The Professional Geographer Radial Growth and … growth...Radial Growth and Increased Water-Use Efﬁciency for Ponderosa Pine Trees in Three Regions in the Western United States*

PLEASE SCROLL DOWN FOR ARTICLE

This article was downloaded by: [Soul, Peter T.]On: 16 May 2011Access details: Access Details: [subscription number 937008037]Publisher RoutledgeInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

The Professional GeographerPublication details, including instructions for authors and subscription information:http://www.informaworld.com/smpp/title~content=t788352615

Radial Growth and Increased Water-Use Efficiency for Ponderosa PineTrees in Three Regions in the Western United StatesPeter T. Souléa; Paul A. Knappb

a Appalachian State University, b University of North Carolina Greensboro,

First published on: 28 April 2011

To cite this Article Soulé, Peter T. and Knapp, Paul A.(2011) 'Radial Growth and Increased Water-Use Efficiency forPonderosa Pine Trees in Three Regions in the Western United States', The Professional Geographer,, First published on:28 April 2011 (iFirst)To link to this Article: DOI: 10.1080/00330124.2011.574088URL: http://dx.doi.org/10.1080/00330124.2011.574088

Full terms and conditions of use: http://www.informaworld.com/terms-and-conditions-of-access.pdf

This article may be used for research, teaching and private study purposes. Any substantial orsystematic reproduction, re-distribution, re-selling, loan or sub-licensing, systematic supply ordistribution in any form to anyone is expressly forbidden.

The publisher does not give any warranty express or implied or make any representation that the contentswill be complete or accurate or up to date. The accuracy of any instructions, formulae and drug dosesshould be independently verified with primary sources. The publisher shall not be liable for any loss,actions, claims, proceedings, demand or costs or damages whatsoever or howsoever caused arising directlyor indirectly in connection with or arising out of the use of this material.

http://www.informaworld.com/smpp/title~content=t788352615

http://dx.doi.org/10.1080/00330124.2011.574088

http://www.informaworld.com/terms-and-conditions-of-access.pdf

Radial Growth and Increased Water-Use Efficiency for

Ponderosa Pine Trees in Three Regions in the Western

United States*

Peter T. SouleAppalachian State University

Paul A. KnappUniversity of North Carolina Greensboro

We examined changes in and relationships between radial growth and intrinsic water-use efficiency (iWUE)of ponderosa pine (Pinus ponderosa) trees, climate, and atmospheric CO2 in the western United States sincethe mid-nineteenth century. We developed tree-ring chronologies for eight sites in three climate regionsand used carbon isotope data to calculate pentadal values of iWUE. We examined relationships amongradial growth, climate, iWUE, and CO2 via correlation and regression analyses. Significant upward trends iniWUE occurred at all sites, and despite an absence of climate changes that would favor growth, upward radialgrowth trends occurred at five sites. Our findings suggest that increased iWUE associated with rising CO2can positively impact tree growth rates in the western United States and are thus an evolving component offorest ecosystem processes. Key Words: atmospheric CO2, climate, drought, intrinsic water-use efficiency,Pinus ponderosa.

Examinamos los cambios entre crecimiento radial y la eficiencia intrınseca del uso del agua (iWUE, sigla eningles) del pino ponderosa (Pinus ponderosa), y las relaciones con el clima y el CO2 presente en la atmosferadel occidente de los Estados Unidos, desde mediados del siglo XIX. Desarrollamos cronologıas basadas enlos anillos del tronco de los arboles para ocho sitios de tres regiones climaticas y utilizamos datos de isotoposde carbono para calcular los valores pentadales de la iWUE. Tambien, examinamos las relaciones entrecrecimiento radial, clima, iWUE y CO2, por medio de analisis de correlacion y regresion. En todos lossitios se detectaron tendencias significativas de aumento de la iWUE y a pesar de la inexistencia de cambiosclimaticos que hubiesen favorecido el crecimiento, en cinco sitios se registraron tendencias de aumento enel crecimiento radial. Nuestros hallazgos sugieren que el incremento de la iWUE asociado con elevacion del

∗This project was funded by the USDA NRI Competitive Grant Program, Plant Adaptations to the Environment #2005-35100-15226, andFaculty Research Grants from both Appalachian State University and the University of North Carolina at Greensboro. The EnvironmentalIsotope Laboratory at the Laboratory of Tree-Ring Research, University of Arizona, conducted the isotopic analyses for use under the guidanceof Dr. Steve Leavitt. Steve Shelly, USDA Forest Service Region 1 RNA coordinator, helped us identify study sites and helped us in the field. Alsoassisting in field and laboratory work were Justin Maxwell, Jason Ortegren, Ian Snider, William Tyminski, and Philip White. We thank all ofthese individuals and the two anonymous reviewers of the article.

The Professional Geographer, 63(3) 2011, pages 1–13 C© Copyright 2011 by Association of American Geographers.Initial submission, December 2009; revised submission, April 2010; final acceptance, May 2010.

Published by Taylor & Francis Group, LLC.

Downloaded By: [Soul, Peter T.] At: 17:58 16 May 2011

2 Volume 63, Number 3, August 2011

CO2 pueden impactar positivamente las tasas de crecimiento de los arboles en la region occidental de EE.UU.y son, por lo tanto, un componente en evolucion de los procesos que ocurren en el ecosistema de bosque.Palabras clave: CO2 atmosferico, clima, sequıa, eficiencia intrınseca del uso del agua, Pinus ponderosa.

