Age class, longevity and growth rate relationships:

protracted growth increases in old trees in the eastern United States

SARAH E. JOHNSON1,2 and MARC D. ABRAMS1

1 School of Forest Resources, Forest Resources Building, The Pennsylvania State University, University Park, PA 16802, USA

2 Corresponding author (sej141@psu.edu)

Received April 17, 2009; accepted August 1, 2009

Summary This study uses data from the International

Tree-Ring Data Bank website and tree cores collected in

the field to explore growth rate (basal area increment,

BAI) relationships across age classes (from young to old)

for eight tree species in the eastern US. These species

represent a variety of ecological traits and include those in

the genera Populus, Quercus, Pinus, Tsuga and Nyssa. We

found that most trees in all age classes and species exhibit

an increasing BAI throughout their lives. This is partic-

ularly unusual for trees in the older age classes that we

expected to have declining growth in the later years, as

predicted by physiological growth models. There exists an

inverse relationship between growth rate and increasing

age class. The oldest trees within each species have

consistently slow growth throughout their lives, implying

an inverse relationship between growth rate and longev-

ity. Younger trees (< 60 years of age) within each species

are consistently growing faster than the older trees when

they are of the same age resulting from a higher

proportion of fast-growing trees in these young age

classes. Slow, but increasing, BAI in the oldest trees in

recent decades is a continuation of their growth pattern

established in previous centuries. The fact that they have

not shown a decreasing growth rate in their old age

contradicts physiological growth models and may be

related to the stimulatory effects of global change

phenomenon (climate and land-use history).

Keywords: basal area increment, dendrochronology, globalchange, old-growth trees, sigmoidal growth model.

Introduction

Trees undergo physiological changes as they age, including

lower photosynthetic rates, decreased growth rates, shifting

of carbon resources to different parts of the plant and

reductions in foliar efficiency, leaf size and gas exchange

rates (Kaufmann 1996, Ryan and Yoder 1997, Carrer

and Urbinati 2004, Martınez-Vilalta et al. 2007). The larger

size and the structural complexity usually associated with

tree aging increase the maintenance respiration costs and

reduce the efficiency of water transport; these both tend

to reduce growth (Weiner and Thomas 2001, Carrer and

Urbinati 2004, Mencuccini et al. 2005, Pennisi 2005). Old

Quercus rubra L. trees, for example, had moderate to high

growth rates for the first 50–100 years, followed by a persis-

tent, slow growth over the subsequent 200 years (Orwig

et al. 2001). This growth trend may be related to an increas-

ing tree canopy during early age, a constant canopy volume

during middle age and then a physiological decline in old

trees (Spiecker et al. 1996). This is expected with changing

resource allocation in older trees, which maximizes life

expectancy by focusing the energy on defense and mainte-

nance rather than on growth (Loehle 1988, Herms and

Mattson 1992). Weiner and Thomas (2001) and Weiner

(2004) discuss a sigmoidal model in which the size of a plant

increases slowly at an early age, becomes exponential during

middle age and plateaus during old age. The latter phase is

attributed to age- and size-related growth declines. In this

study, we adopt a slightly revised version of the sigmoidal

model (referred to here as the ‘sigmoidal growth model’)

to include a late stage decline in tree growth rate after it

plateaus in old age (as predicted in the physiological growth

models). Thus, we are interested in testing whether the sig-

moidal growth model applies to a variety of trees where

there exists a large amount of dendrochronological data.

Vegetation is highly dynamic because the environment in

which it grows is constantly altered by natural and anthro-

pogenic factors. Anthropogenic disturbances such as

aboriginal burning, land clearing for agriculture, wide-

spread logging, introduction of exotic species and cata-

strophic wildfire followed by fire suppression have

dramatically altered the structure and function of forests

(Bazzaz 1990, Boisvenue and Running 2006, Abrams and

Nowacki 2008). In the United States, it is thought that

the European settlement legacy or ecological footprint is

very large, typified by intensive logging practices and subse-

quent slash wildfires (Williams 1982, Abrams 2003,

Nowacki and Abrams 2008). After European settlement,

a period of resource exploitation produced changes in

Tree Physiology Page 1 of 12

doi:10.1093/treephys/tpp068

� The Author 2009. Published by Oxford University Press. All rights reserved.For Permissions, please email: journals.permissions@oxfordjournals.org

Tree Physiology Advance Access published September 4, 2009

community composition and structure. Anthropogenic and

naturally caused disturbance events are discernable in the

dendrochronological record as they produce large increases

in radial growth in residual and newly established trees

(Baker 1995, Nowacki and Abrams 1997). In contrast,

the impacts of greenhouse gases and the corresponding,

often subtle, changes in climate are not as easily detectable

in the tree-ring record (Jacoby and D’Arrigo 1997, Mann

et al. 1999). Some studies have shown increased radial

growth in response to the global change factors such as

nitrogen fertilization and increased CO2 levels (Briffa

et al. 1998, Voelker et al. 2006). Thus, changes in land-

use history and climate may be impacting long-term growth

trends in trees. However, this impact may vary by species

and age class.

In this study, we explore the long-term growth patterns

for tree species of the eastern US to assess the relationship

between growth rate and age class, and whether trees in

varying age classes are following or deviating from the sig-

moidal growth model. We take the approach of examining

basal area increment (BAI) changes over time across all age

classes from young to old trees, rather than just studying

the oldest individuals of each species where sample biases

may exist (Cherubini et al. 1998, Voelker et al. 2006). Eight

tree species were chosen for this study to encompass a range

of ecological and life history attributes in the eastern forest

biome: bigtooth aspen (Populus grandidentata Michx.),

black oak (Quercus velutina Lam.), red oak (Q. rubra),

chestnut oak (Quercus Montana L.), white oak (Quercus

alba L.), pitch pine (Pinus rigida Mill.), hemlock (Tsuga

canadensis L. Carr.) and blackgum (Nyssa sylvaticaMarsh.;

Table 1). The primary objectives of this study are to exam-

ine the following:

1. The relationships between growth rate, longevity and

age class for contrasting tree species both within and

between sites.

