Deep Soil Horizons: Contribution and Importance to Soil Carbon Pools and in Assessing Whole-Ecosystem Response to Management and Global Change

Abstract

Most of the C in terrestrial ecosystems is found in the soil. Although C calculations indicate that soils are more important than plants as reservoirs of C, soil rarely receives the attention given aboveground ecosystem components when C budgets are calculated. When soil pools are quantified they are typically sampled to relatively shallow depths. Shallow soil sampling in research includes studies that estimate C and nutrient pools and studies assessing the response of terrestrial ecosystems (i.e., forests, grasslands, and agricultural fields) to management treatments. Although many soils have sola that are substantially deeper than 20 cm and C accumulates well below these depths in many soils, the majority of studies of soil C sample to depths of 20 cm or less, generally because of the difficulty and cost of sampling the soil profile deeper. Shallow soil sampling is often justified by assuming that deeper soil horizons are stable and will not change over time, although some medium- and long-term studies do not support this assumption. Shallow soil sampling can result in both a major underestimate of soil C present in the soil profile and an inability to adequately measure the impacts of both treatments for specific goals (i.e., tillage, fertilization, and vegetation management) or other changes (i.e., global change and atmospheric inputs) over time in whole-ecosystem studies. We assessed the potential of shallow soil sampling to underestimate C in the soil profile as well as to change the conclusions of studies of management treatments on soil C. Results showed that where soils were sampled to at least 80 cm or more depth 27-77% of mineral soil C was found >20 cm in depth. In addition, analysis of results for 105 different studies of N fertilization in forests and N fertilization or conversion to switchgrass in agricultural studies shows that deeper sampling can actually change the conclusions of results of some research studies of net C accumulation or loss. Researchers wishing to either quantify soil C pools or measure changes of soil C over time are cautioned to sample soil profiles as deeply as possible and not assume that deeper soil horizons are not a critical part of adequate ecosystem analysis.

Full text

Deep Soil Horizons: Contribution and Importance to Soil Carbon Pools
and in Assessing Whole-Ecosystem Response to Management and
Global Change
Robert B. Harrison, Paul W. Footen, and Brian D. Strahm
Abstract: Most of the C in terrestrial ecosystems is found in the soil. Although C calculations indicate that soils
are more important than plants as reservoirs of C, soil rarely receives the attention given aboveground ecosystem
components when C budgets are calculated. When soil pools are quantified they are typically sampled to
relatively shallow depths. Shallow soil sampling in research includes studies that estimate C and nutrient pools
and studies assessing the response of terrestrial ecosystems (i.e., forests, grasslands, and agricultural fields) to
management treatments. Although many soils have sola that are substantially deeper than 20 cm and C
accumulates well below these depths in many soils, the majority of studies of soil C sample to depths of 20 cm
or less, generally because of the difficulty and cost of sampling the soil profile deeper. Shallow soil sampling
is often justified by assuming that deeper soil horizons are stable and will not change over time, although some
medium- and long-term studies do not support this assumption. Shallow soil sampling can result in both a major
underestimate of soil C present in the soil profile and an inability to adequately measure the impacts of both
treatments for specific goals (i.e., tillage, fertilization, and vegetation management) or other changes (i.e., global
change and atmospheric inputs) over time in whole-ecosystem studies. We assessed the potential of shallow soil
sampling to underestimate C in the soil profile as well as to change the conclusions of studies of management
treatments on soil C. Results showed that where soils were sampled to at least 80 cm or more depth 27–77% of
mineral soil C was found ⬎20 cm in depth. In addition, analysis of results for 105 different studies of N
fertilization in forests and N fertilization or conversion to switchgrass in agricultural studies shows that deeper
sampling can actually change the conclusions of results of some research studies of net C accumulation or loss.
Researchers wishing to either quantify soil C pools or measure changes of soil C over time are cautioned to
sample soil profiles as deeply as possible and not assume that deeper soil horizons are not a critical part of
adequate ecosystem analysis. FOR.SCI. 57(1):67–76.
Keywords: carbon, deep soil, subsoil, soil sampling
EMISSIONS OF CO
2
INTO THE ATMOSPHERE from fos-
sil fuel combustion and other sources continue to
increase globally, with fossil fuel combustion esti-
mated to be 6.3 Tg CO
2
-C in 2006 (US Environmental
Protection Agency 2009). In the United States, forests and
other lands have been significant sinks for atmospheric
CO
2
, due in large part to forest management practices that
result in rapid accumulation of C. It is estimated that ap-
proximately 2.3 Tg of the total emissions of CO
2
-C by fossil
fuel combustion is sequestered by oceans and approxi-
mately 2.3 Tg CO
2
-C by terrestrial systems, with an emis-
sion of approximately 1.6 Tg CO
2
-C from terrestrial sys-
tems due to land-use changes. Terrestrial ecosystems offer a
high potential for increasing C sequestration though altering
management, and, in many cases, those increases can be
brought about by increasing the growth of plants in those
ecosystems, which can have other important benefits to
society. Soil, which provides many of the critical growth
factors for plants in terrestrial ecosystems, represents a
slowly renewable, essential resource in which the more
rapidly renewable plant resources are grown. Currently, soil
provides the basis for production of most of our food and
nearly all of our fiber and wood and is a critical resource for
ecosystems set aside for biodiversity, water production, and
other beneficial ecosystem services. In the future, it is
predicted that terrestrial plant growth will also provide a
larger share of our energy supplies, particularly through the
production of liquid fuels such as ethanol, methanol, and
biodiesel. The production of energy and other resources that
displace the use of fossil fuels has a twofold potential
advantage in mitigating CO
2
emissions to the atmosphere,
first by displacing the combustion of fossil fuels for energy
and also by increasing the amount of C sequestered on a
given area of land over time through increased plant pro-
ductivity of unharvested plant parts and additional associ-
ated residues, particularly underground.
