Dynamic Article LinksC<Journal ofEnvironmentalMonitoringCite this: J. Environ. Monit., 2012, 14, 2438

www.rsc.org/jem PAPER

Publ

ishe

d on

06

July

201

2. D

ownl

oade

d by

Nor

thea

ster

n U

nive

rsity

on

30/1

0/20

14 1

6:11

:34.

View Article Online / Journal Homepage / Table of Contents for this issue

Soil organic carbon sequestration as affected by afforestation: the Darab Kolaforest (north of Iran) case study

Yahya Kooch,a SeyedMohsen Hosseini,b Claudio Zaccone,†*b Hamid Jalilvandc and Seyed Mohammad Hojjatic

Received 30th January 2012, Accepted 5th July 2012

DOI: 10.1039/c2em30410d

Following the ratification of the Kyoto Protocol, afforestation of formerly arable lands and/or

degraded areas has been acknowledged as a land-use change contributing to the mitigation of

increasing atmospheric CO2 concentration in the atmosphere. In the present work, we study the soil

organic carbon sequestration (SOCS) in 21 year old stands of maple (Acer velutinum Bioss.), oak

(Quercus castaneifolia C.A. Mey.), and red pine (Pinus brutia Ten.) in the Darab Kola region, north of

Iran. Soil samples were collected at four different depths (0–10, 10–20, 20–30, and 30–40 cm), and

characterized with respect to bulk density, water content, electrical conductivity, pH, texture, lime

content, total organic C, total N, and earthworm density and biomass. Data showed that afforested

stands significantly affected soil characteristics, also raising SOCS phenomena, with values of 163.3,

120.6, and 102.1Mg C ha�1 for red pine, oak and maple stands, respectively, vs. 83.0 Mg C ha�1 for the

control region. Even if the dynamics of organic matter (OM) in soil is very complex and affected by

several pedo-climatic factors, a stepwise regression method indicates that SOCS values in the studied

area could be predicted using the following parameters, i.e., sand, clay, lime, and total N contents, and

C/N ratio. In particular, although the chemical and physical stabilization capacity of organic C by soil

is believed to be mainly governed by clay content, regression analysis showed a positive correlation

between SOCS and sand (R ¼ 0.86**), whereas a negative correlation with clay (R ¼ �0.77**) was

observed, thus suggesting that most of this organic C occurs as particulate OM instead of mineral-

associated OM. Although the proposed models do not take into account possible changes due to

natural and anthropogenic processes, they represent a simple way that could be used to evaluate and/or

monitor the potential of each forest plantation in immobilizing organic C in soil (thus reducing

atmospheric C concentration), as well as to select more appropriate species during forestation plan

management at least in the north of Iran.

aFaculty of Natural Resources and Marine Sciences, Tarbiat ModaresUniversity, 46417-76489, Noor, Mazandaran, IranbDepartment of Agro-Environmental Sciences, Chemistry and PlantProtection, University of Foggia, Via Napoli 25, I-71122 Foggia, Italy.E-mail: c.zaccone@unifg.it; Web: http://www.claudiozaccone.net; Tel:+39 0881 589119cSari Agricultural Sciences and Natural Resources University, Iran

† Present address: Department of the Sciences of Agriculture, Food andEnvironment.

Environmental impact

Afforestation is considered an option to reduce the concentration o

Data obtained in the present paper are very important to clarify the

scale. The introduction of suitable species is the most important fac

more indicated than other stands (i.e., oak and maple), as it allows

2438 | J. Environ. Monit., 2012, 14, 2438–2446

1. Introduction

Carbon cycles globally among three distinct pools, i.e., the

atmosphere, the ocean, and terrestrial ecosystems. In detail, the

soil C pool contains ca. 2300 Pg of C, most of which (ca. 1550 Pg)

is organic C.1 Being three times the atmospheric pool of 770 Pg

and ca. four times the vegetation pool of 610 Pg,2 soils represent

the largest C reservoir in terrestrial ecosystems, with forest soils

holding about 40% of all belowground C.3 However, approxi-

mately half of soil C in managed ecosystems has been lost to the

f atmospheric CO2 by increasing soil C stocks at the local level.

effect of afforestation on soil C sequestration on a 21 year time

tor. In particular, according to our study, planting of red pine is

an increase by 96% in terms of oil organic C sequestration.