I n 2008, atmospheric CO2 concentra-tions (hereafter CO2) from the Mauna

Loa, Hawaii, Observatory records exceeded385 ppmv (http://www.esrl.noaa.gov/gmd/ccgg/trends/ [last accessed 1 December 2009]),representing a 22 percent increase since 1959.In forested ecosystems, one of the primaryeffects related to higher levels of CO2 on treegrowth is decreased stomatal conductance(Tognetti et al. 1998). As CO2 has increased,most tree species have been able to use watermore efficiently as they experienced changesin stomatal conductance; that is, leaf stom-atal apertures narrow during photosynthesis(Ainsworth and Rogers 2007). Consequently,the uptake of CO2 during photosynthesis cur-rently and in recent decades may have occurredat rates equivalent to a lower CO2 environmentbut with less transpirational water loss perbiomass gained. Thus, tree growth during pe-riods of temporary moisture limitations in thetwenty-first and late twentieth century shouldbe greater than comparable periods priorto the 1950s because of increased water-useefficiency. Several studies have documentedsignificant increases in intrinsic water-useefficiency (iWUE; i.e., the ratio of net CO2 as-similation to stomatal conductance) for varioustree species in many parts of the world (Bert,Leavitt, and Dupouey 1997; Feng 1999; Tang,Feng, and Funkhouser 1999; Arneth et al.2002; Saurer, Siegwolf, and Schweingruber2004; Waterhouse et al. 2004; Liu et al. 2007),and others (Soule and Knapp 2006; Knapp andSoule 2008) have documented radial growthrate increases and linked these increases tochanging CO2 concentrations, which in turnlikely increased iWUE rates for those species.

For the western United States, the consen-sus of global circulation model predictions isfor substantially warmer and drier summers(Christensen et al. 2007). Increasing aridity cancause a decline in forest productivity (Hansonand Weltzin 2000; Ciais et al. 2005; Breda et al.2006) or continue a trend of increasing treemortality already observed in the region (vanMantgem et al. 2009). Increasing iWUE as-sociated with rising CO2 concentrations, how-ever, could potentially help offset any declines

associated with possible climate changes detri-mental to tree growth and forest health. Ex-amining the historical responses to changingatmospheric conditions such as CO2 will helpus understand how western conifers might re-spond if climate change results in an environ-ment less favorable for growth. In this studywe examined changes in iWUE, radial growthrates, and associated climate and atmosphericvariables for ponderosa pine (Pinus ponderosa)trees growing in three distinct climatic regionsof the western United States. Our objectiveswere to (1) determine if long-term (>100 years)trends exist for ponderosa pine for iWUE rates,radial growth rates, and the primary climaticdrivers for radial growth of ponderosa pine; and(2) examine the relationships among these threevariables.

Methods

Chronology DevelopmentWe selected ponderosa pine because it isan ecologically important species occurringthroughout the American West with a ge-ographical range exceeding 20◦ of latitudeand longitude. Occupying over 10.9 millionhectares (http://www.wpa.org/ppine.htm [lastaccessed 3 March 2010]), ponderosa pine isa major species for U.S. forestry (WesternWood Products Association 2007). Ponderosapine forests are among the most valued inNorth America because of the diverse ben-efits they provide including “timber, for-age, wildlife habitat, recreation . . . and scenicbeauty” (Youngblood 2005, 49–50). Finally,ponderosa pine trees grow within semiarid siteswhere the combined effects of warmer and drierconditions might make them among the firstspecies to be affected by climate variability.

We sampled live ponderosa pine from twosubspecies (P. ponderosa var. ponderosa and P.ponderosa var. scopulorum) at eight sites in threeregions of the western United States (Figure1, Table 1). We selected open-canopy (i.e.,park-like) woodland sites on federally ownedlands with minimal anthropogenic disturbancesfrom logging, domestic livestock grazing, fire


Radial Growth and Water-Use Efficiency for Ponderosa Pines 3

Figure 1 Location of the eight study sites.

management, and nitrogen deposition (Fennet al. 2003). We sampled more than forty treesper site, taking two increment cores per treeat approximately 1.4 m above ground level(Phipps 1985). We used a selective samplingstrategy to avoid as many site-related con-founding influences to radial growth as possibleand excluded trees with large fire scars, dead ordamaged tops, and abnormal foliar growth as-sociated with dwarf mistletoe (Stanton 2007).All cores were processed using standard labo-ratory techniques for sample preparation andsanding (Stokes and Smiley 1968). We usedthe list method for cross-dating (Yamaguchi1991), a technique that allows for each ringon each core to be assigned an exact calendardate, which accounts for the possibility of miss-ing rings (i.e., years with zero radial growth)or false rings. We measured ring widths on allcores to 0.001 mm precision using a linear en-coder and developed the tree-ring chronologybased on the measured values. We confirmedthe accuracy of cross-dating at each site us-

ing the program COFECHA (Holmes 1983;Grissino-Mayer 2001) and developed a tree-ring chronology based on cores selected forisotopic analysis using the program ARSTAN(Cook 1985). A tree-ring chronology showedthe average annual rate of growth for a studysite. To account for age-related declines in ra-dial growth we conservatively standardized theraw ring width measurements using negativeexponential, negative linear, or line throughthe mean. We then selected the STANDARDchronology version for final analysis to preservelow-frequency growth variance needed to ex-amine long-term relationships between radialgrowth and climate (Grissino-Mayer 1996).