Use of C fixed into plants by photosynthesis represents
“recent” rather than “fossil” C and does not contribute to the
net atmospheric CO
2
increase over time, provided that
plants are regrown sustainably at the same rate on the same
Robert B. Harrison, University of Washington, College of Forest Resources, Box 352100, Seattle, WA 98195-2100 —Phone: (206) 685-7463; Fax: (206)
685-3091; robh@u.washington.edu. Paul W. Footen, University of Washington—pwf@u.washington.edu. Brian D. Strahm, Cornell University—bds92@
cornell.edu.
Acknowledgments: We acknowledge the support of the PNW Stand Management Cooperative and the National Council for Air and Stream Improvement
for financial support in producing this research.
Manuscript received March 25, 2009, accepted July 22, 2010
This article was written by U.S. Government employees and is therefore in the public domain.
Forest Science 57(1) 2011 67

area of land (Birdsey et al. 2006). Production from terres-
trial ecosystems can also be used to produce C-containing
materials that are stored in the long term and displace or
“substitute” for other materials (i.e., wood substituting for
concrete and steel in construction) that result in higher
emissions of fossil CO
2
when produced (Perez-Garcia et al.
2005). Maintaining soil fertility and capability to grow
plants in the long-term thus plays an important role in any
plans to mitigate global CO
2
emissions either through se-
questration or by substituting agricultural and forestry pro-
duction for fossil fuel use directly.
One of the most important properties of soil both in
terms of its C sequestration and potential productivity is
soil organic matter content. Tracking changes in both the
amount of and nature of soil C (as either soil organic
matter or mineral C) is considered critical to understand
changes in soil quality over time and under changing
management regimes (Arshad and Coen 1992, National
Research Council 1993, Doran and Parkin 1994, Manley
et al. 1995, Pikul and Aase 1995, Karlen et al. 1996, Chan
et al. 2002, Tolbert et al. 2002). One estimate is that each
ton of soil organic C per ha may increase crop yields
20–40kgha
⫺1
year
⫺1
for wheat and 10 –20 kg ha
⫺1
year
⫺1
for corn (Lal 2005). Thus, losses in soil C not only
contribute CO
2
directly to the atmosphere but may also
increase the use of fossil fuels to produce fertilizers to
replace lost soil productivity.
Given the critical role of soil C both as a pool of C and
for its role in soil productivity, this synthesis seeks to
evaluate the adequacy of current studies on quantifying soil
carbon pools and tracking changes over time. The current
reality is that because of the difficulties of sampling and
analyzing soil, many researchers studying soil sample rela-
tively shallowly, often to 20 cm or less of soil depth. This
approach may or may not be justified. The objectives of this
article were to review soil studies that can assess the
justification for sampling soil to different depths for
specific purposes such as evaluating C contents and
changes, determining the advantages and disadvantages
of past and current soil sampling for C assessments, and
to providing general guidelines for adequacy of sampling
methodology.
Quantifying Soil Carbon Pools
It is well-known that most global terrestrial ecosystem C
is found in soil pools (Lal 2005), generally with the relative
importance of soil versus vegetation C increasing with
distance from the equator (Dixon et al. 1994). For instance,
Table 1 shows C contents versus depth for 10 individual soil
profiles sampled to at least 80 cm depth for mineral soil
(Rumpel et al. 2002, Harrison et al. 2003, Adams et al.
2005, Ares et al. 2007, Ga´l et al. 2007). Also shown in Table
1 are average soil C contents versus depth for studies that
averaged several soils (Canary et al. 2000, Liebig et al.
2005, P. W. Footen, unpublished observations). When the
equivalent C in the top 20 cm of mineral soil was compared
with the maximum depth sampled in these studies, the 0 –20
cm soil contained 27–76%% of the total C in the sampled
profile, with an average of 44% of C in the 0 –20 cm mineral
soil. Clearly, the results of these studies do not support the
adequacy of sampling shallow soil depths (i.e., 20 cm) in
terms of assessing soil pools of total organic C.
The results of these studies are similar to those seen in
other studies. In an analysis of ⬎2,700 soil profiles in three
global databases (Batjes 1996, Jobba´gy and Jackson 2000)
the 0 –20 cm soil depth contained 42, 33, and 50% of the
total C compared with a depth of 100 cm in shrublands,
grasslands, and forests, respectively. There is also evidence
that in many cases researchers should be sampling to depths
⬎100 cm in some deeper soils, although this is rarely
implemented. For instance, when soil was sampled to 300
cm (Jobba´gy and Jackson 2000), the 0 –20 cm soil depth
only contained 29, 19, and 32% of the total organic C in
shrublands, grasslands, and forests, respectively. It should
be noted that not all soils were 300 cm deep and that this
information was not always available, so some of the soils
sampled to less than 300 cm may have included the entire
soil profile.
Based on the results of all of these studies, shallow soil
sampling to 20 cm would result in major underestimates of
total soil C pools and entire-ecosystem C pools. In some
cases, the C not measured by sampling only to shallow
depths greatly exceeds the entire aboveground C pools
(Homann et al. 2005, Lorenz and Lal 2005, Nicoloso et al.
2009). Given that many published studies sample soils only
to a depth of 20 cm or less, it is clear why shallow soil
sampling has gained such wide popularity among so many
past and current researchers as an adequate measure of soil
C and other factors. First, it is clearly easier and less
expensive to sample and analyze soil to shallow depths
compared with sampling soil much more deeply. When one
directly observes the presence of plant roots in mineral soil
by soil depth, fine roots are typically much more abundant
at shallow depth. Because roots can also hold the mineral
soil material together physically, there often appears to be a
“break” between layers of soil.