This journal is ª The Royal Society of Chemistry 2012

Fig. 1 (a) Site location of the Darab Kola forest. Parcels 33, 34 and 35

(green circles) were afforested by maple (Acer velutinum Bioss.), oak

(Quercus castaneifolia C.A. Mey.), and red pine (Pinus brutia Ten.),

respectively, and constitute the area of the present study. Parcel 32

(brown circle), that was not afforested, is now covered by herbaceous

species and, in the present study, serves as the control area. (b) Schematic

representation of the experimental design adopted for each stand and for

the control area (figure not to scale).

Publ

ishe

d on

06

July

201

2. D

ownl

oade

d by

Nor

thea

ster

n U

nive

rsity

on

30/1

0/20

14 1

6:11

:34.

View Article Online

atmosphere during the past two centuries due to cultivation

practices.4

The soil C storage potential depends mainly on climate (e.g.,

temperature and precipitation), nature of parent geological

materials and soil features (e.g., texture, pH and pE), vegetation

type (e.g., grassland vs. woodland and broadleaf vs. conifer) and

distribution, and land management practices.3–7 The conscious-

ness that increasing the organic matter (OM) content of soil

contributes to removal of CO2 from the atmosphere has made

soil organic carbon sequestration (SOCS) an important research

topic in environmental sciences, especially in recent years.

Forest plantations have become common landscapes across

many parts of the world. For instance, in 2000, forest plantations

occupied 116million ha inAsia (MohammadnezhadKiasari et al.8

reported that ca. 200 000 ha of degraded forests have been refor-

ested in just the north of Iran), 32 million ha in Europe, 28 million

ha in America, and 8 million ha in Africa.9 A change in land use

from agriculture to forestry implies that the annual cycle of culti-

vating and harvesting crops is replaced by the much longer forest

cycle;10 at the same time, themost evident effect of afforestation on

C sequestration is the net sink for atmospheric CO2 in the growing

biomass.11 Although C sequestration occurs more slowly in soil

than in biomass, C stored in soils would be more resistant to

sudden changes in forest management than C stored in biomass;

consequently, it is extremely important to include changes in soil C

while estimating C sequestration due to afforestation.10

In recent years, policy-makers have been striving to devise

ways to mitigate the effects of rising greenhouse gases (GHGs).

Under the terms of the Kyoto Protocol, plantations established

after 1990 may be counted as offsets to GHG emissions and

contribute to countries meeting their international commitments

to address climate change. For such offsets to be admissible,

amounts of C sequestered after afforestation need to be verified

through a credible C accounting procedure.12

Although C sequestration in forest soils is beneficial to site

productivity and reduction of GHG content in the atmosphere,

the role of afforested types on SOCS is not fully understood.13 In

fact, results present in the literature about the effect of affores-

tation on SOCS are quite contrasting. Some researchers pointed

out that afforestation could, in some cases, decrease the SOCS

ability,14,15 whereas other studies reported the positive effects of

afforestation on SOCS.16–19 Guo and Giffort,20 reviewing the

literature on the influence of land use changes on soil C stocks

and applying a metaanalysis on data obtained from 74 publica-

tions, reported that soil C stocks decline after land use changes

from pasture to plantation (�10%), native forest to plantation

(�13%), native forest to crop (�42%), and pasture to crop

(�59%), whereas they increase after land use changes from native

forest to pasture (+8%), crop to pasture (+19%), crop to plan-

tation (+18%), and crop to secondary forest (+53%).

However, while projections of the amounts of C accumulated

by vegetation are often available or readily predictable, con-

trasting results exist for associated changes in soils.

The present study is an attempt to estimate the SOCS potential

of 21 year old stands of maple (Acer velutinum Bioss.), oak

(Quercus castaneifolia C.A. Mey.), and red pine (Pinus brutia

Ten.) in the Darab Kola region, north of Iran, and to define the

most effective physico-chemical features of soil affecting C

sequestrations after afforestation.

This journal is ª The Royal Society of Chemistry 2012

2. Materials and methods

2.1. Study area: location and history

With an area of 2612 ha, the Darab Kola forest is located in the

north of Iran (south-east of the city of Sari), between 36�280 to36�230 latitude North and 52�140 to 52�310 longitude East

(Fig. 1a). The elevation of the forest area ranges between 180 and

800 m above sea level (a.s.l.).21

According to data of the Gharakheill Meteorological Station,

the mean annual precipitation and temperature were 733 mm and

16.8 �C, respectively. The climate is temperate moist and the dry

months extend from May to September. The soil is forest brown

soil showing a texture that ranges between sandy clay loam to

clay loam.