2. The role of shade tolerance and site class (e.g., rock

outcrop and bog) in growth rate and longevity.

3. Whether BAI trends in trees are following the sigmoi-

dal growth model.

Methods

The International Tree-Ring Data Bank (ITRDB; Grissino-

Mayer and Fritts 1997) was used to compile tree-ring

chronologies for eight eastern North American tree species

(bigtooth aspen, blackgum, black oak, chestnut oak, hem-

lock, pitch pine, red oak and white oak; see Table 2 for

tree-ring series used). Further details pertaining to specific

sites can be found in appendix of Johnson (2007). The

majority of the contributing investigators were personally

contacted to ensure that they posted complete chronologies

extending from the bark to (or very near) the pith of the

trees recorded. Both young and old trees were obtained

from the ITRDB.

When faced with the decision of which measurement type

to use for this study, raw-ring widths, ring-width index

(RWI) and BAI were considered. The BAI is typically used

in forest growth and modeling studies because it provides

an accurate quantification of wood production due to the

ever-increasing diameter of a growing tree (Rubino and

McCarthy 2000). In contrast, RWI is typically used in den-

droclimatological studies to standardize ring widths into

indices to highlight above- or below-average periods of

growth through time in relation to climate (Esper et al.

2002). The RWI was calculated for the 20 oldest trees of

four of the study species at individual study sites. Raw-ring

widths were fitted with negative exponential curves using

the subtraction method in the program ARSTAN (Cook

and Holmes 1984). A comparison of BAI and RWI data

for two of these species revealed obvious long-term growth

trends evident in the BAI data that are masked in the RWI

(Figure 1). Because the focus of this study is on ecological

growth trends by age class in relation to the sigmoidal

growth model, rather than climate impacts on growth indi-

ces, we chose to use BAI, in addition to raw-ring width,

rather than RWI (Stan 2008).

Raw-ring-width chronologies of the selected species were

used to calculate tree age and average yearly growth rates in

BAI (expressed as mm2 year�1). The BAI measurements

were calculated for each tree using the tree radius and the

formula for circle area. Cores of each species of interest pre-

viously collected in our laboratory were reanalyzed for age

and growth rate bymeasuring the annual growth increments

on a TA Unislide Velmex machine (0.002 mm precision;

Velmex Inc., Bloomfield, NY). This data was used in

conjunction with the chronologies obtained from the

ITRDB. Many of these cores represented the oldest individ-

uals used for this study. These cores were taken as low

as possible on each tree above buttressing, mounted

on wood blocks and sanded with increasingly finer grit

Table 1. Maximum lifespan (Burns and Honkala 1990) and

oldest recorded individuals (Pederson 2009) of the study species

according to successional status.

Typical

maximum age

(years)

Oldest

recorded

(years)

Early successional

Bigtooth aspen 70–100 113

Pitch pine 200–300 375

Mid-successional

Black oak 150–200 257

Chestnut oak 300–400 427

Red oak 200–250 326

White oak 400–450 464

Late successional

Blackgum 500+ 679

Hemlock 500+ 555

2 JOHNSON AND ABRAMS

TREE PHYSIOLOGY

sandpaper (60–3200 grit in some cases, i.e., blackgum cores)

in preparation for measurement with the Velmex machine.

The cores were crossdated by hand using skeleton plotting

to check for missing and false rings using the identification

of signature years (Stokes and Smiley 1968). A skeleton plot

evaluates the tree-ring width in relation to the rest of the

rings of each tree in a stand. Signature years are consistently

small or large rings indicative of extreme climatic or distur-

bance events that allow for the crossdating of trees. We used

the computer programCOFECHA for crossdating (Holmes

1983, Grissino-Mayer 2001), which analyzes each measured

ring-width series individually. COFECHA bases its analysis

on a master chronology of all the cores compiled and calcu-

lates a correlation coefficient indicating how well the inter-

annual variability in ring widths correlates with the other

ring-width series. Ring-width chronologies obtained from

the ITRDB have been subjected to rigorous crossdating

standards and checked with COFECHA. The raw-ring

widths recorded for remeasured and newly collected core

sets were converted to BAI in the same manner as ring

widths collected from the ITRDB (see the above procedure).

The relationship between growth rate and age class was

analyzed based on 30-year age classes for the entire life of

the trees. Each tree was assigned to a 30-year age class

Table 2. Number of tree-ring series used in each age class for each of the study species.

Species 30–60

years

60–90

years

90–120

years

120–150

years

150–180

years

180–210

years

210–240

years

240–270

years

270–300

years

300+

years

Bigtooth aspen 8 37 4

Black oak 10 15 14 9 4 4 3

Red oak 6 28 19 20 12 16 5

Chestnut oak 39 55 35 24 24 44 29 41 29 3

White oak 14 55 99 139 133 221 239 239 136 11

Pitch pine 13 56 30 17 12 7 8 5 7 3

Hemlock 4 10 27 39 58 78 87 13

Blackgum 3 33 16 8 9 4 3 6 7 2

B

Tree age (years)0

BA

I (m

m2 /y

ear)

0

200

400

600

800

1000

1200

A

Rin

g W

idth

Inde

x (b

y ca

lend

ar y

ear)

0.94

0.96

0.98

1.00

1.02

1.04

D

Tree age (years)0

0

100

200

300

400

500

600

C

0.8

0.9

1.0

1.1

1.2

50 100 150 200 250 100 200 300

Figure 1. A comparison of RWI (Figure 1A and C; calculated by calendar year of growth) and BAI (Figure 1B and D) for the 20 oldesttrees of chestnut oak from Detweiler Run in central Pennsylvania (Figure 1A and B) and of blackgum from Mohonk State Park, NewYork (Figure 1C and D).

