.8 m

Root depths increased with MAP and soil coarseness

Maintaining root potentials demands cavitation avoidanceGeometry of xylem

Soil hydraulic conductance goes to 0SW USA co-dominants can withstand cavitation (embolism) from -4 to -9

Mpa

(

extremely

low)

The conceptual model requires sustained -1 MPa at the root zone

Sperry et al. (2002) – Functional EcologySlide14

Hydrologic Buffering

Arid lands are hydrologically variable

Model requires constancy

As such, moisture conditions at 2-3 m below grade must be buffered from the surface variability

Water budget suggests that Dd (deep drainage) occurs only when I > (A

ET + ΔS)

Δ

S

max

is ~ 350 mm (over 3 m root depth)Dd occurs only when I is 350 mm > A

ET (basically never)Arid land plants use

excess water in wet yearsMultiple wet years create rapid vegetation adjustmentsExpansion of deep rooted shrubsExplosion of annuals

As such, the base of the root zone basically are continuously water starved

So why invest in deep roots?Slide15

Deep Roots

C expenditure that needs to be net benefit

Existing profile data suggest very little plant available water below 60 cm

31

yr data record at 2 wk interval

Living roots at 180 cmLow potentials (< -1.5 Mpa)

Net energy benefit?

Plants may actually be hydrologically subsidizing deep roots to maintain their viability during wet periods

Confers benefits during periods of deep drought

Seyfried et al. (2001) – WRRSlide16

Implications

Natural deep rooted perennial communities are most frequently replaced with shallow rooted annuals

Grazing, planting

Changes episodic downward fluxes of water and therefore salt, nutrients, hazardous materialsExtremely difficult to reestablish deep rooted vegetationAdaptive significance of this “ecological inheritance”

Low water potentials, high Cl peaksSlide17

Alternative Stable StatesSlide18

What is a “State”

A state is a configuration of a community of organisms that can be described by a set of (dynamic) “state” variables

Biomass, density, species (ages, guilds, abundances), nutrients/organic matter

A “stable state” is a particular configuration that possesses self-reinforcing feedbacks that confer some resilience in the presence of perturbationsSlide19

Origins of the Idea

Stability of populations (constant environment)

“If the system of equations describing the transformation of state is nonlinear…there may be multiple stable points with all species present so that local stability does not imply global stability”

(

Lewontin 1969)

Overfishing, invasions, predator removalStability of ecosystems (variable environment)

(Lord) Robert May (1974)Slide20

Stable and Non-Stable Equilibria

dP

/

dt

= B(P) – D(P)

D = a*P + c

B = q + K/(1+e

-rP

)

Solve for when

dP

/dt = 0or B(P) = D(P)Slide21

The Demographic Transition

Introduces ideas:

Equilibria

Disequilibrated systemsTime lags between statesPath dependency?Slide22

Demographic Stability – Allee Effects

Per capita birth rates decline at low density

Mate finding, dispersal

Creates two population equilibriaWhich is stable? Unstable?Slide23

Balls-in-cups

Ball represents an ecosystem state

Cup represents the domain (or basin) of attraction

Remember that there are feedbacks that create “resilience”

Regime shifts occur by:Perturbation to the variables (i.e., move the ball)

Perturbation to the “fitness landscape” (i.e., change the basin)Slide24

Conceptual Model – Stability LandscapesSlide25

Ecosystem Response to Change

a. One equilibrium for each environmental condition

b. One equilibrium, but rapid rates of change over short span of conditions

c. Three equilibria can persist for one state (two stable, one unstable), strong

hysteresis (path dependency, temporal lags) in the transition between statesSlide26

Resilience – A Related Concept

Two aspects of the basin of attraction matter

Width of stability domain [Over what range of conditions/perturbations will the community recover?]

Steepness of the stability domain [How quickly will the system recover?]

Resilience can be used to define both properties The former is dubbed “ecological resilience” the latter “engineering resilience”Basically some measure of a systems capacity to buffer the effects of exogenous changesSlide27

Hysteresis – A Key Emergent Property

A systems trajectory over time may depend on its current and historical position

Changes in system configuration may delay or even prevent recovery

Suding et al. (2005)

Trends in Ecology and EvolutionSlide28

Why Should We Care?