The acceptance of shallow sampling is also probably
rooted in history on the basis of the depth to which a typical
moldboard plow would mix soil (Brady 1990). It is impor-
tant to note that much of the early study of soil science was
rooted in agriculture. It is also necessary to reintroduce
nonmetric terms in understanding the likely historical basis
for undersampling the soil profile. Most initial and current
studies of soil fertility were done under controlled condi-
tions in pots, which is still widely done. Because the amount
of nutrient per volume of soil could be calculated, the desire
was to replicate the same concentrations in field studies and
agricultural practice. The term “acre-furrow-slice,” referred
to the amount of soil that was mixed by a typical moldboard
plow, usually about 6 inches or 15 cm in depth. The original
use of the term is uncertain, but its practical definition is
clear (Brady 1990).
Tillage of an acre (approximately 0.4 ha) was thought to
result in approximately 2 million pounds (approximately
900 Mg) of soil mixed per acre. It is thus relatively easy to
calculate how much fertilizer would need to be added to an
acre and mixed into this soil depth to achieve the optimum
concentration of the nutrient. Because of the practical focus
68 Forest Science 57(1) 2011

on the zone of soil that was disturbed and at least partially
mixed, the assumptive depth of agricultural soil sampling
was often based on the acre-furrow-slice. Much of the early
research was based on this depth, and researchers in other
fields were often trained as and/or by agricultural scientists
(including the senior author of this article); thus, it is un-
derstandable that the trend of shallow soil sampling was
continued. Many common crops (i.e., onions) have rooting
systems with relatively shallow depths; however, some (i.e.,
alfalfa) root much more deeply in deeper soils.
It is also becoming clear that this historical sampling
depth, although perhaps adequate for the goals of the re-
search of the time on which it was based and certainly easier
to implement than the alternative of deep soil study, does
not adequately address many of the questions that are
currently being asked by researchers studying soil. For
Table 1. Carbon content of soils sampled to a depth of at least 80 cm
Soil depth
Soil genetic
horizon
Control total
soil C
N-fertilized
soil C
Change in
soil C
Total
soil N
N-fertilized
soil N
Change in
soil N Reference
...........(Mg ha
⫺1
)........... .........(kg ha
⫺1
) .........
Everett very gravelly sandy loam Harrison et
al. 20030–15 cm A 114.0
15–32 cm Bs 31.0
32–48 cm BC 23.0
48–85 cm C1 31.0
85–105 cm C2 17.0
Indianola loamy sand Harrison et
al. 20030–20 cm A 99.0
20–50 cm Bw1 40.0
50–80 cm BC 20.0
80–120 cm C1 14.0
120–180 cm C2 14.0
Boistfort clay loam Ares et al.
200712.3–22.3 cm AB 44.9
22.3–42.3 cm AB 69.4
42.3–62.3 cm B 42.2
62.3–80.0 cm B 24.5
Steinkreuz (Dystric Cambisol) Rumpel et
al. 20022.0–0.0 cm O 17.1
0.0–5.0 cm Ah 42.9
5.0–24.0 cm Bv 24.1
24.0–50.0 cm SwdBv1 7.9
50.0–80.0 cm SwdBv2 4.0
80.0–115.0 cm IIICv 1.4
115.0–150.0 cm IVCv 0.7
Walestein (Haplic Podzol) Rumpel et
al. 20028.5–0.0 cm O 54.6
0.0–10.0 cm Ah 29.0
10.0–12.0 cm Bv 10.4
12.0–30.0 cm SwdBv1 54.7
30.0–55.0 cm SwdBv2 20.9
55.0–70.0 cm IIICv 2.5
70.0–80.0 cm IVCv 1.9
Barneston gravelly silt loam Adams et
al. 20057.5–0.0 cm O 10.9 10.7 ⫺0.2 320 490 170
0.0–12.3 cm A 31.3 28.0 ⫺3.3 850 1,160 310
12.3–62.3 cm Bs 34.0 29.0 ⫺5.0 980 1,300 320
62.3–100.0 cm BC 8.5 7.8 ⫺0.7 470 510 40
Poulsbo gravelly sandy loam Adams et
al. 20052.0–0.0 cm O 18.4 10.0 ⫺8.4 760 390 ⫺370
0.0–3.0 cm E 15.2 24.5 9.3 420 850 430
3.0–21.0 cm Bs 11.7 28.8 17.1 480 1,460 980
21.0–100.0 Bsm 28.9 49.1 20.2 3,520 3,450 ⫺70
Winston ashy loam Adams et
al. 20055.5–0.0 cm O 41.1 35.0 ⫺6.1 1,430 1,510 80
0.0–8.8 cm A 34.9 148.4 113.5 1,030 4,720 3,690
8.8–14.0 cm AB 12.4 65.4 53.0 450 1,990 1,540
14.0–61.8 cm Bs 38.7 133.0 94.3 2,430 5,150 2,720
61.8–100.0 cm 2C 31.6 23.5 ⫺8.1 2,280 1,500 ⫺780
Unnamed Typic Hapludand, silt loam Adams et
al. 20050.0–21.3 cm A 105.4 114.9 9.5 4,470 4,610 140
21.3–33.3 cm AB 42.7 224.8 182.1 1,840 8,060 6,220
33.3–100.0 cm Bw/BC 203.9 79.1 ⫺124.8 8,700 4,150 ⫺4,550
Forest Science 57(1) 2011 69

instance, Hamburg (2000) showed how inadequate soil
sampling gave rise to inaccurate conclusions in several
studies of ecosystem processes and gives three general rules
for adequacy of soil sampling as follows:
1. All soil horizons must be considered (mineral and
organic).
2. Soils must be considered to at least a depth of1mor
the top of the C horizon.
3. Measurements of soil bulk density and carbon concen-
trations must be from the same samples.
These are given as general minimums for adequate soil
assessment, depending on the goals of a particular study.