AGE CLASS, LONGEVITY AND GROWTH RATE RELATIONSHIPS 3

TREE PHYSIOLOGY ONLINE at http://www.treephys.oxfordjournals.org

(e.g., 30–60 years, 61–90 years, etc. to 300+ years; based on

tree age at sampling, not calendar year of growth). Tree

growth (both raw-ring width and BAI) was measured as

the average of 10-year intervals across all trees in each

age category. A repeated-measures analysis of covariance

was used to compare each tree’s decadal mean growth rates

both between age classes of a single species and among spe-

cies. Adjusted P values were generated for each age class

comparison (a = 0.05).

Results

Bigtooth aspen and pitch pine, both shade intolerant and

considered early successional, showed very different growth

rates as well as different maximum ages (120 and 300+

years, respectively). Overall, both bigtooth aspen and pitch

pine trees showed an inverse relationship between growth

rate and increasing age class (Figure 2A and C). Even so,

in both species older trees continued to increase in average

BAI throughout their life. For the first 50 years of aspen

growth, trees in the 30–60 year age class grew faster than

the trees in the 60–90 and 90–120 year age classes

(P = 0.004 and 0.017, respectively). Pitch pine trees 30–

90 years old exhibited the fastest growth rates, peaking at

BAIs at nearly 1400 mm2 year�1. Significant differences

in pitch pine BAI growth rates exist between disparate

age classes. For example, trees 30–150 years of age grew sig-

nificantly (P < 0.05) faster than the trees over 150 years of

age. The youngest pitch pine trees peaked at around

1400 mm2 year�1, whereas even though the youngest big-

tooth aspen trees also peaked at around this value, older

trees continued to increase in growth rate until leveling

off occurred in very old specimens. Bigtooth aspen trees

aged 90–120 years and pitch pine that lived to be over

300 years showed the slowest growth rates overall, indicat-

ing that maximum longevity is obtained through slow

growth. Young bigtooth aspen and pitch pine trees grew

significantly faster than did the older trees at the same

respective age. For example, when bigtooth aspen trees in

the 90–120 year age class were 30 years old, they grew at

� 200 mm2 year�1. In contrast, trees in the 30–60 year

age class were growing at about 1000 mm2 year�1 when

they were 30 years old. In terms of raw-ring data (mm

year�1), aspen trees in the 30–60 year age class showed a

growth decline over their lives, whereas trees in the 61–

120 year age class had an increasing growth (Figure 2B).

Pitch pine raw growth rates increased with age in younger

trees, but were fairly constant throughout the older age

classes (Figure 2D).

A

Bas

al a

rea

grow

th r

ate

(mm

2 /yea

r)

0

500

1000

1500

2000

2500

Tree age (years)0

Tre

e rin

g gr

owth

rat

e (m

m/y

ear)

1.0

1.5

2.0

2.5

3.0

30-60 years60-90 years90-120 years

B

C

0

200

400

600

800

1000

1200

1400

Tree age (years)0 100 200 300

0.5

1.0

1.5

2.0

2.5

30-60 years60-90 years90-120 years120-150 years150-180 years180-210 years210-240 years240-270 years270-300 years300+ years

D

20 40 60 80 100 120 140

Figure 2. Bigtooth aspen decadal average growth rates in BAI (A) and raw-ring widths (B), and pitch pine decadal average growthrates in BAI (C) and raw-ring widths (D) for each age class. Each decade indicates a 10-year period of averaged growth over all trees ineach data set.

4 JOHNSON AND ABRAMS

TREE PHYSIOLOGY

The four intermediately shade tolerant oak species were

separated based on family groups (red oak family versus

white oak family). In the red oak group, black oak trees

in the four youngest age classes (30–150 years old) had a

significantly greater BAI over time than the trees in the old-

est age classes (150 years old and above; Figure 3A). Young

trees exhibited large increases in growth rate with increasing

age. Growth increased over time in older black oak, but in

a much less dramatic fashion (� 200% in the youngest four

age classes compared to < 50% in the oldest four age clas-

ses). Black oak in older age classes grew much more slowly

than the younger black oak, both at the same respective age

and throughout the lives of these older trees (i.e., growth

rates of 200 mm2 year�1 in older black oak versus 2000–

3000 mm2 year�1 in younger black oak; P < 0.05). Black

oak raw-ring growth declined over time in the two youngest

age classes, but was highly variable (with no significant

trend) in trees in the older age classes (> 90 years; Figure

3B). It is important to note that constant raw-ring growth

over time results in an increasing BAI (i.e., it takes more

wood to produce the same size tree ring as the diameter

of a tree increases). In northern red oak, BAI increased sig-

nificantly from past to present in all age classes (Figure 3C).

Northern red oak trees between 30 and 120 years of age

typically had the fastest growth rates (P < 0.05), although

trees 120–150 years of age ultimately produced the highest

growth rate of about 5000 mm2 year�1. The youngest

northern red oak trees had increasing raw-ring growth over

time, whereas the older age classes had variable, but overall

flat growth trends (Figure 3D).