Isotopic AnalysisAll trees selected for isotopic analysis had (1)a clear ring structure and growth patternsclosely matched to the full chronology pattern(i.e., they were included in the full chronologyby virtue of their interseries correlation), (2)



Table 1 Study site and tree-ring chronology information

SiteLatitude/longitude

Elevation(m)

Treesubspecies

Beginning year oftree-ring data

Ending year oftree-ring dataa

Series intercorrelation/

mean sensitivityb

RCN 37.75 2,261 scopulorum 1480 2005 0.65112.32 0.26

USC 37.95 2,528 scopulorum 1620 2005 0.74111.6 0.36

DRY 40.65 2,736 scopulorum 1720 2005 0.75109.75 0.25

FLR 47.33 1,214 ponderosa 1560 2004 0.61114.87 0.26

RCP 46.95 1,683 ponderosa 1602 2007 0.65114.33 0.28

BCR 45.83 1,555 ponderosa 1650 2004 0.62114.25 0.26

SLX 43.05 1,486 ponderosa 1659 1999 0.72121.1 0.30

FMC 45.35 1,265 ponderosa 1750 2000 0.53121.43 0.24

aWe collected all samples in the following year.bThe values are within the range of medians reported for this species by the NOAA Paleoclimatology Program (seehttp://www.ncdc.noaa.gov/paleo/treering/cofecha/speciesdata.html).

two cores available for destructive analysis, and(3) interior dates as old as possible (preferablypre-1800) at approximately 1.4 m aboveground level. Using six matched tree cores persite (thus n = 12 cores, except for RCN wheren = 11), we calculated iWUE by pentad from1850 (i.e., 1850–1854) through 1995–1999 or2000–2004 (depending on the length of thetree-ring record). The isotopic laboratory workwas performed by the Environmental IsotopeLaboratory of the Department of Geosciencesat the University of Arizona using standardtechniques (Sternberg 1989; Leavitt & Danzer1993; Coplen 1996). The laboratory analysesprovided us with pentadal measurements ofcarbon isotope composition (δ13C), a measurereflecting temporal variability or directionalchanges in physiological responses that includestomatal conductance (Seibt et al. 2008). Wethen used the δ13C measurements derivedfrom the core samples, δ13Cplant, to calculateiWUE (A/gH20) where A = CO2 assimilationuptake and gH20 = by the following (Farquhar,Ehleringer, and Hubick 1989):

� ‰ = δ13Cair − δ13Cplant

1,000 + δ13Cplant× 1,000 (1)

where � is carbon isotopic discrimination ex-pressed in parts per mille and δ13Cair andδ13Cplant are carbon isotopic composition ofthe atmosphere and tree rings, respectively.Using Equation 1, iWUE (A/gH20) can bedetermined:

�(‰) = a + (b − a)(1 − 1.6A) CagH20(2)

where a and b represent fractionation throughthe stomata (4.4‰) and during carboxylation(27‰) for C3 plants (Farquhar and Richards1984), Ca is CO2 concentration (Etheridgeet al. 1998; Keeling and Whorf 2005), A is theCO2 uptake rate, and gH20 is leaf conductanceto water. Thus:

iWUE = A/gH20 = ca(b − �)/1.6(b − a)(3)

Although each chronology began in the eigh-teenth century or earlier (Table 1), some in-dividual samples dated to the early 1800s. Toavoid the “juvenile effect” that could artificiallybias our results (McCarroll and Loader 2004,



789) we excluded from our analyses all pentadaldata prior to 1850.

Statistical AnalysisWe identified the dominant climatic drivers ofradial growth using monthly climatic division-level data (NOAA-ESRL 2009) for tempera-ture, precipitation, and the Palmer DroughtSeverity Index (PDSI; Palmer 1965). We exam-ined concurrent relationships between annualradial growth and climate; lagged relationshipsup to one year; and relationships with monthly,annual, and multimonth averages using Pear-son correlation. Once we identified the climatevariable most strongly related to radial growth,we created bivariate regression models withradial growth as the dependent variable and cli-mate as the independent variable. We examinedthe residuals from each model for temporaltrends using Pearson correlation. The presenceof residual trends has been used (Graumlich1991; Knapp, Soule, and Grissino-Mayer 2001)to suggest that either model over- or under-prediction is caused by a factor(s) not capturedby the (climatic) variables considered. To de-termine the potential impact of CO2 on radialgrowth we added annual CO2 values as a secondindependent variable in the regression modelsand examined the partial R2 values and residualtrends. From 1959 through the end of thetree-ring record (Table 1), we obtained annualCO2 data from the Mauna Loa record (Keelingand Whorf, 2005). For the period from 1895through 1958 we obtained pentadal CO2 datafrom the Law Dome ice cores (Etheridge et al.1998) and used a linear interpolation to obtainannual values between pentadal values.

To identify possible significant (p < 0.05)temporal trends in radial growth and iWUE bypentad we used Pearson correlation over threetime periods: 1850–1995/2000 pentads, 1850–1945 pentads, and 1950–1995/2000 pentads.We selected 1950 as a dividing point becausethe largest increases in CO2 occurred duringthis period and prior studies have used the samethreshold to analyze associated changes in CO2and plant growth (Kienast & Luxmoore 1988;Graumlich 1991; Knapp, Soule, and Grissino-Mayer 2001; Soule and Knapp 2006). Weused Curve Estimation (SPSS 2009) to deter-mine the best curve fit (e.g., linear, quadratic,power function) for trends in iWUE. We used

Pearson correlation to determine the relation-ship between iWUE and radial growth by pen-tad and to identify the presence of trends in theannual climate data, the annual radial growthdata, and the residuals from the linear regres-sion models.

We used pre- and post-1950 subsets todetermine if radial growth responses dur-ing drought periods have changed throughthe instrumental time period (i.e., 1895–end).We tested for significant differences in radialgrowth and drought severity as measured bya twelve-month mean of PDSI from prior-year October to current-year September usingMann–Whitney tests. We retained years ineach subset with PDSI values ≤ –1.0 (milddrought or worse) and tested for differencesin radial growth and drought severity pre- andpost-1949 for the drought years only.