Assessing Soil Carbon Changes over Time
Despite the importance of soil as a pool of total ecosys-
tem C, soil characterization often gets much less consider-
ation than aboveground C pools in assessments of land
management and other impacts (i.e., global warming) on
ecosystem C. There are many ways in which these changes
can alter the gain/loss/movement of C within a given soil
profile. For instance, Figure 1 shows some basic C fluxes
for a forest ecosystem. Any management change that affects
productivity (i.e., stocking levels, genetic improvement, fer-
tilization, thinning, and competing vegetation control) can
also change the relative amounts of CO
2
input into the forest
by photosynthesis. In addition, increases in atmospheric
CO
2
can have a fertilization effect because C is a nutrient
taken up as CO
2
from the atmosphere and CO
2
exchange
often limits photosynthetic rates (De Luis et al. 1999). There
can also be changes in the cycling of C within the forest,
including differential allocation of photosynthate to below-
ground pools (ain Figure 1). An example would be the
common observation that forest fertilization often causes (or
allows) trees to focus more C in aboveground biomass while
allocating less C to the root system.
Such changes are usually attributed to less need for trees
to develop an extensive rooting system when the soil nutri-
ent availability is higher. In the longer term, however, it is
often noted that after forest fertilization any increased
growth of the trees ultimately results in higher overall rates
of growth in the root system even though a higher percent-
age of total growth is still allocated to aboveground growth
(Friend et al. 1994, Coyle and Coleman 2005). Changes in
direct additions through plant root growth (and mortality) in
the soil profile and litterfall to the soil surface (bin Figure
1) are not the only ways in which changes in conditions can
affect C in soil over time. Any physical, chemical, or
biological change in the nature of the organic matter, the
soil in which it is located (cin Figure 1), or other substances
moving through the soil profile (i.e., gases and moisture)
can affect the rates at which soil C is decomposed, retained,
or leached differentially (d,e, and fin Figure 1) (Kalbitz and
Kaiser 2008).
Where inorganic C is present, organic C already in the
soil and changing conditions can also affect soil respiration
rates. In that case, the CO
2
partial pressure in the soil
Table 1. (Continued)
Soil depth
Soil genetic
horizon
Control total
soil C
N-fertilized
soil C
Change in
soil C
Total
soil N
N-fertilized
soil N
Change in
soil N Reference
...........(Mg ha
⫺1
)........... .........(kg ha
⫺1
).........
Chalmers silty clay loam Gál et al.
20070.0–5.0 cm 22.5* 13.6 ⫺8.9 1,800* 1,100
5.0–15.0 cm 36.4* 27.9 ⫺8.5 2,900* 2,200
15.0–30.0 cm 48.2† 42.8 ⫺5.4 3,800† 3,300
30.0–50.0 cm 33.3† 43.9 10.6 3,000† 3,700
50.0–75.0 cm 16.7 18.2 1.5 1,900 2,000
75.0–100.0 cm 12.2 12.8 0.6 1,600 1,300
31 sites ND, SD, and MN switchgrass
versus previous crops
Control Switchgrass Liebig et al.
2004
0.0–5.0 cm 13.4† 15.0 1.6
5.0–10.0 cm 14.6 13.8 ⫺0.8
10.0–20.0 cm 22.6 23.4 0.8
20.0–30.0 cm 17.1 17.9 0.8
30.0–60.0 cm 38.1† 48.9 10.8
60.0–90.0 cm 33.3† 37.6 4.3
90.0–120.0 cm 36.1 37.0 0.9
Three forest soils averaged: Alderwood gravelly loam, Everett very gravelly sandy loam, Teneriffe loamy sand Canary et al.,
2000
5.25–0.0 cm O 15.0 20.0 5.0
0–1.5 cm 12.0 11.0 ⫺1.0
1.5–26.5 cm 39.0 41.0 2.0
7.3–56.5 cm 24.0 22.0 ⫺2.0
34.0–86.5 cm 15.0 14.0 ⫺1.0
Five forest soils averaged Footen,
unpublished
— O 6.6 8.1 1.5
0.0–10.0 cm 58.1 54.6 ⫺3.5
10.0–50.0 cm 112.2 84.7 ⫺27.5
50.0–100.0 cm 40.0 47.7 7.7
* Means of C or N treatments were significantly different at the 0.01 level.
† Means of C or N treatments were significantly different at the 0.05 level.
70 Forest Science 57(1) 2011

atmosphere and the tendency for soil atmosphere or dis-
solved CO
2
to form carbonates, diffuse out of the soil, or
leach deeper into soil as H
2
CO
3
can be altered. In addition,
fire and harvest removals can also remove C directly from
the site and amendments (i.e., biosolids, pulp and paper mill
residuals, biochar, compost, and wood residues) can add C
directly to the soil, either at the surface or into the soil
profile when incorporated. Changes in aboveground pools
of C affect the production of soluble C compounds that
might leach into the deeper soil horizons (Strahm et al.
2009) and can be used by soil macroorganisms (e.g., earth-
worms, insects, and mites) for energy and as a nutrient
substrate as well as move them physically (Darwin 1882).
These and other mechanisms can act to change the C dy-
namics of the soil and over time and affect the amounts,
distribution, and composition of C within the soil profile.