Species in the white oak group, chestnut oak and white

oak, displayed slower growth rates overall than the species

of the red oak family, except in the youngest chestnut oak

(Figure 4). Chestnut oak trees in the 30–60 and 60–90 year

age classes grew significantly faster in terms of BAI than

trees 120 years of age and older (P < 0.05), peaking near

8000 mm2 year�1 (Figure 4A). Chestnut and white oak

trees of all age classes had a variable but an increasing

BAI growth over time; however, white oak trees in the

300+ age class exhibited a growth plateau after 200 years

of age (Figure 4C). Young chestnut oak trees grew much

faster than did the older trees at the same respective age

(P < 0.05); young white oak showed a similar trend; how-

ever, the growth rates were less disparate when compared

with other oak species. Both chestnut oak and white oak

raw growth rates showed fairly constant trends; however,

chestnut oak 300+ years of age had a decline in raw-ring

growth over time (Figure 4B and D). Nonetheless, this

was enough to result in an increasing BAI curve in all

age classes. The decline in raw growth of white oak is more

A

Bas

al a

rea

grow

th r

ate

(mm

2 /yea

r)

0

2000

4000

6000

8000

Tree age (years)

Tre

e rin

g gr

owth

rat

e (m

m/y

ear)

0.5

1.0

1.5

2.0

30-60 years60-90 years90-120 years120-150 years150-180 years180-210 years210-240 years240-270 years

B

C

0

1000

2000

3000

4000

5000

Tree age (years)

0 50 100 150 200 2500 50 100 150 200 250

0.5

1.0

1.5

2.0

2.5

30-60 years60-90 years90-120 years120-150 years150-180 years180-210 years210-240 years

D

Figure 3. Black oak decadal average growth rates in BAI (A) and raw-ring widths (B), and red oak decadal average growth rates inBAI (C) and raw-ring widths (D) for each age class. Each decade indicates a 10-year period of averaged growth over all trees in eachdata set.

AGE CLASS, LONGEVITY AND GROWTH RATE RELATIONSHIPS 5

TREE PHYSIOLOGY ONLINE at http://www.treephys.oxfordjournals.org

apparent in the 300+ year old trees until the most recent

years when growth abruptly increased (Figure 4D).

Late-successional shade tolerant species (hemlock and

blackgum) displayed the slowest overall BAI growth rates

of the species studied (Figure 5A and B). Hemlock trees

exhibited a significantly increasing BAI growth over time

in all age classes, except 300+ year old trees; these trees

reached a growth plateau after 200 years. Blackgum trees

in all age classes exhibited an increasing BAI with increas-

ing age, including trees over 300 years old. The growth pla-

teau seen in the oldest trees of some of the study species was

not evident in blackgum even after 400 years of age. Fast-

growing hemlock trees were seen in the four youngest age

classes (90–210 years old), whereas slow-growing hemlock

trees were in the four oldest age classes (210–300+ years

old). Significant differences in hemlock BAI growth rates

were seen between these younger versus older age classes

(P < 0.05), whereas significant differences in blackgum

growth rates were seen only between disparate age classes,

e.g., 30–60 versus 300+ years old (P < 0.05). Hemlock

and blackgum trees in the older age classes also grew signif-

icantly slower than the younger trees at the same respective

age. Both species had a highly variable raw-ring growth in

younger age classes (� 30–120 years old; Figure 5C and D).

Hemlock trees in the 120–300 year age classes typically had

flat raw growth curves, whereas trees in the 300+ year age

class had a slightly declining raw growth in the later years.

Raw-ring growth in blackgum was variable but tended to

be flat over time, although trees in the oldest age classes

(240+ years) had a trend of increasing raw growth.

A comparison between the BAI growth rate and the

tree age for the highest age class of each species indicates

that old pitch pine, blackgum and hemlock trees are the

slowest growing of the eight study species (Figure 6).

Hemlock trees typically grew faster than pitch pine and

blackgum at younger ages, although a large decline in

hemlock growth occurred after age 300. The oldest black-

gum, hemlock and pitch pine are growing significantly

slower than the four oak species (P < 0.05). Bigtooth

aspen in the highest age class typically had the highest

growth rate and the fastest increase in growth as trees

aged (P < 0.05). The oldest trees of the four oak species

were intermediate in growth, but had some spikes in

growth that equaled or exceeded that of bigtooth aspen.

Among the oldest oaks, chestnut oak typically had a

slower growth than the others, whereas black and north-

ern red oak displayed higher growth rates.

There exists the possibility that some of these patterns of

growth could be the result of tree age and site quality inter-

actions, for example, older trees are typically growing on

poorer (rock outcrops with thin soils or bogs/swamps with

organic soils), inaccessible sites. Therefore, we analyzed age

A

Bas

al a

rea

grow

th r

ate

(mm

2 /yea

r)

0

2000

4000

6000

8000

Tree age (years)0

Tre

e rin

g gr

owth

rat

e (m

m/y

ear)

1

2

3

4

30-60 years60-90 years90-120 years120-150 years150-180 years180-210 years210-240 years240-270 years270-300 years300+ years

B

Tree age (years)

0.6

0.8

1.0

1.2

1.4

1.6

1.8

2.0

2.230-60 years60-90 years90-120 years120-150 years150-180 years180-210 years210-240 years240-270 years270-300 years300+ years

C

0

500

1000

1500

2000

D

100 200 300 400 0 100 200 300 400

Figure 4. Chestnut oak decadal average growth rates in BAI (A) and raw-ring widths (B), and white oak decadal average growth rates inBAI (C) and raw-ring widths (D) for each age class. Each decade indicates a 10-year period of averaged growth over all trees in each data set.