In the 1980s, this forest was divided into several parcels by the

Forests and Rangelands Organization of Iran (FROI) that were

partially destroyed because of extensive exploitation carried out

by local residents. Consequently, in 1987 these parcels were

‘‘clear-cut’’, stumps eradicated, and then afforested by the FROI

in 1991. The dominant forest types, which were planted at a

spacing of 2 � 2 m, included maple (Acer velutinum Bioss.), oak

(Quercus castaneifolia C.A. Mey.), and red pine (Pinus brutia

Ten.); a natural forest of hornbeam (Carpinus betulus L.) – elm of

Siberia (Zelkova carpinifolia Pall.) with oak (Quercus castanei-

folia C.A. Mey.) is also present.21 However, some parcels were

not afforested in 1991 and are now covered by sparse herbaceous

species including Asperula odorata L., Euphorbia amygdaloides

L., Hypericum androsaemum L., and Polystichum sp. Since 2006,

J. Environ. Monit., 2012, 14, 2438–2446 | 2439

Publ

ishe

d on

06

July

201

2. D

ownl

oade

d by

Nor

thea

ster

n U

nive

rsity

on

30/1

0/20

14 1

6:11

:34.

View Article Online

this forest represents an Experimental Forest Station of the

Mazandaran University.

The treatments investigated in the present research consisted

of 21 year old stands of maple, oak, and red pine (i.e., parcels 33,

34 and 35), whereas barren lands located near the afforested

stands (i.e., parcel 32) were selected as the control region

(Fig. 1a). The study area, located between 240 and 270 m a.s.l.,

shows very similar climatic conditions and management

practices.

Fig. 2 Average values of bulk density (a), water content (b), EC (c), and pH

Standard deviations values are also reported.

2440 | J. Environ. Monit., 2012, 14, 2438–2446

2.2. Soil sampling and analysis

Four hectare areas (200� 200 m) were selected for each stand. In

order to decrease the border effects, surrounding rows of stands

were not considered during sampling.

Soil sampling was carried out during the summer time using a

randomly systematic method. Four soil profiles (50� 50� 40 cm)

were dug along the four parallel transects in the central part of

each afforested stand (Fig. 1b). Soil samples were collected at

(d) in afforested stands (on the left) and at different depths (on the right).

This journal is ª The Royal Society of Chemistry 2012

Table 1 Average total organic C (TOC) content, bulk density, and soilorganic C sequestration (SOCS) values in different stands and at differentdepths

Afforested standSoil depth(cm)

TOC content(g kg�1)

Bulk density(g cm�3)

SOCS(Mg C ha�1)

Maple 0–10 38.2 1.23 47.2010–20 16.2 1.28 21.4020–30 13.7 1.34 18.5530–40 11.0 1.36 14.97Mean 19.7 1.30 25.53Tot 79.1 — 102.12

Oak 0–10 50.1 1.23 61.9210–20 22.9 1.26 29.0020–30 11.9 1.28 15.3230–40 11.1 1.29 14.40Mean 24.0 1.26 30.17Tot 96.0 — 120.64

Red pine 0–10 62.5 1.15 71.8510–20 34.8 1.23 42.8020–30 23.9 1.32 31.2730–40 12.9 1.32 16.92Mean 33.5 1.25 40.71Tot 134.1 — 163.27

Control area 0–10 24.4 1.48 36.2010–20 12.0 1.50 18.0020–30 9.5 1.52 14.4730–40 9.4 1.52 14.35Mean 13.8 1.50 20.75Tot 55.3 — 83.02

Publ

ishe

d on

06

July

201

2. D

ownl

oade

d by

Nor

thea

ster

n U

nive

rsity

on

30/1

0/20

14 1

6:11

:34.

View Article Online

0–10, 10–20, 20–30, and 30–40 cm depths, thus resulting in 64 soil

samples for each stand at four different depths. The same

sampling procedure was carried out also for the control area.

Litter was removed from each profile, as well as large plant

material (e.g., root and shoots) occurring in each soil sample.

Then, soil samples were air-dried and 2 mm sieved.

Each soil sample was characterized with respect to the water

content (i.e., by drying soil samples at 105 �C for 24 hours),

electrical conductivity (EC) (soil : water ratio, 1 : 2), bulk density

(by the clod method22), texture (by the Bouyoucos hydrometer

method23), lime content (by the titration method), pH in water

(soil : water ratio, 1 : 2), total organic C (by the Walkey and

Black method24), and total N (by the Kjeldahl method25).