As mentioned earlier, in some ecosystem studies of soil
C change, when soil is studied at all, it is often sampled
shallowly. Some researchers have considered that subsur-
face soil C is a relatively stable pool, and in certain cases
and conditions it may be. Early evidence offered for the
stability of subsoil C results from studies identifying aver-
age residence times of hundreds or even thousands of years
(Oades 1988, Jenkinson et al. 1992, Righi et al. 1995,
Sollins et al. 1996, Baldock et al. 1997, Paul et al. 1997,
Kaiser and Zech 1998, Nierop and Buurman 1999, Baldock
and Skjemstad 2000, Kaiser and Guggenberger 2000, Chris-
tensen 2001, Kaiser et al. 2002, Rumpel et al. 2002, 2004,
Kiem and Ko¨ gel-Knabner 2003, Baldock et al. 2004, Fang
and Moncrieff 2005, Scho¨ ning and Ko¨gel-Knabner 2006,
Chabbi et al. 2009). Unfortunately, the long residence time
of soil C is partly due to the environment in which the C
existed, and if that environment changes, the content and
nature of soil C that has existed for long periods of time may
rapidly change. This is certainly true for soil minerals,
where rocks that may be millions or even billions of years
old can weather rapidly when exposed to new weathering
conditions. The specific types of ecosystems and soil envi-
ronments most susceptible to relatively rapid changes in
previously stable organic matter are difficult to predict.
Much more research is required to be able to predict con-
tinued organic matter stability or instability.
The introduction of new factors can rapidly change the
movement of C into and loss of C from soil material that has
apparently remained stable for considerable periods of time.
For instance, Fontaine et al. (2007) found that the introduc-
tion of fresh organic substrate primed the rapid decompo-
sition of ancient soil C in soil sampled from 60 – 80 cm soil
depth. They reasoned that microbes in the 60 – 80 cm soil
layer were not able to decompose the stabilized organic
matter alone, but when other, less recalcitrant, organic mat-
ter was added to the deeper soil horizons, the ancient soil C
readily decomposed at more rapid rates.
Lower soil decomposition at depth may also be more of
a factor of the environment in which the organic matter is
stored than its inherent physical and chemical properties.
For example, Fang et al. (2005) found that labile and resis-
tant soil organic matter pools had similar responses to
Figure 1. Mechanisms for differential changes in C changes in the soil profile with depth.
Forest Science 57(1) 2011 71

changes in soil temperature. Risk et al. (2008) found with in
situ studies that organic C decomposition in soil C at 35 cm
was more than 100 times less active than surface soils.
However, when they incubated the subsurface and surface
substrates under the same laboratory incubation conditions,
this large difference in decomposition rates disappeared,
indicating that the soil organic C in subsoils was very
susceptible to higher rates of decomposition when the en-
vironmental conditions under which the soil C stabilized
initially were changed. Other studies have shown significant
variability in response of organic matter retention and loss
when environmental variables are changed (Dixon et al.
1994, Schuman et al. 1999, Prichard et al. 2000, Agren and
Bosatta 2002, Rothe et al. 2002, Swanston et al. 2002, Fierer
et al. 2003, 2005, Torn et al. 2005, Turner et al. 2005, Van
Miegroet et al. 2005).
There is evidence in soil studies that directly measure C
changes over time that both shallow and deep soil horizons
can change relatively rapidly when management practices
are altered and that deep soil sampling may be essential to
understanding the full impacts of any management changes.
For instance, Table 1 includes several studies that measured
changes in soil C associated with management treatments.
Canary et al. (2000), Adams et al. (2005), and P. W. Footen
(unpublished observations) studied the impacts of forest N
fertilization, initially on tree growth, but also quantified the
impacts of N fertilization on soil C over periods of at least
20 years. Overall, study results show a high degree of
variation in forest soil composition but also show that
N-fertilized versus control treatments had observed differ-
ences in subsoil C. In the case of the well-replicated studies
of Ga´l et al. (2007) and Liebig et al. (2005), soil horizons
tested significantly different (Liebig et al. 2005) as deeply
as 90 cm for soil C and as deeply as 50 cm for C and N (Ga´l
et al. 2007).
In a synthesis of the impact of N fertilization on long-
term C sequestration in 48 sites in the United States, Can-
ada, Denmark, England, and Australia (Khan et al. 2007),
less than half of the scientists sampled soil below 20 cm,
and only five researchers sampled below 30 cm. Studies
were initiated as early as 1881, and most research was
performed for several decades or more between the initia-
tion of treatments and final sampling. Unexpectedly, there
was a strong apparent impact of depth of sampling on the
interpretation of results of the study, much greater than the
actual treatments applied. Researchers who sampled soil at
more shallow depths had relatively small differences noted
between N-fertilized and unfertilized sites, whereas there
was an increasingly negative impact of N fertilization on
total C loss with increasing depth (Figure 2).
In the case of the researchers that sampled ⱕ20 cm of
depth, the average of gain/loss of C was relatively small,
with 1.2 Mg C ha
⫺1
lost over study periods of up to 77
years. The researchers that sampled below 20 cm noted a
much greater loss of total soil C, 9.2 Mg C ha
⫺1
(more than
7 times the average for 0 –20 cm) for time periods up to 99
years associated with N fertilization. The single researcher
who sampled to a depth of 120 cm noted the largest total
change due to N fertilization of the entire study of 48 sites,
a loss of 39 Mg C ha
⫺1
, more than 30 times the average of
studies that sampled 0 –20 cm soil depths. Unfortunately,
there were only five studies in this compilation that sampled
below 60 cm and only one below 80 cm.
In a more complete study (all soils were sampled both
shallowly and deeply to 120 cm) of the impact of switch-
grass versus cropland cultivation that included 42 sites, a
similar but opposite trend in C gain/loss was noted with the
switchgrass treatment compared with the other cropland
treatments (Liebig et al. 2005). When a depth of 20 cm was
used as a reference point from which to draw study conclu-
sions, a net increase of 1.6 Mg C ha
⫺1
was estimated,
whereas the estimated increase was greater at 18.4 Mg C
ha
⫺1
when the soil was sampled to 120 cm, more than 10
times more soil C gained overall than if the soil profile was
shallowly sampled. Clearly, both of these studies, which
combined a large number of smaller studies in different
locations, show that the interpretation of results (amounts of
C and changes in C contents by treatment over time) were
highly dependent on the depth to which soil was sampled.