6 JOHNSON AND ABRAMS

TREE PHYSIOLOGY

and BAI growth rate relationships for chestnut oak, white

oak, pitch pine, hemlock and blackgum trees growing on

the same site (Figure 7). Chestnut oak trees near Detweiler

Run in central Pennsylvania (Figure 7A) and white oak at

Lake Ahquabi, Iowa (Figure 7B) had increasing growth as

trees aged (with the exception of the oldest white oak),

decreasing growth with increasing age class and younger

trees grew faster than older trees at the same respective

age. Pitch pine trees on the Shawangunk Ridge, New York,

were highly variable in their BAI growth rates both within

and between age classes (Figure 7C). Nonetheless, pitch

trees typically grew faster up until about 200 years of age

and younger pitch pine trees had higher growth rates than

older trees, including a higher growth at the same respective

Tree age (years)

0

Bas

al a

rea

grow

th r

ate

(mm

2 /yea

r)

0

500

1000

1500

2000

Bigtooth aspenBlack oakBlackgumChestnut oakHemlockPitch pineRed oakWhite oak

100 200 300 400 500

Figure 6. Decadal average BAIgrowth trends of the highest ageclasses of each of the studyspecies.

AB

asal

are

a gr

owth

rat

e (m

m2 /y

ear)

0

500

1000

1500

2000

2500

3000

3500

Tree age (years)0 100 200 300 400 500 0 100 200 300 400 500

Tre

e rin

g gr

owth

rat

e (m

m/y

ear)

0.5

1.0

1.5

2.0

2.5

90-120 years120-150 years150-180 years180-210 years210-240 years240-270 years270-300 years300+ years

B

Tree age (years)

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8

2.030-60 years60-90 years90-120 years120-150 years150-180 years180-210 years210-240 years240-270 years270-300 years300+ years

C

0

200

400

600

800

1000

1200

D

Figure 5. Hemlock decadal average growth rates in BAI (A) and raw-ring widths (B), and blackgum decadal average growth rates inBAI (C) and raw-ring widths (D) for each age class. Each decade indicates a 10-year period of averaged growth over all trees in eachdata set.

AGE CLASS, LONGEVITY AND GROWTH RATE RELATIONSHIPS 7

TREE PHYSIOLOGY ONLINE at http://www.treephys.oxfordjournals.org

age as the older trees. Hemlock trees growing at Ramseys

Draft, Virginia (Figure 7D) as well as blackgum growing

at Mohonk State Park, New York (Figure 7E) showed

increasing growth with age within each age class, decreasing

growth from young to old age class and younger trees

growing faster than older trees at the same respective age.

Therefore, similar tree age, longevity and growth rate

relationships exist for these five species when trees of a

species grew on the same site as well as when the trees grew

across different sites throughout their range. A complete

listing of all statistical results can be found in Johnson

(2007; Tables 3–10). Also included are figures containing

standard error bars for each decadal data point across all

age classes and species (Figures 2, 4, 6, 8, 10, 12 and 14).

Discussion

The major results of this study are that (1) BAI is a robust

measure of long-term growth trends for contrasting tree

Detweiler Run Chestnut Oak (A)

0

200

400

600

800

1000

1200

1400

1600

100-150 years 150-200 years 200-250 years 250-300 years 300+ years

Lake Ahquabi State Park White Oak (B)

Gro

wth

rat

e (m

m2 /y

ear)

0

1000

2000

3000

4000

5000

50-100 years 100-150 years150-200 years200-250 years250-300 years300+ years

Shawangunk Ridge Pitch Pine (C)

Tree age (years)0

0

200

400

600

800 50-100 years100-150 years150-200 years200-250 years250-300 years300+ years

100 200 300 400

Figure 7. Decadal averagegrowth rates (BAI) for individualsites of chestnut oak at DetweilerRun in central Pennsylvania (A),white oak at Lake Ahquabi, Iowa(B), pitch pine at ShawangunkRidge, New York (C), hemlock atRamseys Draft, Virginia (D) andblackgum at Mohonk State Park,New York (E).

8 JOHNSON AND ABRAMS

TREE PHYSIOLOGY

species and age classes; (2) an inverse relationship exists

between BAI and increasing age class for all species (both

within and between sites); (3) trees in the oldest age class

for each species grew at the slowest rate throughout their

life, implying an inverse relationship between growth rate

and longevity; (4) the majority of trees in all age classes

had increasing BAI throughout their life, including most

of the oldest trees, which represents a continuation of their

established growth patterns and a deviation from the sig-

moidal growth model; (5) over the last 50–100 years, youn-

ger trees within a species grew faster than did the older trees

when they were of the same respective age; and (6) highly

shade tolerant trees and trees growing on poor sites (e.g.,

rock outcrops and bogs) have an inherently slow growth,

implying a relationship among tree life history, site quality

and growth rate.

The BAI measurements in this study consistently

increased over the life of the trees, including all age classes

from young to old. However, BAI plateaued in the oldest

trees (240–300+ years) of black oak, chestnut oak, white

oak and hemlock over their last 50–150 years of growth.

The increase in growth in young trees is expected from

the sigmoidal growth model because BAI should increase

as young trees produce an increasingly larger leaf canopy

(Spiecker et al. 1996). However, a remarkable finding of

this study is that even the oldest trees of several species

had slow but increasing BAI values, which continued

throughout the life of most trees. This contradicts the

sigmoidal growth model that predicts growth rate should

plateau and then decline, as middle age trees approach

old age (Ryan and Yoder 1997, Weiner and Thomas

2001). Patterns of continued carbon sequestration have

been noted in old-growth forests, indicating a continued

biomass accumulation (Luyssaert et al. 2008); these results

may provide substantiation for such findings. The increas-

ing BAI exhibited by both young and old trees in this study

is also indicated by a quasi-constant raw-ring-width trend,

rather than decreasing over time. The latter case is expected

when a constant amount of wood is distributed over an

increasingly larger tree diameter. A constant raw-ring width

over time means that the tree is producing an increasingly

larger amount of wood (BAI) each year.

