1 Nicholas School of the Environment and Earth Sciences, Duke University 2 Department of Civil and Environmental Engineering, Duke University Bringing.

1Nicholas School of the Environment and Earth Sciences, Duke University

2Department of Civil and Environmental Engineering, Duke University

Bringing Photosynthesis to the Atmosphere: a feedback on terrestrial

water cycling

Gabriel Katul1,2 & Ram Oren1

Photosynthetic Apparatus (~3.5 Billion years)

Land Plants (~350 million years; need for hydraulic apparatus)

TIME in BILLIONS OF YEARS

From D. Christian (2004)

Background:

The evolution of land plants required hydraulic adjustments that permitted the display of the photosynthetic machinery within a desiccating atmosphere.

Background:

A productive display of this machinery necessitates a vertical distribution of chlorophyll so that the light-use efficiency of the organism is increased relative to a concentrated display of chlorophyll.

RandomPorousMedia

CrownClumping

From C. Song

Background

This vertical distribution required certain “engineering solutions” to permit water to be supplied to leaves enclosed in an envelope that permits a controlled exchange of CO2 for

water vapor

Question

Is there some connections between these ‘engineering solutions’ and the photosynthetic properties of the plant?

Contemporary approaches neglect this connection.

Introduction – 1:

There is growing evidence that ecophysiological properties are correlated to hydraulic properties.

Brodribb and Feild (2000) reported strong correlation between maximum carboxylation capacity and stem hydraulic conductivity.

Background - 1:Field Experiments

Data include conifers, vessel-less and vessel bearing taxa from New Caledonia and Tasmania (data from Brodribb and Feild, 2000)

Leuning (2002)

Background – 2:Field Experiments

• Numerous ecophysiological studies reported linkages between canopy photosynthesis and conductance.

• Many hydraulic studies reported linkages between plant hydraulics and photosynthesis (e.g. Hubbard et al., 2001).

Relationship between stomatal conductance (gs) and photosynthesis (A) and plant hydraulic conductivity (KL).

Data from Hubbard et al. (2001)

Background – 4:Missing Link

To date, no analytic framework has been developed to relate hydraulic and ecophysiological properties.

This study proposes an “equilibrium hypothesis” for coupled CO2 & H2O to arrive at analytic expressions between soil-plant hydraulics and ecophysiological

properties

Hypothesis - 1:

The basic hypothesis is that hydraulic and eco-physiological properties co-evolve at time scales relevant to stand development.

Hypothesis – cont’d:

• The hydraulic and ecophysiological parameters

evolve so that the maximum biochemical demand

for carbon uptake is in equilibrium with the

maximum carbon gain permissible by the soil-root-

xylem hydraulics.

Hydraulic Supply: Water Transport in Plants

resr

esw rr

J

Soil Water Status

Soil Water CharacteristicCurve

Soil Hydraulic Conductivity Function and RAI

Hydraulic Supply: Water Transport in Plants

e

reG

tl

Onset of Cavitation

Root-to-Epidermis [Simplest Model]

Maximum Hydraulic Capacity is achieved when:

tle

maxGGre

Maximum Hydraulic Conductivity

From Hacke et al. (2000): Pinus taeda

Sperry et al. (2000) demonstrated that plant pressure is close the cavitation threshold within the root-xylem system.

tle

Stomatal Closure – and Xylem Pressure

Stomatal closure observed during drought is a result of a decline in plant hydraulic conductance in the xylem of the roots.

This can explains why stomatal conductance shuts down at suction levels well below the turgor suction.

Vascular Plants

New processes associated with the evolution of a larger

and more complex plant body became more limiting to

gas exchange than those occurring within individual

cells (Sperry, 2000; Katul et al., 2003)

Gas Exchange - Stomata

ss DgLAIE 6.1

Water Loss from Epidermis to Atmosphere:

Setting E = Jw (Steady State) and noting that:

is

ns CC

Ag

s

is

resr

tlsn D

CC

rrLAIA

6.1

1

min,

Hence, the maximum hydraulic supply of CO2 (per LAI)

d

i

in R

C

CA

2

*1

Biochemical Demand for CO2

Farquhar et al. (1980):

Equate maximum hydraulic supply to maximum biochemical demand and solve for Ci

2

*1

min, 6.1

1

i

i

s

ia

resr

tlsn C

C

D

CC

rrLAIA

200 m

Duke Pine Forest

Implications to FACE: Conductance/Down-regulation

Reference State: The nothing changed except atmospheric CO2 scenario:

amba

amb

ambambc

elva

elv

elvelvc

ambamba

ambs

elvelva

elvs

ambn

elvn

C

V

C

V

Cg

Cg

A

A

2

max,

2

max,

1

1

15.1max,

max, amb

c

elvc

V

V

Implications to FACE

Rogers and Ellsworth (2002) report a reduction in Vcmax of 25% (for older foliage). Photosynthesis-weighted Vcmax

drops by ~ 17%.