Such studies should serve as a source of caution for any
researchers planning on monitoring the impacts of any
ecosystem pools over time from either land management
treatments aimed to meet particular C goals or from factors
that could change soil C (i.e., invasive plants, air pollution,
global warming, acidic deposition, or others). Unfortu-
nately, even with significant data showing the potential for
changes in deep soil C pools and other properties over time,
many researchers are still not sampling soil deeply enough
to include half of the total soil C pool, and, without this
sampling, the resultant research studies cannot assess the
potential for changes in the deep soil, which often contains
the largest stocks of C in the ecosystems being studied.
Many researchers may be locked into attempting to assess
Figure 2. Impact of soil sampling depth on results of agro-
nomic treatments on soil C loss or gain. A represents 42 paired
switchgrass versus cropland studies in North Dakota, South
Dakota, and Minnesota from Liebig et al. 2005. B represents a
reanalysis of the impact of sampling depth on C loss or gain in
48 sites in the United States, Canada, Denmark, England, and
Australia (Khan et al. 2007).
72 Forest Science 57(1) 2011

whole-ecosystem changes in C after sampling only a frac-
tion of the soil C pool and not sampling soil horizons that
contain both the bulk of ecosystem C as well as horizons
that can change with different management regimes over
relatively modest periods of time.
A current example of the lack of sampling of deep soil as
a pool, source, or sink of C and other constituents is in the
Forest Inventory and Analysis (FIA) National Program (US
Forest Service 2010), which states that it “provides the
information needed to assess America’s forests.” Unfortu-
nately, the protocol for FIA soil sampling does not include
sampling soil below 20 cm (O’Neill et al. 2005) and does
not acknowledge that this may be a problem in assessing
soil pools and changes in soil over time. Clearly, it must be
noted that any monitoring system, including the FIA proto-
col, has limitations, and for the FIA that limitation is a
compromise of lower sampling intensity at a given site to
allow a higher number of sites to be sampled. Unfortu-
nately, and unjustifiably, only a small portion of the total
FIA budget is focused on monitoring of soil, probably the
reservoir of most of the soil C and the basic resource for the
growth of all plants in forest ecosystems. Inadequate fund-
ing is clearly the source of the undersampling of soil in the
FIA. Unfortunately, unrepresentative estimates can be at
best worthless or at worst misleading in making judgments
of the effects of management on soil (Johnson 1992) or
even for initial estimates of the relative amounts of
carbon in ecosystems. We do not mean in this discussion
to single out the FIA for specific criticism; however, it is
the largest single recipient of US funding for forest soil
research, and results of previous research presented indi-
cate that this large expenditure may not be meeting the
FIA’s stated goals.
In particular, taking a large number of consistently bi-
ased samples can result in tight confidence intervals
around inaccurate values and poor estimates of the effects
of management or other impacts (i.e., global warming) on
whole-ecosystem carbon and nutrients. When researchers
report the results of shallow studies of soils, the depth
ranges from which the soil is sampled should be promi-
nently indicated, and discussion of the results of shallow
studies should include caveats of the fact that unsampled
soil zones could contain substantial additional pools and
could have also been subject to differential changes.
Elements of Successfully Quantifying Soil C
Most commonly, the measures of C sought in research
studies are total C pools expressed on a per-area basis (often
MgCha
⫺1
orgCm
2
), which are often scaled to larger areas
by multiplying those per-area pools by the land areas or area
of similar ecosystem types and stages to calculate larger
pools. Likewise, fluxes into and out of pools are often
calculated, although sometimes they are directly measured,
by quantifying C pools at the beginning and end of a study
period, calculating a C flux per hectare per unit time (often
a year) and estimating how a pool might change over a
longer period of time. In most cases, a soil sample repre-
sentative of a particular soil horizon or depth increment is
measured for C concentration and that concentration mul-
tiplied by the soil’s bulk density as
MgCha
⫺1⫽
冉
mg C
g soil
冊
1
soil C
conc.
冉
g soil
cm3soil
冊
1
bulk
density
冉
cm
1
冊
1
soil
depth
冉
Mg C
109mg C
冊冉
108cm2
ha
冊
(1)
In using the techniques of Equation 1, the three most
common errors typically made in quantifying C pools in this
way are to sample soil shallowly and ignore C in the deeper
soil profile, to ignore certain fractions of soil (i.e., soil
particles ⬎2 mm) when analyzing samples for soil C
(Butcher et al. 1984, Huntington et al. 1988, 1989, Canary
1994, Corti et al. 1998, Canary et al. 2000, Plante et al.
2006), and to not obtain an accurate estimate of the soil bulk
density (Butcher et al. 1984, Blake and Hartge 1986, Hun-
tington et al. 1988, 1989, Harrison et al. 1994, Page-
Dumroese et al. 1999). The second and third errors are often
related and are made at the same time, particularly in rocky
and gravelly soil with a large percentage of the soil matrix
⬎2 mm in diameter. Rocky and compacted soils are also
more difficult to sample deeply, which probably leads re-
searchers to sample the soil more shallowly.
For instance, Harrison et al. (2003) found that bulk
density samples were impossible to take accurately by a
widely used soil core technique in an extremely rocky forest
soil (Everett series very gravelly sandy loam). The use of
large core diameters or pit excavation techniques can incor-
porate larger rocks into the determination of soil bulk den-
sity (Jurgensen et al. 1977, Huntington et al. 1988, 1989,
Canary 1994, Page-Dumroese et al. 1999, Canary et al.
2000, Homann et al. 2001). Including the normally dis-
carded ⬎2 mm soil fraction (by pulverizing large rocks to
powder) resulted in a more than doubling of the measured
organic C content of the soil. Rocks in this soil profile had
weathered and developed a porous “rind” that contained
organic C in the approximately 10,000 years since they were
deposited during the last glacial retreat. In this study, using
only the ⬍2 mm soil fraction would have resulted in an
estimate of 81 Mg ha
⫺1
for the 0 –105 cm mineral soil in the
Everett series soil, but when the ⬎2 mm fraction is added,
the estimate increases to 221 Mg ha
⫺1
, a 63% increase.