Fast growth in young trees versus old trees when they

were of the same age is intriguing yet confounded by

several factors. It is important to note that trees can only

reach the maximum longevity for a species by having slow

growth (i.e., grow slow, live long, or the converse–grow

fast, die young; cf. Schulman 1954, Makela 1986, Kelly

Ramseys Draft Hemlock (D)

0

1000

2000

3000

4000100-150 years 150-200 years 200-250 years 250-300 years 300+ years

Mohonk State Park Blackgum (E)

Tree age (years)

0 100

Gro

wth

rat

e (m

m2 /y

ear)

0

200

400

600

800

1000

100-200 years 200-300 years 300-400 years 400+ years

200 300 400 500

Fig 7. Continued.

AGE CLASS, LONGEVITY AND GROWTH RATE RELATIONSHIPS 9

TREE PHYSIOLOGY ONLINE at http://www.treephys.oxfordjournals.org

et al. 1994, Abrams and Orwig 1995, Larson 2001,

Abrams 2007, Black et al. 2008). Age class-related growth

declines in each species are likely due to a lower propor-

tion of fast-growing trees with increasing age class. Very

old trees are rare within a species, whereas younger trees

are common for most species. Younger trees are repre-

sented by both fast-growing and slower-growing individu-

als, of which only a few slow growers will probably obtain

old age. The proportion of fast-growing trees is higher in

the younger age classes; this trend is reversed in the older

age classes resulting in significant decreases in the average

growth rate (cf. Abrams and Orwig 1995, Orwig et al.

2001, Black et al. 2008). A slow growth rate may come

about from genetics, physiology or from site factors. Site

factors such as poor site quality, deep shade and intense

competition will result in slow tree growth (Nowacki

and Abrams 1997, Pederson 2005, Abrams 2007). Only

a small percentage of young trees will have the right com-

bination of genetics and site conditions (which may

include inaccessibility to logging) to result in slow growth

rates, and thus the opportunity to reach maximum poten-

tial lifespan. Our results suggest that growth rate and

longevity relationships hold true for both between- and

within-site comparisons (Figure 6); however, site condi-

tions are certainly a factor in the analysis.

Ecophysiological differences between species also play a

role in a variety of growth rates and longevities seen over

the study species. For example, bigtooth aspen is an early

successional, shade intolerant species that displays high

growth rates and short lifespan. The oak species studied

here are typically intermediate in shade tolerance and suc-

cessional status, while blackgum and hemlock are highly

shade tolerant and have among the slowest growth rates

and the longest lifespan observed of the study species.

Hemlock and blackgum also have the ability to occupy

nutrient-poor, upper slopes and bogs, where they grow very

slowly and may approach maximum longevity (Abrams

et al. 2001, Abrams 2007). However, poor site quality

resulting in a very slow growth can result in great longevity,

even in early successional trees species such as pitch pine

and northern white cedar (Abrams and Orwig 1995, Larson

2001). This phenomenon can be seen with this analysis, as

the pitch pine and blackgum had slow growth rates on rock

outcrops and bogs, respectively, resulting in a longer life-

span on these types of sites.

Intrinsic differences in growth are shown irrespective of

the climate, as noted in previous studies, because tree

growth rate data before the year 1800 still shows age class

differences (Black et al. 2008). Very old age classes in this

data also have a high proportion of very slow-growing trees

even before the Industrial Revolution, suggesting that old

trees exhibited slow growth throughout their entire life, as

reported in this study. However, a surprising trend is that

older trees in this study had slow but increasing BAI over

the last century or more, which contradicts the sigmoidal

growth model. We do not know the exact cause of the

increasing BAI in the oldest trees; however, it seems reason-

able to assume that it may be due to a stimulatory effect of

anthropogenic global change defined in the broadest sense,

i.e., including both land-use history (land clearing and

changes in natural disturbance regimes) and atmospheric

factors (cf. Innes 1991, Briffa et al. 1998, Voelker et al.

2006). Atmospheric factors include increased CO2 levels,

warming temperatures, increased precipitation, and

changes in precipitation chemistry (Aber et al. 1989, Bazzaz

1990). Yearly average temperatures, atmospheric CO2 and

nitrogen levels have increased in the eastern US (as well

as much of the rest of the world) over the last 50–100 years

(Aber et al. 1989, Lindroth et al. 1993, Nadelhoffer et al.

1999, Korner 2000, Galloway et al. 2003, IPCC 2007).

The fertilizing effect of these factors is present in many,

but not all, tree-ring series as increases in radial growth

(Briffa 1992, Hattenschwiler et al. 1996, Rathgeber et al.

2000, Esper et al. 2002, Voelker et al. 2006). The relation-

ship between tree growth rates and these global change fac-

tors requires much more study to define the various factors

associated with changing global climate and environments,

and the effect that these may have on tree growth. It may be

that global change phenomena, including land-use history,

have increased the growth rate of old trees during the last

century. This leads to the intriguing hypothesis that the

continuation of global change factors that have a stimula-

tory effect on tree growth may act to reduce tree longevity

in the future, as fast-growing trees are less likely to obtain

the maximum longevity for the species.

Funding

Support was provided by the Department of Defense, Ft.

Indiantown Gap National Guard Base and Pennsylvania

State University.

Supplementary data

Supplementary data for this article are available at Tree

Physiology Online.