Schafer et al. (2002) – bulk canopy conductance increased by ~ 25% but no change in LAI.

0.6

0.8

1.0

1.2

1.4

1.6

1997 1998 1999

Rat

io o

f ele

vate

d-t

o-am

bien

t

0.6

0.8

1.0

1.2

1.4

1.6

Normalized conductance Normalized Ci/Ca

Ellsworth

Schäfer

Vcmax

1997 1998 1999

An

To Match a 17% reduction in Vcmax and a 1.38 increasein bulk stomatal conductance,

Ci/Ca must increase by 1.21 (~ 0.67 0.80).

Experiment: SETRES II

SouthEast Tree Research and Education Site (SETRES) – 2.

A large-scale genotype nutrition interaction experiment designed to quantify the effects of fertilization on C- cycling in a managed southern pine forest in North Carolina

(operated by the U.S. Forest Service).

SETRES I & II Fertilization Experiment

After 6 years of fertilization

Fertilization Rate (~11 g N m-2 y-1):

Leaf Area Index Doubled (1.65 3.51)

Maximum Leaf Carboxilation Capacity Increased

by about 20%.

Respiring Biomass Increased by ~ 48%.

Case Study: Duke Forest and SETRES

Mode of soil moisturepdf’s during growingSeason.

SETRES ~ 0.09

DUKE FOREST~0.2

Stable Isotope MeasurementsCi/Ca ~ 0.66

Fertilization Increases Vcmax

Leaf Nitrogen

)2(

)1(

)1(

)2(

)2(max,

)1(max,

)1(

)2(

)2(max

)1(max

ref

ref

s

s

c

c

g

g

LAI

LAI

g

g

LAI

LAI

V

V

Effect of N-Fertilization on Vcmax (SETRES –I & II)

18.150

125

51.3

65.1)(

max

)(max c

c

fc

V

V

Using LAI (from SETRES – II) and sapflux based measurements (from SETRES – I ) of

17.14.852.100)(max

)(max c

cf

c VV

Gas Exchange Measurements (SETRES – II):

refg

General Remark:

Equilibrium model provides some constraints on how shifts in plant hydraulic and physiological properties affect photosynthesis and water uptake.

Hydraulic shifts already measured in the Prototype (Schafer, Oren) and physiological shifts reported in FACE, the next logical question is whether N x CO2 effects can be predicted by the equilibrium model.

Engineering Solutions – Climate /Reproduction?

Emergence of land plants from aquatic habitat also required engineering solutions for re-production (mainly - pollen and its dispersal).

Is there some connection between the ‘engineering solutions’ for plant hydraulics and pollen dispersal?

Timely Topic: Pollen Dispersal of GM Trees

Increased use of GM trees is now being considered and debated within the policy arena.

The “finger-print” of this increase is perhaps evident in the number of permit applications to the USDA-APHIS, now exceeding 150 types of transgenic trees (Mann and Plummer 2002).

Quantifying GM pine pollen and seed dispersal pattern is a necessary first step to assessing gene flow, landscape change, and other unforeseen ecological risk – especially with elevated atmospheric CO2 (LaDaux and Clark, 2001).

x=LongitudinalDistance (m)

y=LateralDistance (m)

z=Vertical Distance (m)

Uplifting

Mean WindDirection

d

d

Pr(d)DispersalKernel

List

Litu

Reproductive (16 years) and Harvesting Ages (25 years)are used in model calculations. Stand attributes are based on Leaf area measurements and site index for the stand at Duke Forest.

POLLEN SIMULATIONS FOR PINE PLANTATIONS

Duke Forest, NC

Dispersal Kernels for the two stand ages

Kernel calculations excluded pollen that escaped the atmospheric boundary layer top or experienced travel times in excess of 1 hour.

Dispersal Kernels Excluding Pollen that Escaped the Canopy Volume

Oilseed rape Frequency of hybrid formation (GM and non-GM) was 0.156% at 200 m

Maize (corn ) Most pollen fell within 5 m, 98% of the pollen remained within a 25–50 m.

Cultivated rice Pollen horizontal movement was limited to 10 m.

Potato Isolation distance of 20 m appears adequate for transgenic potatoes. However, maximum distances for which pollen particles were detected is 80 m.

Cotton Pollen trapped decrease to 0.03% at 50 m from the source.

Apples Maximum distance pollen particles were detected is 56 m

Crop Comments

Wheat Maximum distance pollen were detected is 20 m

Biocontainment zones for few GM crops

1. Conifer pollen is NOT likely to be of negligible viability at those distances (viability diminishes due to excess UV-B, cold air temperatures, or dehydration).

2. Given the long dispersal distances reported here, a regulatory framework that distinguishes between annual (and perennial) crops and forests is only logical.

Results

Broader Conclusions

The evolution of land plants required hydraulic adjustments that permitted the display of the photosynthetic machinery within a desiccating atmosphere.

Does this ‘equilibrium’ between plant hydraulic and photosynthetic properties also govern reproductive strategies?