However, including the ⬎2 mm fraction in the Indianola
series soil only increases the estimate from 207 to 214 Mg
ha
⫺1
, a 3% increase. These results show that the potential
contribution of ⬎2 mm soil cannot automatically be dis-
counted but must be assessed before being ignored in stud-
ies of soil carbon in coarse-textured soils.
Results of a number of other studies also showed that
ignoring the ⬎2 mm soil fraction can result in an underes-
timate of C content, depending on soil conditions. Whitney
and Zabowski (2004), in a study of 17 varied soils, found
that not sampling and analyzing the ⬎2 mm fraction under-
estimated the soil N content between 0.3 and 37%. Al-
though Whitney and Zabowski (2004) did not report C
contents, normally N content and organic C contents are
strongly related, so C would probably be underestimated in
Forest Science 57(1) 2011 73

a similar way. A number of other studies also showed that
the coarse soil coarse matrix is often important for a number
of soil functions and pools (Federer et al. 1993, Ugolini et
al. 1996, Amelung et al. 1998, Dodd et al. 2000, Koele and
Hildebrand 2008). Unfortunately, as for deep sampling,
incorporating coarse-fraction soil C into ecosystem C esti-
mates is very difficult and time-consuming and is not often
implemented. Fortunately, the importance of coarse-soil
fractions contributing to substantial pools of soil C can
potentially be estimated by the quantity of the soil coarse
fraction.
Conclusions and Suggested Soil Sampling
Guidelines
The issue of most soil C analyses not analyzing the entire
soil matrix can be approached by determining whether a
substantial quantity of coarse particles are present. Clearly
the coarse-particle issue is not a problem in all soils. How-
ever, where coarse particles are either abundant and/or have
specific properties (i.e., weathering rinds and porosity),
which may result in higher accumulation of C, the coarse
fraction must be sampled and analyzed to adequately quan-
tify soil C. In all cases, a significant bias is introduced when
the use of techniques that do not sample ⬎2 mm soil
material is forced for acquiring a particular sample (i.e.,
sampling finer material between coarse particles of a soil pit
face), and this should be avoided. In many cases, estimates
of soil C should definitely include sampling and analyzing
soil material ⬎2 mm. Unfortunately, this has only been
done in a very limited number of research studies.
The issue of sampling only the soil surface is serious for
nearly all soils, as most soil profiles are deeper than 15–20
cm. Clearly, the sampling depth of a soil for total ecosystem
estimates of C should not be limited to the approximate
mixing zone of a moldboard plow but should be selected to
correspond to ecosystem realities as much as possible. The
belowground portion of terrestrial ecosystems represents a
difficult pool to estimate accurately. However, based on the
results presented in this synthesis, which show the impor-
tance of deep soil C, this must be done in any study
quantifying ecosystem C pools. In studies that seek to
determine changes in soil C over time, it cannot be assumed
that the deep C pools will not change over time, as they
often do, and in studies of soil change they also must be
sampled and compared. Given the indeterminate nature of
many soil profiles, it is uncertain what the maximum depth
necessary for soil sampling actually is, and it may be
necessary in future studies to determine this measure for any
particular system. We would caution researchers that, al-
though studying and sampling soil deeply may not be easy
to carry out practically, such deep sampling may be abso-
lutely essential for accurate estimates of pools and
processes.
Literature Cited
ADAMS, A.B., R.B. HARRISON, R.S. SLETTEN, B.D. STRAHM, E.C.
TURNBLOM,AND C.M. JENSEN. 2005. Nitrogen-fertilization im-
pacts on carbon sequestration and flux in managed coastal
Douglas-fir stands of the Pacific Northwest. For. Ecol. Manag.
220:313–325.
AGREN, G.I., AND E. BOSATTA. 2002. Reconciling differences in
predictions of temperature response of SOM. Soil Biol. Bio-
chem. 34:129 –132.
AMELUNG, W., W. ZECH,X.ZHANG, R.F. FOLLETT,H.TIESSEN,E.
KNOX,AND K.W. FLACH. 1998. Carbon, nitrogen, and sulfur
pools in particle-size fractions as influenced by climate. Soil
Sci. Soc. Am. J. 62:172–181.
ARES, A., T.A. TERRY, K.B. PIATEK, R.B. HARRISON, R.E.
MILLER,B.FLAMING,C.LICATA,B.STRAHM, C.A. HAR-
RINGTON,R.MEADE,H.W.ANDERSON, L.C. BRODIE,AND J.M.
KRAFT. 2007. The Fall River long term site productivity study
in coastal Washington: Site characteristics, experimental de-
sign, and biomass, carbon and nitrogen stores before and after
harvest. US For. Serv. Gen. Tech. Report PNW-GTR-691,
Portland, OR. 85 p.
ARSHAD, M.A., AND G.M. COEN. 1992. Characterization of soil
quality: Physical and chemical criteria. Am. J. Altern. Agric.
7:25–32.
BALDOCK, J.A., C.A. MASIELLO,Y.G
´
ELINAS,AND J.I. HEDGES.
2004. Cycling and composition of organic matter in terrestrial
and marine ecosystems. Mar. Chem. 92:39 – 64.
BALDOCK, J.A., J.M. OADES, P.N. NELSON, T.M. SKENE,A.
GOLCHIN,AND P. CLARKE. 1997. Assessing the extent of de-
composition of natural organic materials using solid-state
13
C
NMR spectroscopy. Aus. J. Soil Res. 35:1061–1083.
BALDOCK, J.A., AND J.O. SKJEMSTAD. 2000. Role of the matrix and
minerals in protecting natural organic materials against biolog-
ical attack. Org. Geochem. 31:697–710.