Earthworms were collected during soil sampling, washed with

deionised water and weighed. Their biomass was then defined as

the weight of the worms after drying for 48 hours on filter paper

at room temperature.26

The following formula was used to calculate C accumulation

at different soil depths:

SOCSL ¼ C � Bd � e � 0.1

where the SOCSL indicates the organic C sequestration at each

soil layer (Mg ha�1);C is the organic C content (g kg�1); Bd is the

bulk density (g cm�3); e is the thickness of the layers (cm), and 0.1

is a conversion factor.

2.3. Elemental composition

Total C and N contents in litter samples were determined in

quadruplicate, using dry combustion with an elemental analyzer

(Fisons EA1108, Milan, Italy). The instrument was calibrated by

the BBOT [2,5-bis-(5-tert-butyl-benzoxazol-2-yl)-thiophen]

standard (ThermoQuest Italia s.p.a.). The obtained data were

corrected for the moisture content.

2.4. Statistical analysis

The normality of the variables was checked by the Kolmogorov–

Smirnov test, while Levene’s test was used to examine the equality

of the variances. Differences in soil characteristics among affor-

ested stands and depths were tested with two-way analysis

(ANOVA) using the General Linear Model (GLM) procedure,

with stands (maple, oak, red pine, and control region) and depths

(0–10, 10–20, 20–30, and 30–40 cm) as independent factors.

Interactions between independent factors were also tested. Dun-

can’s test was used to separate the averages of the dependent

variables which were significantly affected by treatment.

Significant differences among treatment averages for different

parameters were tested at P # 0.05.

A stepwise regression method was used to define the most

important soil features which were effective on organic C seques-

tration. Clearly, in building regressions, all parameters determined

were considered, but those that did not give very relevant results,

and/or that were highly correlated with other parameters already

considered in themodels,were systematically excluded.As a result,

in the first step, sand (%) was introduced into the model; in the

second one, totalN,whereas, in the subsequent steps,C/N ratio (in

the third model), lime (in the fourth model) and clay (in the fifth

model) contents were implemented, respectively.

This journal is ª The Royal Society of Chemistry 2012

SPSS v. 11.5 software was used for all statistical analysis.27

3. Results and discussion

3.1. Soil features

Obtained data indicate that afforested stands significantly

affected most of the soil characteristics under investigation.

Bulk density in the studied stands shows significantly

(P < 0.01) lower values when compared with the control region,

and, as expected, these values increase with depth (Fig. 2a and

Table 1). The observed trend is clearly due to the OM input

coming from litter decomposition that lowered the density of the

upper layers. Because of the higher OM percentage, the water

content is significantly (P < 0.01) higher in the studied stands

when compared with the control, as well as in the top soil layers

rather than in the deeper ones (Fig. 2b). The EC significantly

(P < 0.01) increases in the red pine stand, with the lowest values

observed for the oak stand (Fig. 2c); at the same time, the red

pine stand features also slightly, but significantly (P < 0.01),

lower pH values. In fact, it is well known that conifer litter is

more acidic than deciduous leaf litter,28 thus giving substance to

the theory that litter pH affects soil pH. These results are in

agreement with previous works carried out by several other

authors including Sariyildiz et al.29 and Marcos et al.30

Also physical features, including texture and lime content,

show significantly statistical differences (P < 0.01) both among

stands vs. the control, and with depth (Fig. 3a–d). This suggests a

different evolution of the soil profile when covered by forest,

especially in terms of erosion.

Total organic C and total N show significantly (P < 0.01)

higher values in afforested soils compared to the control, and

J. Environ. Monit., 2012, 14, 2438–2446 | 2441

Fig. 3 Average percentages of sand (a), silt (b), clay (c), and lime (d) in afforested stands (on the left) and at different depths (on the right). Standard

deviation values are also reported.

Publ

ishe

d on

06

July

201

2. D

ownl

oade

d by

Nor

thea

ster

n U

nive

rsity

on

30/1

0/20

14 1

6:11

:34.

View Article Online

their content significantly decreases with depth (Fig. 4a and b).

Highest organic C and total N were found in red pine and maple

stands, respectively. As for pH, also organic C and total N

contents, they are strictly related to the composition of the

original plant material (Table 2). In fact, significantly higher

(P < 0.01) C/N ratios were found in the red pine stand (Fig. 4c);

these results underline the more recalcitrant nature of coniferous

2442 | J. Environ. Monit., 2012, 14, 2438–2446

litter, probably due to the hard cuticle of needles,31 and could

suggest a longer mean residence time of this OM.