Results

Changes in iWUEA positive quadratic model provided the bestfit for the significant positive trends from 1850through 1995/2000 and the highest iWUEvalues were recorded during the last pentadat six of eight sites and during the last twodecades at all eight study sites (Figure 2, Ta-ble 2). Fifty percent of the sites had signif-icant positive trends in iWUE for the 1850through 1945 pentads and all sites had sig-nificant positive trends for the 1950 through1995/2000 pentads. The magnitude of the re-lationship also increased except for the SLXsite in southern Oregon (Table 2). Because thesignificance, direction, and magnitude of tem-poral trends established by Pearson correlationfollowed those established by the nonparamet-ric Spearman correlation (e.g., see Table 2 for1850–end iWUE pentadal trends), we use onlythe Pearson results for consistency in determin-ing the presence of trends.

Radial GrowthUsing the pentadal averages over the full studyperiod (i.e., 1850–end), four sites had signif-icant positive trends in radial growth, withtwo sites having positive trends in the early(1850–1945) and late (1950–1995/2000) pen-tadal periods. Using yearly data (1895–end), we



RCN

70

80

90

100

110

120

130

18501860

18701880

18901900

19101920

19301940

19501960

19701980

19902000

Pentad

iWU

E

0

0.5

1

1.5

2

2.5

Gro

wth

USC

70

80

90

100

110

120

130

18501860

18701880

18901900

19101920

19301940

19501960

19701980

19902000

Pentad

iWU

E

0

0.5

1

1.5

2

2.5

Gro

wth

DRY

70

80

90

100

110

120

130

18501860

18701880

18901900

19101920

19301940

19501960

19701980

19902000

Pentad

iWU

E

0

0.5

1

1.5

2

2.5

Gro

wth

FLR

70

80

90

100

110

120

130

18501860

18701880

18901900

19101920

19301940

19501960

19701980

19902000

Pentad

iWU

E

0

0.5

1

1.5

2

2.5

Gro

wth

RCP

70

80

90

100

110

120

130

18501860

18701880

18901900

19101920

19301940

19501960

19701980

19902000

Pentad

iWU

E

0

0.5

1

1.5

2

2.5

Gro

wth

BCR

70

80

90

100

110

120

130

18501860

18701880

18901900

19101920

19301940

19501960

19701980

19902000

Pentad

iWU

E

0

0.5

1

1.5

2

2.5

Gro

wth

SLX

70

80

90

100

110

120

130

18501860

18701880

18901900

19101920

19301940

19501960

19701980

1990

Pentad

iWU

E

0

0.5

1

1.5

2

2.5

Gro

wth

FMC

70

80

90

100

110

120

130

18501860

18701880

18901900

19101920

19301940

19501960

19701980

1990

Pentad

iWU

E

0

0.5

1

1.5

2

2.5

Gro

wth

Figure 2 Pentadal values of intrinsic water-use efficiency (iWUE; orange squares) and radial growth(blue diamonds) from 1850 to end. The trend lines represent a second-order polynomial (gold for iWUE;light blue for radial growth).

found positive radial growth trends at five sites(Table 3), and two and three sites had positivetrends in the last fifty and thirty years, respec-tively (Table 3). At FLR (Montana), the long-

term positive trend in growth was replaced witha negative trend over the last fifty- and thirty-year periods, even though radial growth indexesremained above 1.0 since 1945 (Figure 2).


Tab

le2

Tem

pora

ltre

nds

ofin

trin

sic

wat

er-u

seef

ficie

ncy

(iWU

E)a

ndra

dial

grow

thba

sed

onpe

ntad

alav

erag

esan

dre

latio

nshi

psbe

twee

niW

UE

and

radi

algr

owth

Sit

e

iWU

Ew

ith

tim

e

All

pen

tad

s1850–

(SLX

,FM

C

n=

30;all

oth

ers

n=

31)

iWU

Ew

ith

tim

ea

All

pen

tad

s1850–

(SLX

,FM

C

n=

30;all

oth

ers

n=

31)

iWU

Eb

est

reg

ressio

nm

od

el

wit

hti

me

b,c

All

pen

tad

s1850–

(SLX

,

FM

Cn

=30;

all

oth

ers

n=

31)

iWU

Ew

ith

tim

e

All

pen

tad

s

1850–1945

(n=

20)

iWU

Ew

ith

tim

e

All

pen

tad

s

1950–

(SLX

,

FM

Cn

=10;

all

oth

ers

n=

11)

Rad

ial

gro

wth

wit

hti

me

All

pen

tad

s

1850–

(SLX

,

FM

Cn

=30;

all

oth

ers

n=

31)

Rad

ial

gro

wth

wit

hti

me

All

pen

tad

s

1850–1945

(n=

20)

Rad

ial

gro

wth

wit

hti

me

All

pen

tad

s

1950–

(SLX

,

FM

Cn

=10;

all

oth

ers

n=

11)

iWU

Ean

dg

row

th

All

pen

tad

s1850–

(rvalu

e;

pvalu

e,

n)

(SLX

,FM

Cn

=30;

all

oth

ers

n=

31)

RC

N0.

79∗∗

0.74

∗∗0.

76∗∗

0.24

0.84

∗∗0.

310.

280.

26−0

.05

US

C0.

88∗∗

0.89

∗∗0.

82∗∗

0.67

∗∗0.

82∗

0.00

0.06

0.39

−0.2

5D

RY

0.74

∗∗0.

75∗∗

0.66

∗∗0.

45∗

0.82

∗∗0.

39∗

0.00

0.69

∗0.

28FL

R0.

76∗∗

0.78

∗∗0.

7∗∗

0.22

0.74

∗∗0.