Costly surprises (catastrophic shifts, weak ecosystem responses to management)

Collapse of fishery stocks

Outbreaks of disease

Exotic species invasionsEcosystem shifts and loss of services

Underscore the complex, contingent (dependent of antecedent conditions) nature of environmental systemsThresholds imply the need for caution when dealing with complex systems

Climate change, biodiversity loss, stock markets, land use changeSlide29

Tidal Marsh Alternative EquilibriaSlide30

Example: Shallow Lakes

Submerged Aquatics

Rooted plants little phytoplankton

Low turbidity due to:Plant filtrationHigh sediment cohesion limits resuspension due to wind action

Algal-dominatedPhytoplankton shades SAVHigh turbidity

Reduced filtrationLoose, flocculent sediments that can be easily wind-entrainedSlide31

Empirical Evidence

Enrichment of P leads to increased production

Eventually there is sufficient water column P to start to erode the controls exerted by SAV

Catastrophic flip (short time) to phytoplankton dominance

Reductions in P necessary to reverse the process are much largerSlide32

Incipient Shifts – Where are the Tipping Points?

Key problem:

It would be nice to know when these are about to happen

Interesting theoretical wrinkles:

Are the conditions for incipient regime shift the same?Volatility? Stability?

Of sufficient interest to be the subject of an NPR pieceSlide33

What Kinds of Incipient Shifts?

Multiple settings

Medicine: Onset of asthma and epileptic seizures

Finance: Stock market crashesAgriculture: Drought-society Environment: Ocean circulation, fish stocks, rangeland woody cover, boreal climate feedbacks

“Canaries in the coal mine” – predicted catastrophic bifurcationsVolatility/varianceSkewnessSlide34

Critical Slowing

At critical thresholds, small changes in state or conditions can induce a dramatic shift

That is, the system becomes increasingly “slow” in recovering

Scheffer et al. (2009)

NatureSlide35

How Do You Measure Slowing Down?

Response to experimental perturbations

Measure the rate of return

Impossible to control in large systems, but these are always perturbed

Measure rate at which system fluctuates around the meanAre the fluctuations autocorrelated

?Are they increasing in time?Are they changing in their statistical properties?

Scheffer et al. (2009)

NatureSlide36

Autocorrelation, Variance and Skewness

Guttal and Jayaprakash 2008

Ecology Letters

Scheffer et al. (2009)

NatureSlide37

Scheffer

et al. (2009)

NatureSlide38

Confronting Theory With Data

Detecting alternative attractors (observational)

Time series shifts

Bimodality in system states (across systems)

Dual relationships to a given control factor

Scheffer and Carpenter (2003)

TREESlide39

Confronting Theory with Data

Experimentally

Different initial conditions lead to different final states

Disturbance triggered transitions

Path dependency (hysteresis)

Scheffer

and Carpenter (2003)

TREESlide40

Alt. Stable State – Desert Streams

Wetlands used to be a major component of desert streams (Sycamore Creek, AZ)

Reduced to zero over the late 19

th and 20th C.

Recently become reestablished with controls on grazingSlide41

Basic Mechanism

Wetland plants control sediment dynamics

Density dependent stabilization

Floods and drying are the major hydrologic disturbance in deserts

This makes areas with dense vegetation better able to persist through floods and dry periods, and areas without vegetation less ableTwo predicted states – vegetated and gravel bed

Region of global bi-satbility defined by:

G – veg growth

S – vegetation mortality

K

s

– channel sediment stability w/o veg

C

s

– per capita stabilization of sediment

r

s

– scour vegetation mortality

V – vegetation

Q – flood frequencySlide42

Empirical Support

Water and Vegetation cover over time

Major Resetting FloodSlide43

Vegetation Effects on Vegetation Loss

(the

biogeomorphic

feedback)Vegetation density affects cover loss

Stream sites self-organize into 1 of 2 modesSlide44

Idle Restoration Thoughts

Two strategies for restoration

Change conditions (e.g., reduce P)

Force state (e.g., massive disturbance)Are there other alternative stable states lurking?Slide45

Management and A.S.S.

Native SAV

Dominated

Hydrilla

Dominated

Blue-Green

Dominated

Ecological release

Disturbance

Changes in herbivory

Nutrient enrichment

Physical removal

Chemical removal

?

Should massive disturbance be part of the restoration tool-box?

Are there states less desirable than the one we’re trying to restore?Slide46

In a Spatial Domain…(Next Time)