Earthworm density and biomass were significantly higher

(P < 0.01) in the maple stand vs. other stands and in deeper layers

(Fig. 5a and b). It is well-known that earthworm distribution and

biomass are affected by changes of vegetation (e.g., differences in

litter quality) and/or soil features (e.g., pH, nutrient availability).

This journal is ª The Royal Society of Chemistry 2012

Fig. 4 Average values of total organic C (a) and total N (b) contents, C/N ratio (c), and (d) soil organic C sequestration (SOCS) in afforested stands (on

the left) and at different depths (on the right). Standard deviation values are also reported.

Publ

ishe

d on

06

July

201

2. D

ownl

oade

d by

Nor

thea

ster

n U

nive

rsity

on

30/1

0/20

14 1

6:11

:34.

View Article Online

In the present research, the highest values of earthworm density

and biomass under the maple stand can be related to a higher

moisture percentage and total N content, as well as to a lower

C/N ratio, whereas litter features and the lower pH of the red

pine stand negatively affected earthworm presence. Because soil

temperature is the factor that affects earthworm migration to the

greatest extent,32,33 and considering the sampling season (i.e.,

This journal is ª The Royal Society of Chemistry 2012

summer), the higher gathering of earthworms in deeper soil

layers was predictable.

3.2. Carbon sequestration by soil

Table 1 shows the amounts of C sequestered by the soil in each

afforested stand and at different depths. Results clearly indicate

J. Environ. Monit., 2012, 14, 2438–2446 | 2443

Table 2 Total C and N content (avg. � st. dev.; n ¼ 4) in litter samples.C/N ratio values have been also reported

Litter type C (%) N (%) C/N

Maple 51.4 � 1.5 2.08 � 0.13 24.9Oak 64.6 � 3.6 0.91 � 0.03 70.9Red pine 67.6 � 3.6 0.80 � 0.04 84.7

Table 3 Stepwise regression analysis of soil organic C sequestration(SOCS, dependent variable) and soil features (independent variables)

Modelno. Regression equation F-value

1 UCa Y ¼ �2.65 + 2.54Sand 179.52c

1 SCb Y ¼ 0.86Sand

2 UC Y ¼ �6.67 + 1.82Sand + 86.5N 158.76c

2 SC Y ¼ 0.62Sand + 0.39N3 UC Y ¼ �25.31 + 0.46Sand + 141.72N

+ 1.82C/N321.87c

3 SC Y ¼ 0.15Sand + 0.65N + 0.48C/N4 UC Y ¼ �14.21 + 0.39Sand + 133.09N

+ 1.75C/N � 3.44Lime267.20c

4 SC Y ¼ 0.13Sand + 0.61N + 0.47C/N� 0.10Lime

5 UC Y ¼ �7.26 + 0.15Sand + 130.66N+ 1.79C/N � 2.98Lime � 0.14Clay

226.94c

5 SC Y ¼ 0.05Sand + 0.60N + 0.48C/N� 0.09Lime � 0.11Clay

6 UC Y ¼ �6.04 + 134.05N + 1.86C/N� 3.0Lime � 0.17Clay

285.87c

6 SC Y ¼ 0.61N + 0.50C/N � 0.09Lime� 0.13Clay

a Unstandardized coefficients. b Standardized coefficients. c Regressionmodel is significant at the 0.01 level.

Publ

ishe

d on

06

July

201

2. D

ownl

oade

d by

Nor

thea

ster

n U

nive

rsity

on

30/1

0/20

14 1

6:11

:34.

View Article Online

that afforestation increased SOCS, although its magnitude

depends on the species utilized, with the red pine stand seques-

tering significantly higher C in soil (P < 0.01) than oak and maple

ones. In detail, SOCS values were 163.27, 120.64 and 102.12 Mg

C ha�1 for red pine, oak and maple stands, respectively, against

83.02 Mg C ha�1 for the control region. Thus, afforestation of

soils with red pine, oak and maple raised SOCS by ca. 96%, 45%,

and 23% when compared with the control region.

3.2.1. Regression models. The stepwise regression method

was used to define SOCS prediction models, as well as the most

important soil features affecting SOCS phenomena (Table 3 and

4). In the first step, sand (%) was introduced into the model (adj.

R2 ¼ 0.74). In the second step, total N was added to the model

and the adj. R2 increased to 0.83. In the subsequent steps, C/N

ratio (in third model), lime (in fourth model) and clay (in fifth

model) contents were implemented, respectively, and the adj. R2

increased to 0.95 (Table 4). Each of the prediction models

featured a significant and linear correlation between SOCS

(dependent variable) and soil features (independent variables)

(Table 3). The models 5 and 6 seem to provide the best estimation

of SOCS considering both the resulting adj. R2 and the standard

error of estimated amounts (Table 3 and 4).