79∗∗

0.49

∗−0

.51

0.57

∗∗R

CP

0.73

∗∗0.

63∗∗

0.74

∗∗0.

160.

89∗∗

0.75

∗∗0.

280.

63∗

0.62

∗∗B

CR

0.73

∗∗0.

66∗∗

0.83

∗∗−0

.02

0.87

∗∗0.

70∗∗

0.57

∗∗0.

020.

38∗

SLX

0.94

∗∗0.

95∗∗

0.91

∗∗0.

87∗∗

0.85

∗∗0.

22−0

.62∗

∗0.

420.

14FM

C0.

83∗∗

0.86

∗∗0.

87∗∗

0.59

∗0.

84∗∗

0.19

0.03

0.18

0.18

Not

e:A

llva

lues

repr

esen

tPe

arso

nr

valu

esex

cept

asfo

otno

ted.

a Spe

arm

anco

rrel

atio

nr-

valu

es.

bR

2va

lues

.c A

llbe

stm

odel

squ

adra

ticin

form

.∗ p

<0.

05.

∗∗p

<0.

01.

7


Tab

le3

Rad

ialg

row

th/c

limat

ere

latio

nshi

psan

dlin

ear

tren

dsof

radi

algr

owth

,clim

ate,

and

resi

dual

sfr

omth

egr

owth

/clim

ate

and

grow

th/c

limat

e+

CO

2re

gres

sion

mod

els

Sit

eC

lim

ate

va

ria

ble

Be

st

gro

wth

-

cli

ma

te

rela

tio

nsh

ipa

Lin

ea

r

tre

nd

of

cli

ma

te

va

ria

ble

b

Lin

ea

rtr

en

d

of

cli

ma

te

va

ria

ble

(la

st

50

ye

ars

)

Lin

ea

rtr

en

d

of

cli

ma

te

va

ria

ble

(la

st

30

ye

ars

)

Gro

wth

-

tim

e

rela

tio

nsh

ip

18

95

–e

nd

Gro

wth

-

tim

e

rela

tio

nsh

ip

La

st

50

ye

ars

Gro

wth

-

tim

e

rela

tio

nsh

ip

La

st

30

ye

ars

Lin

ea

rtr

en

do

f

resid

ua

lsfo

rth

e

gro

wth

-cli

ma

te

reg

ressio

n

mo

de

l

Gro

wth

-

cli

ma

te+

CO

2

rela

tio

nsh

ipa

Lin

ea

rtr

en

d

of

resid

ua

ls

for

the

gro

wth

-cli

ma

te

+C

O2

mo

de

l

Imp

rove

me

nt

inR

2fo

r

the

+CO

2

mo

de

l

RC

NA

nnua

lcP

DS

I0.

37∗∗

0.14

0.25

0.06

0.15

0.07

−0.2

80.

080.

370.

040.

0011

011

050

3011

150

3011

011

011

0U

SC

July

PD

SI

0.31

∗∗0.

110.

220.

040.

020.

23−0

.16

−0.0

50.

31−0

.03

0.00

111

111

5030

111

5030

111

111

111

DR

YJa

nuar

y+

May

+0.

17∗∗

0.11

0.14

0.20

0.27

∗∗0.

53∗∗

0.45

∗0.

120.

3∗∗

−0.1

30.

13Ju

nepr

ecip

itatio

n11

111

150

3011

150

3011

111

111

1FL

RA

nnua

lPD

SI

0.13

∗∗−0

.12

−0.4

2∗∗

−0.3

8∗∗

0.59

∗∗−0

.29∗

−0.4

5∗0.

68∗∗

0.38

∗∗0.

21∗

0.25

109

109

5030

110

5030

109

109

109

RC

PA

pril

+M

ay+

June

+0.

1∗∗

0.05

0.14

−0.1

60.

67∗∗

0.75

∗∗0.

72∗∗

0.64

∗∗0.

64∗∗

−0.0

40.

54Ju

lypr

ecip

itatio

n11

311

350

3011

350

3011

211

311

3B

CR

May

+Ju

ne+

July

0.24

∗∗−0

.01

0.12

−0.0

40.

34∗∗

0.05

−0.3

20.

39∗∗

0.31

∗∗0.

120.

07pr

ecip

itatio

n11

011

050

3011

050

3011

011

011

0S

LXPr

ior

Nov

embe

rto

0.36

∗∗0.

06−0

.08

−0.0

20.

38∗∗

0.27

0.4∗

0.43

∗∗0.

54∗∗

−0.0

60.

18cu

rren

tA

ugus

t10

410

450

3010

550

3010

410

410

4P

DS

IFM

CM

ay+

June

+Ju

ly0.

13∗∗

0.17

0.15

0.37

∗−0

.04

0.16

0.33

−0.0

90.

13−0

.10

0.00

prec

ipita

tion

106

106

5030

106

5030

106

106

106

Not

e:P

DS

I=Pa

lmer

Dro

ught

Sev

erity

Inde

x.aR

2va

lues

and

allr

elat

ions

hips

are

posi

tive.

bD

eter

min

edvi

aPe

arso

nco

rrel

atio

n(r

valu

ean

dsa

mpl

esi

ze).

c Ann

ualc

alcu

late

das

the

12-m

onth

aver

age

ofpr

ior-y

ear

Oct

ober

tocu

rren

t-ye

arS

epte

mbe

rva

lues

.∗ p

<0.

05.

∗∗p

<0.

01.

8



Figure 3 A comparison of radial growth and annual Palmer Drought Severity Index (PDSI) values(prior-year October through current-year September) between the early (1896–1949) and late (1950–end)periods using all annual data and data from years when annual PDSI values were ≤ –1.00 (moderatedrought severity). Significant differences based on a Mann–Whitney U test are highlighted (*p < 0.05;∗∗ p < 0.01).