Acknowledgments

The authors thank Drs. Charles Ruffner, Mary Ann Fajvan, Neil

Pederson, Ed Cook, Bryan Black, Carolyn Copenheaver, Steve

Signell and Rebekah Wagner for research advice and for providing

the tree core sets; Drs. Margot Kaye, Kim Steiner and Jacob

Weiner for ideas on data analysis and for a critical review of the

manuscript; Christine Shook and Glenna Malcolm for field, statis-

tical and graphical help.

References

Aber, J.D., K.J. Nadelhoffer, P. Steudler and J.M. Melillo. 1989.

Nitrogen saturation in northern forest ecosystems. Bioscience

39:378–386.

10 JOHNSON AND ABRAMS

TREE PHYSIOLOGY

Abrams, M.D. 2003. Where has all the white oak gone?

Bioscience 53:927–939.

Abrams, M.D. 2007. Tales from the blackgum, a consummate

subordinate tree. Bioscience 57:347–359.

Abrams, M.D. and G.J. Nowacki. 2008. Native Americans as

active and passive promoters of mast and fruit trees in the

eastern United States. The Holocene 18:1123–1137.

Abrams, M.D. and D.A. Orwig. 1995. Structure, radial growth

dynamics and recent climate variation of a 320-year-old Pinus

rigida rock outcrop community. Oecologia 101:353–360.

Abrams, M.D., C.A. Copenheaver, B.A. Black and S. van de

Gevel. 2001. Dendroecology and climatic impacts for a relict,

old-growth, bog forest in the Ridge and Valley province of

central Pennsylvania, USA. Can. J. Bot. 79:58–69.

Baker, W.L. 1995. Longterm response of disturbance landscapes

to human intervention and global change. Landsc. Ecol.

10:143–159.

Bazzaz, F.A. 1990. Response of natural ecosystems to the rising

global CO2 levels. Annu. Rev. Ecol. Syst. 21:167–196.

Black, B.A., J.J. Colbert and N. Pederson. 2008. Relationships

between radial growth rates and lifespan within North

American tree species. Ecoscience 15:349–357.

Boisvenue, C. and S.W. Running. 2006. Impacts of climate

change on natural forest productivity – evidence since

the middle of the 20th century. Global Change Biol. 12:

862–882.

Briffa, K.R. 1992. Increasing productivity of ‘natural growth’

conifers in Europe over the last century. Proc. Int. Dendrol.

Symp. 1990:64–71.

Briffa, K.R., F.H. Schweingruber, P.D. Jones, T.J. Osborn, I.C.

Harris, S.G. Shiyatov, E.A. Vaganov, H. Grudd and J.

Cowie. 1998. Trees tell of past climates: but are they speaking

less clearly today? [and discussion]. Philos. Trans. R. Soc.

Lond. B Biol. Sci. 353:65–73.

Burns, R.M. and B.H. Honkala. 1990. Silvics of North America.

Agricultural handbook 654. US Department of Agriculture,

Washington, DC.

Carrer, M. and C. Urbinati. 2004. Age-dependent tree-ring

growth responses to climate in Larix deciduas and Pinus

cembra. Ecology 85:730–740.

Cherubini, P., M. Dobbertin and J.L. Innes. 1998. Potential

sampling bias in long-term forest growth trends reconstructed

from tree-rings: a case study from the Italian Alps. For. Ecol.

Manag. 109:103–118.

Cook, E.R and R.L. Holmes. 1984. Program ARSTAN user

manual. Laboratory of tree ring research. University of

Arizona, Tucson, AZ.

Esper, J., E.R. Cook and F.H. Schweingruber. 2002.

Low-frequency signals in long tree-ring chronologies for

reconstructing past temperature variability. Science 295:

2250–2253.

Galloway, J.N., J.D. Aber, J.W. Erisman, S.P. Seitzinger, R.W.

Howarth, E.B. Cowling and B.J. Cosby. 2003. The nitrogen

cascade. Bioscience 53:341–356.

Grissino-Mayer, H.D. 2001. Evaluating crossdating accuracy: a

manual and tutorial for the computer program COFECHA.

Tree-Ring Res. 57:205–221.

Grissino-Mayer, H.D. and H.C. Fritts. 1997. The International

Tree-Ring Data Bank: an enhanced global database serving

the global scientific community. The Holocene 7:235–238.

Hattenschwiler, S., F.H. Schweingruber and C. Korner. 1996.

Tree ring responses to elevated CO2 and increased N

deposition in Picea abies. Plant Cell Environ. 19:1369–1378.

Herms, D.A. and W.J. Mattson. 1992. The dilemma of plants: to

grow or defend. Q. Rev. Biol. 67:283–335.

Holmes, R.L. 1983. Computer-assisted quality control in tree-

ring dating and measurement. Tree-Ring Bull. 43:69–78.

Innes, J.L. 1991. High altitude and high latitude tree growth in

relation to past, present and future global climate change. The

Holocene 1:168–173.

IPCC (Intergovernmental Panel for Climate Change). 2007.

Climate Change 2007: Impacts, Adaptation, and Vulner-

ability. Contribution ofWorking Group II to the Fourth Assess-

ment Report of the IPCC. Eds. M.L. Parry, O.F. Canziani,

J.P. Palutikof, P.J. van der Linden and C.E. Hanson. Cambridge

University Press, Cambridge, UK, 1000 p.

Jacoby, G.C. and R.D. D’Arrigo. 1997. Tree rings, carbon

dioxide, and climate change. Proc. Natl. Acad. Sci. USA

94:8350–8353.

Johnson, S.E. 2007. Comparison of growth rate and longevity

among a variety of eastern tree species. Masters Thesis.

Pennsylvania State University, 68 p.

Kaufmann, M.R. 1996. To live fast or not: growth, vigor, and

longevity of old-growth ponderosa pine and lodgepole pine

trees. Tree Physiol. 16:139–144.