Fig. 5 Average soil earthworm density (a) and biomass (b) in afforested stan

values are also reported.

2444 | J. Environ. Monit., 2012, 14, 2438–2446

This study clearly shows that the quantity (and the quality) of

organic C sequestered by soils depends on chemical and physical

features of the litter-forming plant material. In detail, the red

pine stand shows higher SOCS ability when compared with oak

and maple stands, probably because of both the greater litter

deposition on the soil surface (e.g. ref. 34) and the higher recal-

citrance to decomposition of needle-leaves vs. broad-leaves,35 as

ds (on the left) and at different depths (on the right). Standard deviation

This journal is ª The Royal Society of Chemistry 2012

Table 4 Statistical parameters for predicted models of soil organic Csequestration (SOCS) on the basis of soil featuresa

Model no. R Adjusted R2Standard errorof the estimate

1 0.862 0.739 9.2432 0.916 0.834 7.3833 0.970 0.939 4.4864 0.973 0.944 4.2775 0.975 0.947 4.1596 0.975 0.948 4.142

a Note: the Durbin–Watson test was calculated with a value of 1.98, andthe maximum of variance inflation factor (VIF) was detected less than 6(collinearity statistic) for predicted models.

Publ

ishe

d on

06

July

201

2. D

ownl

oade

d by

Nor

thea

ster

n U

nive

rsity

on

30/1

0/20

14 1

6:11

:34.

View Article Online

also underlined by the C/N ratio values reported in Table 2 (85

vs. 25 in red pine and maple, respectively).

The stepwise regression method underlined how some soil

characters are able to affect the pool of soil organic C. In the

present study, models 5 and 6 can be proposed in the studied area

to estimate SOCS with a good approximation.

Regression analysis shows positive correlation between SOCS

and sand (R ¼ 0.86**), whereas a negative correlation with clay

(R ¼ �0.77**) and lime (R ¼ �0.67**) was observed. The

chemical and physical stabilization capacity of organic C by soil

is believed to be mainly governed by clay content because of both

the increase of specific surface area of mineral particles with

decreasing particle size, and the role of clays in aggregation and

the related indirect effect on enhancing C storage by occluding

organic materials.36–38 Results of the present study, however,

show a different situation underlining a higher SOCS phenom-

enon in soils richer in sand rather than in clay. This could suggest

that most of this organic C occurs as particulate OM instead of

mineral-associated OM.39

Regression analysis shows also positive correlation between

SOCS and total N (R ¼ 0.77**), and between SOCS and C/N

ratio (R ¼ 0.63**). With increasing soil N, growth and produc-

tion will be strengthened and its following C stocks will enhance

in the long term.40 In fact, since N is often the limiting nutrient in

forests, a higher N content generally results in an increase of

wood production and accumulation of soil OM, thus increasing

C sequestration into the forest. However, according to de Vries

et al.,41 with increasing N-enrichment, N immobilization will

decrease (N leaching will increase) and C/N will decline, and

consequently less C could be sequestered per unit of N.

3.2.2. Limits of regression models. As already reported in

Section 2.4, in building stepwise regressions, parameters that did

not result to be very relevant or were highly variable at short-time

scales (e.g., EC, water content, number and biomass of earth-

worms), and/or that were highly correlated with other parame-

ters already considered in the models (e.g., pH, bulk density),

were systematically excluded.

Furthermore, proposed models could be not able to predict

SOCS if changes caused by anthropogenic (e.g., fires) or natural

processes (e.g., climate changes, floodings) occur. For this

reason, and considering that there are many other factors

affecting the extent of changes in soil organic C (e.g., site prep-

aration and management, previous land use, age of afforestation,

This journal is ª The Royal Society of Chemistry 2012

harvesting), the potential for SOCS varies greatly among

different regions and needs to be investigated at a local scale.

3.3. ‘‘Economic benefits’’ from SOCS

The results showed that afforestation increased the organic C

stocks in soil although to a different extent. Average values along

the profile ranged around 40.71, 30.17, and 25.53 Mg C ha�1 for

red pine, oak and maple stands, respectively. Considering that

ca. 27.3% (in weight) of atmospheric CO2 is C, and that

sequestering 1 Mg C means to avoid ca. 3.7 Mg of CO2 being

released to the atmosphere, studied afforestation stands seem to

be quite effective in sequestrating atmospheric CO2.