Relationships Among Radial Growthand Climate, CO2 , and iWUEThe primary climatic drivers of radial growthwere site specific but included the impactsof moisture during the growing season at allsites (e.g., July PDSI in southern Utah [USC];May–July precipitation at BCR [southwesternMontana] and FMC [northern Oregon]) andmoisture extending to the prior calendaryear at others (e.g., prior-year October tocurrent-year September [i.e., Annual] PDSIvalues at RCN [southern Utah] and FLR[Montana]; Table 3). Radial growth varianceexplained by the primary climate variablesranged from 10 to 37 percent (Table 3). Atfour sites, significant positive trends exist inresiduals from the growth/climate regression

model, indicating that the climate variablesalone underpredict observed radial growth inthe later portion of the study. Adding annualCO2 as a second variable in the regressionmodels revealed that it influences radial growthat five of the eight study sites significantly(Table 3) and increased explained variance atthese sites by an average of 23.4 percent.

None of the climate variables displayed long-term temporal trends, but May through Julyprecipitation in northern Oregon (FMC) in-creased during the last thirty years, and annualPDSI for FLR (Montana) decreased for thelast fifty- and thirty-year periods (Table 3). An-nual CO2 values trend positively through time(R2 = 0.99 for a cubic model), and there are sig-nificant, positive relationships between iWUEand radial growth at three of the sites over the



full study period when using the pentadal data(Table 2).

Radial Growth Comparisons During DroughtYearsMean radial growth was significantly greaterin the post-1950 period compared to the pre-1950 period at five sites. The increases occurredabsent significant differences in moisture con-ditions, with all sites recording mean annualPDSI values in the near normal range (0.49to –0.49). Further, using a subset of droughtyears produced similar results, with four sitesrecording significantly greater radial growth inthe post-1950 period despite no significant dif-ferences in drought conditions. At these foursites, the mean radial growth was at or abovenormal (1.0) despite average drought condi-tions classified as moderate or worse (PDSI ≤–2.0). At DRY (northern Utah), there was a sig-nificant difference in both radial growth ratesand drought severity. Radial growth at DRYincreased post-1950 when drought conditionswere less severe than the pre-1950 period.

Discussion

Our results demonstrate that trends towardhigher rates of iWUE for ponderosa pine arepanregional, occurring at eight sites withinthree distinct climatic regimes and for twosubspecies (Figure 1). The increasing trendsin iWUE at our sites are similar to thosereported by Feng (1999) for several conif-erous tree species found throughout westernNorth America, and increasing iWUE hasbeen reported for conifers at other northernhemisphere locations (e.g., Bert, Leavitt, andDupouey 1997; Saurer, Siegwolf, and Schwe-ingruber 2004). Further, our results showingthe highest iWUE were recorded in the lastpentad at six of eight sites and follow a positivequadratic progression at all sites (Figure 2), sug-gesting that future increases in iWUE are likelyfor ponderosa pine within our study regions asCO2 levels increase.

Five of the eight sites display long-term pos-itive trends in annual radial growth rates post-1895, which have occurred absent long-termtrends toward more favorable climate condi-tions (Table 3). Pairing CO2 with the climate

variables results in a significant improvementin the explanatory power of the model at fivesites. At all but one site, the positive tempo-ral autocorrelation of residuals disappears withthe inclusion of CO2, and at three sites a sig-nificant relationship between pentadal iWUEand radial growth exists. We also found signif-icant improvements in radial growth rates dur-ing drought years after 1950. In a prior studythat included the full chronologies for FMCand SLX plus six additional sites in the Pa-cific Northwest, Soule and Knapp (2006) foundsimilar improvements in radial growth duringdrought periods of the late twentieth to earlytwenty-first century and concluded that CO2enrichment was likely operative. This priorstudy lacked the empirical evidence providedby the iWUE data, however, which were de-rived from collected tree-ring samples.

For three sites (RCN, USC in southernUtah, and FMC in northern Oregon), we foundno evidence of significant radial growth in-creases despite significant increases in iWUE.As the RCN and USC chronologies weredeveloped from P. ponderosa var. scopulorum(Figure 1), we posited that different genotypesmight vary in their responses to increasediWUE, but a similar response occurs with theFMC chronology developed from P. ponderosavar. ponderosa that otherwise was responsiveto increased iWUE, and the DRY (northernUtah) chronology was also developed fromP. ponderosa var. scopulorum. Alternatively, wealso speculated that the timing of precipitationmight be influential, as RCN and USC experi-ence a late-summer maximum, but FMC has alate-summer minima of precipitation and alsolacks long-term trends in radial growth.

Although microscale environmental factorsundetected by our analysis could overridethe potentially positive impacts of increasingiWUE on radial growth in some locations, thetheoretical progression that increased CO2 im-proves iWUE is supported by our results. Inconcert, these findings lead us to conclude that,for open-canopy ponderosa pine trees growingin locations with limited anthropogenic distur-bance, increased iWUE associated with risingCO2 might be a contributing factor to observedincreases in radial growth rates through time.That said, we caution that these conclusionsare based on a limited sample size for each site



and focus on one of many potential vectors ofchange for western U.S. forests.