Kelly, P.E., E.R. Cook and D.W. Larson. 1994. A 1397-year

tree-ring chronology of Thuja occidentalis from cliff-faces of

the Niagara Escarpment, southern Ontario, Canada. Can. J.

For. Res. 24:1049–1057.

Korner, C. 2000. Biosphere response to CO2 enrichment. Ecol.

Appl. 10:1590–1619.

Larson, D.W. 2001. The paradox of great longevity in a short-

lived tree species. Exp. Gerontol. 36:651–673.

Lindroth, R.L., K.K. Kinney and C.L. Platz. 1993. Responses

of deciduous trees to elevated atmospheric CO2: producti-

vity, phytochemistry, and insect performance. Ecology 74:

763–777.

Loehle, C. 1988. Tree life history strategies: the role of defenses.

Can. J. For. Res. 18:209–222.

Luyssaert, S., E. Detlef Schulze, A. Borner, A. Knohl, D.

Hessenmoller, B.E. Law, P. Ciais and J. Grace. 2008. Old-

growth forests as global carbon sinks. Nature 455:213–215.

Makela, A. 1986. Implications of the pipe model theory on dry

matter partitioning and height growth in trees. J. Theor. Biol.

123:103–120.

Mann, M.E., R.S. Bradley and M.K. Hughes. 1999. Northern

hemisphere temperatures during the past millennium: infer-

ences, uncertainties, and limitations. Geophys. Res. Lett.

26:759–762.

Martınez-Vilalta, J., D. Vanderklein and M. Mencuccini. 2007.

Tree height and age-related decline in growth in Scots pine

(Pinus sylvestris L.). Oecologia 150:529–544.

Mencuccini, M., J. Martınez-Vilalta, D. Vanderklein, H.A.

Hamid, E. Korakaki, S. Lee and B. Michiels. 2005.

Size-mediated ageing reduces vigour in trees. Ecol. Lett. 8:

1183–1190.

Nadelhoffer, K.J., B.A. Emmett, P. Gundersen, O.J. Kjonaas,

C.J. Koopmans, P. Schleppi, A. Tietema and R.F. Wright.

1999. Nitrogen deposition makes a minor contribution to

carbon sequestration in temperate forests.Nature 398:145–148.

Nowacki, G.J. and M.D. Abrams. 1997. Radial-growth averag-

ing criteria for reconstruction of disturbance histories from

presettlement-origin oaks. Ecol. Monogr. 67:225–249.

Nowacki, G.J. and M.D. Abrams. 2008. The demise of fire and

the ‘mesophication’ of forests in the eastern United States.

Bioscience 58:123–138.

AGE CLASS, LONGEVITY AND GROWTH RATE RELATIONSHIPS 11

TREE PHYSIOLOGY ONLINE at http://www.treephys.oxfordjournals.org

Orwig, D.A., C.V. Cogbill, D.R. Foster and J.F. O’Keefe. 2001.

Variations in old-growth structure and definitions: forest

dynamics on Wachusett Mountain, Massachusetts. Ecol.

Appl. 11:437–452.

Pederson, N. 2005. Climatic sensitivity and growth of southern

temperate trees in the eastern US: implications for the carbon

cycle. Ph.D. Dissertation. Columbia University, 186 pp.

Pederson, N. 2009. Eastern OldList: A database of maximum

tree ages for eastern North America. Rocky Mountain Tree-

Ring Research, Inc. and the Department of Biological

Sciences of Eastern Kentucky University. http://people.

eku.edu/pedersonn/OLDLISTeast/.

Pennisi, E. 2005. Tree growth: the sky is not the limit. Science

23:1896–1897.

Rathgeber, C., A. Nicault, J. Guiot, T. Keller, F. Guibal and

P. Roche. 2000. Simulated responses of Pinus halepensis forest

productivity to climatic change and CO2 increase using a

statistical model. Global Planet. Change 26:405–421.

Rubino, D.L. and B.C. McCarthy. 2000. Dendroclimatological

analysis of white oak (Quercus alba L., Fagaceae) from an

old-growth forest of southeastern Ohio, USA. J. Torrey Bot.

Soc. 127:240–250.

Ryan, M.G. and B.J. Yoder. 1997. Hydraulic limits to tree

height and tree growth. Bioscience 47:235–242.

Schulman, E. 1954. Longevity under adversity in conifers.

Science 119:396–399.

Spiecker, H., K. Mielikainen, M. Kohl and J.P. Skovsgaard.

(Eds.). 1996. Growth trends in European forests: studies from

12 countries. European Forest Institute Research Report

No. 5. Springer, Berlin, Germany.

Stan, A. 2008. Growth release of trees following fine-scale

canopy disturbances in old-growth forests of coastal British

Columbia, Canada. Ph.D. Dissertation. University of British

Columbia, 140 pp.

Stokes, M.A. and T.L. Smiley. 1968. An introduction to tree

ring dating. The University of Chicago Press, Chicago, IL.

Voelker, S.L., R. Muzika, R.P. Guyette and M.C. Stambaugh.

2006. Historical CO2 growth enhancement declines with age

in Quercus and Pinus. Ecol. Monogr. 76:549–564.

Weiner, J. 2004. Allocation, plasticity and allometry in plants.

Perspect. Plant Ecol. Evol. Syst. 6:207–215.

Weiner, J. and S.C. Thomas. 2001. The nature of tree growth

and the age-related decline in forest productivity. Oikos

94:374–376.

Williams, M. 1982. Clearing the United States forests: pivotal

years 1810–1860. J. Hist. Geogr. 8:12–28.

12 JOHNSON AND ABRAMS

TREE PHYSIOLOGY