Besides C cycling, soil quality, and land management, affor-

estation is an important land use change affecting also regional

socioeconomic development.42

Taking into account the whole 0–40 cm profile, and the

afforested areas of red pine (5.5 ha), oak (7 ha) and maple (7 ha)

stands in the Darab Kola region,21 it is possible to estimate the

influence of each stand in terms of SOCS, that is about 898.0,

844.5, and 714.8 Mg C, respectively.

If we consider amean and rational price of 50US$perMgC,43 the

economical valueofSOCSthroughredpine,oakandmaple stands in

the studied region will be 44900, 42225 and 35700 US$, respectively.

4. Conclusions

Today, afforestation is considered an option to reduce the

concentration of atmospheric CO2 by increasing soil C stocks at

the local level, although it could have a significant effect on the

global C budget.

Data obtained in the present paper are very important to clarify

the effect of afforestation on soil C sequestration on a 21 year time

scale. Furthermore, the present work has at least two very strong

aspects: (1) long-term studies are quite scarce in the literature but

very valuable to improve our understanding of SOM dynamics

and afforestation options for sequestering C in soils; and (2) the

study is focused on thewhole soil profile rather than on the surface

layer only, which makes the data even more precious.

In detail, this research elucidated that forest plantation in the

Darab Kola region (north of Iran) has a good potential in terms of

SOCS, although to a different extent. Thus, the introduction of

suitable species is the most important factor for successful forest

plantation. In particular, according to our study, planting of red pine

is more indicated than other stands (i.e., oak andmaple), as it allows

an increase by 96% in terms of SOCS (with respect to the control

region). That is probably due to the more recalcitrant nature of

coniferous litter resulting in a longermean residence timeof thisOM.

The stepwise regression method used to define SOCS predic-

tion models indicated that SOCS values in the studied area are

predictable using the following parameters, i.e., sand, total N, C/

N ratio, lime and clay. Of particular interest was the positive

correlation between SOCS and sand (R ¼ 0.86**), whereas a

negative correlation with clay (R ¼ �0.77**) was observed.

Anyway, because there are many other factors affecting the

extent of changes in soil C, including site preparation, previous

land use, climate, age of afforestation, site management, and

harvesting, the potential for SOCS varies greatly among different

regions and needs to be investigated at a local scale.

J. Environ. Monit., 2012, 14, 2438–2446 | 2445

Publ

ishe

d on

06

July

201

2. D

ownl

oade

d by

Nor

thea

ster

n U

nive

rsity

on

30/1

0/20

14 1

6:11

:34.

View Article Online

References

1 N. H. Batjes, Eur. J. Soil Sci., 1996, 47, 151–163.2 R. Lal, Clim. Change, 2001, 51, 35–72.3 K. I. Paul, P. J. Polglase, J. G. Nyakuengama and P. K. Khanna,Forest Ecol. Manage., 2002, 168, 241–257.

4 B. A.McCarl, F. B. Metting and C. Rice,Clim. Change, 2007, 80, 1–3.5 H. Yan, M. Cao, J. Liu and B. Tao, Agric. Ecosyst. Environ., 2007,121, 325–335.

6 V. Bellassen, R. J. Manlay, J.-P. Ch�ery, V. Gitz, A. Tour�e,M. Bernoux and J.-L. Chotte, Clim. Change, 2009, 98, 213–243.

7 M. Kaul, G. M. J. Mohren and K. V. Dadhwal, Global Change, 2010,15, 489–510.

8 S. H. Mohammadnezhad Kiasari, K. H. Sagheb-Talebi, R. Rahmaniand O. Ghasemi Chapi, Caspian J. Environ. Sci., 2009, 7, 87–98.

9 D. Nsabimana, L. Klemedtson, B. A. Kaplin and G. Wallin, Afr. J.Environ. Sci. Technol., 2008, 2, 142–149.

10 L. Vesterdal, E. Ritter and P. Gundersen, Forest Ecol. Manage., 2002,169, 137–147.

11 IPCC, in Land use, Land Use Change and Forestry, ed. R. T. Watson,I. R. Noble, B. Bolin, N. H. Ravindranath, D. J. Verardo and D. J.Dokken, Cambridge University Press, Cambridge, 2000, p. 375.

12 K. I. Paul, P. J. Polglase and G. P. Richards, Forest Ecol. Manage.,2003, 177, 485–501.

13 M. D. Busse, F. G. Sanchez, A. W. Ratcliff, J. R. Butnor, E. A. Carterand R. F. Powers, Soil Biol. Biochem., 2009, 41, 220–227.