Multiple, synergistic linkages exist betweenforest processes and climate change (Bonan2008) and between climate change and forestprocesses (Ollinger et al. 2008). For example,future increases in CO2 should lead to increas-ing rates of net primary productivity (NPP) ina variety of forested ecosystems because “thephotosynthetic uptake of carbon that drivesNPP is not saturated at current levels” (Norbyet al. 2005, 18052). In turn, this can poten-tially impact rates of carbon sequestration andfeedback mechanisms between the biosphereand atmosphere that control CO2 concentra-tions. Globally, about one third of the carbonreleased by humans from fossil fuel combus-tion and land clearing is removed from theatmosphere by trees, and forests can impactclimatic conditions through changes in surfacealbedo, evapotranspiration, and runoff (Bonan2008; Ollinger et al. 2008). If potential cli-mate changes lead to increasing aridity in thewestern United States, additional increases iniWUE associated with future increase in CO2might ameliorate growth declines associatedwith drought conditions. As iWUE changes arerecognized as a potentially important compo-nent of global environment change processes(Aber et al. 2001), it is likely that forest ecosys-tems will continue to be influenced by risingCO2. �

Literature Cited

Aber, J., R. P. Neilson, S. McNulty, J. M. Lenihan,D. Bachelet, and R. J. Drapek. 2001. Forest pro-cesses and global environmental change: Predict-ing the effects of individual and multiple stressors.BioScience 51:735–51.

Ainsworth, E. A., and A. Rogers. 2007. The responseof photosynthesis and stomatal conductance to ris-ing [CO2]; mechanisms and environmental inter-actions. Plant, Cell and Environment 30:258–70.

Arneth, A., J. Lloyd, H. Santruckova, M. Bird, S.Girgoryev, Y. N. Kalaschnikov, G. Gleixner, andE. Schulze. 2002. Response of central SiberianScots pine to soil water deficit and long-term trendsin atmospheric CO2 concentration. Global Biogeo-chemical Cycles 16:5.1–5.13.

Bert, D., S. Leavitt, and J.-L. Dupouey. 1997. Vari-ations of wood δ13C and water-use efficiency ofAbies alba during the last century. Ecology 78:1588–96.

Bonan, G. B. 2008. Forests and climate change: Forc-ings, feedbacks, and the climate benefits of forests.Science 320:1444–49.

Breda, N., R. Huc, A. Granier, and E. Dreyer E. 2006.Temperate forest trees and stand under severedrought: A review of ecophysiological responses,adaptation processes and long-term consequences.Annals of Forest Science 63:625–44.

Christensen, J., B. Hewitson, A. Busuioc, A. Chen,X. Gao, I. Held, R. Jones, et al. 2007. Regionalclimate predictions. In Climate change 2007: Thephysical science basis. Contribution of Working GroupI to the Fourth Assessment Report of the Intergovern-mental Panel on Climate Change, ed. S. Solomon, D.Qin, M. Manning, Z. Chen, M. Marquis, K. B. Av-eryt, M. Tignor, and H. L. Miller, 847–940. NewYork: Cambridge University Press.

Ciais, P., M. Reichstein, N. Viovy, A. Granier,J. Ogee, V. Allard, M. Aubinet, et al. 2005. Europe-wide reduction in primary productivity caused bythe heat and drought in 2003. Nature 437:529–33.

Cook. E. 1985. A time series analysis approach totree-ring standardization. PhD thesis, Universityof Arizona, Tucson, AZ.

Coplen, T. 1996. New guidelines for reporting sta-ble hydrogen, carbon and oxygen isotope-ratiodata. Geochimica. et Cosmochimica Acta 602:3359–60.

Etheridge, D., L. Steele, R. Langenfelds, R. Francey,V. Barnola, and J.-M. Morgan. 1998. HistoricalCO2 records from the Law Dome DE08, DE08–2,and DSS ice cores. In Trends: A compendium of dataon global change. Oak Ridge, TN: Carbon DioxideInformation Analysis Center, Oak Ridge NationalLaboratory, U.S. Department of Energy.

Farquhar, G. D., J. R. Ehleringer, and K. T. Hubick.1989. Carbon isotope discrimination and photsyn-thesis. Annual Review of Plant Physiology and PlantMolecular Biology 40:503–37.

Feng, X. 1999. Trends in intrinsic water-use effi-ciency of natural trees for the past 100–200 years: Aresponse to atmospheric concentration. Geochimicaet Cosmochimica Acta 63:1891–1903.

Fenn, M., J. Baron, E. Allen, H. Rueth, K. Nydick,L. Geiser, W. Bowman, J. Sickman, T. Meixner,D. Johnson, and P. Neitlich. 2003. Ecological ef-fects of nitrogen deposition in the western UnitedStates. BioScience 53:404–20.

Graumlich, L. 1991. Subalpine tree growth, cli-mate, and increasing CO2: An assessment of recentgrowth trends. Ecology 72:1–11.

Grissino-Mayer, H. D. 1996. A 2129-year recon-struction of precipitation for northwestern NewMexico, USA. In Tree-rings environment and hu-manity, ed. J. S. Dean, D. M. Meko, and T. W.Swetnam, 191–204. Tucson: Department of Geo-sciences, The University of Arizona.



———. 2001. Evaluating crossdating accuracy: Amanual and tutorial for the computer programCOFECHA. Tree-Ring Research 57:205–21.

Hanson, P., and J. Weltzin. 2000. Drought distur-bance from climate change: Response of UnitedStates forests. The Science of the Total Environment262:205–20.

Holmes, R. 1983. Computer-assisted quality controlin tree-ring dating and measurement. Tree-RingBulletin 43:69–78.

Keeling, C., and T. Whorf. 2005. Atmospheric CO2records from sites in the SIO air sampling network.In Trends: A compendium of data on global change. OakRidge, TN: Carbon Dioxide Information AnalysisCenter, Oak Ridge National Laboratory, U.S. De-partment of Energy.

Kienast, F., and R. Luxmoore. 1988. Tree-ring anal-ysis and conifer growth responses to increased at-mospheric CO2levels. Oecologia 76:487–95.

Knapp, P., and P. Soule. 2008. Use of atmosphericCO2-sensitive trees may influence dendroclimaticreconstructions. Geophysical Research Letters 35. doi:10.1029/2008GL035664

Knapp, P., P. Soule, and H. Grissino-Mayer. 2001.Detecting the potential regional effects of in-creased atmospheric CO2 on growth rates of west-ern juniper. Global Change Biology 7:903–17.

Leavitt, S., and S. Danzer. 1993. Method for batchprocessing small wood samples to holocellulose forstable-carbon isotope analysis. Analytical Chemistry65:87–89.

Liu, X, X. Shao, E. Liang, L. Zhao, T. Chen, D. Qin,and J. Ren. 2007. Species dependent responses ofjuniper and spruce to increasing CO2 concentra-tion and to climate in semi-arid and arid areasof northwestern China. Plant Ecology 193:195–209.

McCarroll, D., and N. Loader. 2004. Stable isotopesin tree rings. Quaternary Science Reviews 23:771–801.

NOAA-ESRL. 2009. US Climate Division: Temper-ature, precipitation and Palmer Drought SeverityIndex (PDSI). http://www.esrl.noaa.gov/psd/data/timeseries/ (last accessed 1 December 2009).

Norby, R. J., E. H. Delucia, B. Glelen, C. Calfaple-tra, C. P. Giardina, J. S. King, J. Ledford et al.2005. Forest response to elevated CO2 is conservedacross a broad range of productivity. Proceedings ofthe National Academy of Science 102:18052–56.

Ollinger, S. V., C. L. Goodale, K. Hayhoe, andJ. P. Jenkins. 2008. Potential effects of climatechange and rising CO2 on ecosystem processes innortheastern U.S. forests. Mitigation and Adapta-tion Strategies for Global Change 13:467–85.

Palmer, W. 1965. Meteorological drought. Washing-ton, DC: U.S. Government Printing Office.

Phipps, R. 1985. Collecting, preparing, crossdating, andmeasuring tree increment cores. U.S. Geological Sur-

vey Water-Resources Investigations Report 85-4148. Reston, VA: Water Resources Division.

Saurer, M., R. Siegwolf, and F. Schweingruber. 2004.Carbon isotope discrimination indicates improv-ing water-use efficiency of trees in northern Eura-sia over the last 100 years. Global Change Biology10:2109–20.

Seibt, U., A. Rajabi, H. Griffiths, and J. A. Berry.2008. Carbon isotopes and water use efficiency:Sense and sensitivity. Oecologia 155:441–54.

Soule, P., and P. Knapp. 2006. Radial growth rateincreases in naturally-occurring ponderosa pinetrees: A late 20th century CO2 fertilization effect?New Phytologist 171:379–90.

SPSS. 2009. SPSS for Windows, Rel. 17.0.0. Chicago:SPSS.

Stanton, S. 2007. Effects of dwarf mistletoe on cli-mate response of mature ponderosa pine trees. TreeRing Research 63:69–80.

Sternberg, L. 1989. Oxygen and hydrogen isotopemeasurements in plant cellulose. In Modern meth-ods in plant analysis, ed. H. F. Linskens and J. F.Jackson, 89–98. New York: Springer-Verlag.

Stokes, M., and T. Smiley. 1968. Introduction to tree-ring dating. Chicago: University of Chicago Press.

Tang, K., X. Feng, and G. Funkhouser. 1999. Theδ13C of trees in full-bark and strip-bark bristleconepine trees in the White Mountains of California.Global Change Biology 5:33–40.

Tognetti, R., A. Longobucco, F. Miglietta, andA. Raschi. 1998. Transpiration and stomatal be-haviour of Quercus ilex plants during the summerin a Mediterranean carbon dioxide spring. Plant,Cell and Environment 21:613–22.

van Mantgem, P., N. Stephenson, J. Byrne, L.Daniels, J. Franklin, P. Fule, M. Harmon, A.Larson, J. Smith, A. Taylor, and T. Veblen. 2009.Widespread increase of tree mortality rates in theWestern United States. Science 323:521–24.

Waterhouse, J., V. Switsur, A. Barker, A. Carter, D.Hemming, N. Loader, and I. Robertson. 2004.Northern European trees show a progressivelydiminishing response to increasing atmosphericcarbon dioxide concentrations. Quaternary ScienceReviews 23:803–10.

Western Wood Products Association. 2009.Ponderosa pine species facts. http://www2.wwpa.org/SPECIESPRODUCTS/PonderosaPine/tabid/298/Default.aspx (last accessed 1 December2009).

Yamaguchi, D. 1991. A simple method for cross-dating increment cores from living trees. CanadianJournal of Forest Research 21:414–16.

Youngblood, A. 2005. Silvicultural systems for manag-ing Ponderosa pine. USDA Forest Service Techni-cal Report PSW-GTR-198, U.S. Department ofAgriculture. Albany, CA: Pacific Southwest Re-search Station.



PETER T. SOULE is Professor in the Depart-ment of Geography and Planning at AppalachianState University, Boone, NC 28608. E-mail:[email protected]. His primary research focusdeals with understanding the linkages betweenchanging climate and atmospheric conditions onforest ecosystems.

PAUL A. KNAPP is Professor in the De-partment of Geography at the University ofNorth Carolina Greensboro, Greensboro, NC27412. E-mail: [email protected]. His researchinterests focus on climatology, biogeography,and tree-ring science projects in the AmericanWest.


Documents

The Professional Geographer Radial Growth and … growth...Radial Growth and Increased Water-Use Efﬁciency for Ponderosa Pine Trees in Three Regions in the Western United States*