14 H. Neufeldt, D. V. S. Resck and M. A. Ayarza, Geoderma, 2002, 107,151–164.

15 F. M. Groenendijk, L. M. Condron and W. C. Rijkse, Geoderma,2002, 108, 91–100.

16 C. R. Chen, Z. H. Xu and N. J. Mathers, Soil Sci. Soc. Am. J., 2004,68, 282–291.

17 B. Lemma, D. B. Kleja, I. Nilsson and M. Olsson, Geoderma, 2006,136, 886–898.

18 M. D. Nosetto, E. G. Jobbagy and J. M. Paruelo, J. Arid Environ.,2006, 67, 142–156.

19 L. Soto-Pinto, M. Anzueto, J. Mendoza, G. J. Ferrer and B. de Jong,Agrofor. Syst., 2010, 78, 39–51.

20 L. B. Guo and R. M. Gifford, GCB Bioenergy, 2002, 8, 345–360.

21 S. A. Hosseini and H. Jalilvand, Pak. J. Biol. Sci., 2007, 10, 1766–1771.

2446 | J. Environ. Monit., 2012, 14, 2438–2446

22 E. J. Plaster, Soil Science and Management, Delmar Publishers Inc.,Albany, NY, 1985, p. 124.

23 G. J. Bouyoucos, Agron. J., 1962, 54, 464–465.24 A. Walkley and I. A. Black, Soil Sci., 1934, 37, 29–38.25 J. Kjeldahl, Anal. Bioanal. Chem., 1883, 22, 366–382.26 Y. Kooch and S. M. Hosseini, Folia Oecologia, 2010, 37, 181–190.27 SPSS, Statistics Package for the Social Sciences 11.5.0 Base

Applications Guide, SPSS Inc., USA, 2002.28 M. J. Swift, O. W. Heal and J. M. Anderson, Decomposition in

Terrestrial Ecosystems, Blackwell Scientific Publications, Oxford,1979.

29 T. Sariyildiz, J. M. Anderson and M. Kucuk, Soil Biol. Biochem.,2005, 37, 1695–1706.

30 E. Marcos, L. Calvo, J. M. Marcos, A. Taboada and R. T�arrega, Eur.J. For. Res., 2010, 129, 25–30.

31 A. Traversa, V. D’Orazio and N. Senesi, Forest Ecol. Manage., 2008,256, 2018–2028.

32 L. Nachtergale, K. Ghekiere, A. De Schrijver, B. Muys, S. Luyssaertand N. Lust, Pedobiologia, 2002, 46, 440–451.

33 Y. Kooch, H. Jalilvand, M. A. Bahmanyar andM. R. Pormajidian, J.Biol. Sci., 2008, 8, 1033–1038.

34 P. Micks, M. R. Downs, A. H. Magill, K. J. Nadelhoffer andJ. D. Aber, Forest Ecol. Manage., 2004, 196, 71–87.

35 F. J. Stevenson, Cycles of Soil. Carbon, Nitrogen, Phosphorus,Sulphur, Micronutrients., Wiley & Sons, New York, 1986.

36 J. Six, R. T. Conant, E. A. Paul andK. Paustian,Plant Soil, 2002, 241,155–176.

37 A. K€olbl and I. K€ogel-Knabner, J. Plant Nutr. Soil Sci., 2004, 167,45–53.

38 A. F. Plante, R. T. Conant, C. E. Stewart, K. Paustian and J. Six, SoilSci. Soc. Am. J., 2006, 70, 287–296.

39 C. T. Garten Jr, Biomass Bioenergy, 2002, 23, 93–102.40 A. Pussinen, T. Karjalainen, R.Makipaa, L. Valsta and S. Kellomaki,

Forest Ecol. Manage., 2002, 158, 103–115.41 W. de Vries, S. Solberg, M. Dobbertin, H. Sterba, D. Laubhann,

M. van Oijen, C. Evans, P. Gundersen, J. Kros,G. W. W. Wamelink, G. J. Reinds and M. A. Sutton, Forest Ecol.Manage., 2009, 258, 1814–1823.

42 T. K. Rudel, O. T. Coomes, E. Moran, F. Achard, A. Angelsen, J. Xuand E. Lambin, Global Environ. Change, 2005, 15, 23–31.

43 P. C. Ben�ıtez, I. McCallum, M. Obersteiner and Y. Yamagata, Ecol.Econ., 2007, 60, 572–583.

This journal is ª The Royal Society of Chemistry 2012