BATJES, N.H. 1996. Total carbon and nitrogen in the soils of the
world. Eur. J. Soil Sci. 47:151–163.
BIRDSEY, R., K. PREGITZER,AND A. LUCIER. 2006. Forest carbon
management in the United States: 1600 –2000. J. Environ.
Qual. 35:1461–1469.
BLAKE, G.R. AND K.H. HARTGE. 1986. Bulk density. In Methods
of soil analysis. Part 1: Physical and mineralogical methods,
2nd ed., Klute, A. (ed.). Agronomy Society of America, Inc.,
Madison, WI.
BRADY, N.C. (1990). The nature and properties of soils, 10th ed.
Macmillan, New York, NY.
BUTCHER, B., J. LEUENBERGER, P.J. WIERENGA,AND F. RICHARD.
1984. Preparation of large core samples from stony soils. Soil
Sci. Soc. Am. J. 48:1460 –1462.
CANARY, J.D. 1994. Carbon and nitrogen cycling following re-
peated urea fertilization of second-growth Douglas-fir stands
in western Washington. MS thesis, Univ. of Washington, Se-
attle, WA.
CANARY, J.D., R.B. HARRISON, R.E. EDMONDS,AND H.N. CHAP-
PELL. 2000. Carbon sequestration following repeated urea fer-
tilization of second-growth Douglas-fir stands in western
Washington. For. Ecol. Manag. 138:225–232.
CHABBI, A., I. K ¨
OGEL-KNABNER,AND C. RUMPEL. 2009. Stabilised
carbon in subsoil horizons is located in spatially distinct parts
of the soil profile. Soil Biol. Biochem. 41:256 –261.
CHAN, K.Y., D.P. HEENAN,AND A. OATES. 2002. Soil carbon
fractions and relationship to soil quality under different tillage
and stubble management. Soil Till. Res. 63:133–139.
CHRISTENSEN, B.T. 2001. Physical fractionation of soil and struc-
tural and functional complexity in organic matter turnover. Eur.
J. Soil Sci. 52:345–353.
CORTI, G., F.C. UGOLINI,AND A. AGNELLI. 1998. Classing the soil
skeleton (greater than two millimeters): Proposed approach and
procedure. Soil Sci. Soc. Am. J. 62:1620 –1629.
74 Forest Science 57(1) 2011

COYLE, D.R., AND M.D. COLEMAN. 2005. Forest production re-
sponses to irrigation and fertilization are not explained by shifts
in allocation. For. Ecol. Manag. 208:137–152.
DARWIN, C. 1882. The formation of vegetable mould, through the
action of worms, with observations on their habits. D. Apple-
ton, New York, NY.
DELUIS, I., J.J. IRIGOYEN,AND M. SANCHEZ.-DIAZ. 1999. Elevated
CO
2
enhances plant growth in droughted N
2
-fixing alfalfa
without improving water status. Physiol. Plan. 107:84 – 89.
DIXON, R.K., S. BROWN, R.A. HOUGHTON, A.M. SOLOMON, M.C.
TREXLER,AND J. WISNIEWSKI. 1994. Carbon pools and fluxes
of global forest ecosystems. Science 263:185–190.
DODD, M.B., W.K. LAUENROTH,AND I.C. BURKE. 2000. Nitrogen
availability through a coarse-textured soil profile in the short-
grass steppe. Soil Sci. Soc. Am. J. 64:391–398.
DORAN, J.W., AND T.B. PARKIN. 1994. Defining and assessing soil
quality. P. 3–21 in Defining soil quality for a sustainable
environment, Doran, J.W., D.C. Coleman, D.F. Bezdicek, and
BA Steward (eds.). SSSA Special Publication No. 35. Soil
Science Society of America, Madison, WI.
FANG, C., AND J.B. MONCRIEFF. 2005. The variation of soil mi-
crobial respiration with depth in relation to soil carbon com-
position. Plant Soil. 268:243–253.
FANG, C., P. SMITH, J.B. MONCRIEFF,AND J.U. SMITH. 2005.
Similar response of labile and resistant SOM pools to changes
in temperature. Nature 433:57–59.
FEDERER, C.A., D.E. TURCOTTE,AND C.T. SMITH. 1993. The
organic fraction—Bulk density relationship and the expression
of nutrient content in forest soils. Can. J. For. Res. 23:1026 –1032.
FIERER, N., J.M. CRAINE,K.MCLAUCHLAN,AND J.P. SCHIMEL.
2005. Litter quality and the temperature sensitivity of decom-
position. Ecology 86:320 –326.
FIERER, N., J.P. SCHIMEL,AND P.A. HOLDEN. 2003. Controls on
microbial CO
2
production in surface and subsurface soil hori-
zons of a California grassland. Global Change Biol. 9:
1322–1332.
FONTAINE, S., S. BAROT,P.BARR ´
E,N.BDIOUI,B.MARY,AND C.
RUMPEL. 2007. Stability of organic carbon in deep soil layers
controlled by fresh carbon supply. Nature 450:277–281.
FRIEND, A.L., M.D. COLEMAN,AND J.G. ISEBRANDS. 1994. Carbon
allocation to root and shoot systems of woody plants. P.
245–274 in Biology of Adventitious Root Formation, Davis,
T.D., and B.E. Haissig (eds.). Plenum Press. New York, NY.
G´
AL, A., T.J. VYN,E.MICH´
ELI, E.J. KLADIVKO,AND W.W.
MCFEE. 2007. Soil carbon and nitrogen accumulation with
long-term no-till versus moldboard plowing overestimated with
tilled-zone sampling depths. Soil Till. Res. 96:42–51.
HAMBURG, S.A. 2000. Simple rules for measuring changes in
ecosystem carbon in forestry-offset projects. Mitig. Adapt.
Strat. Glob. Change 5:25–37.
HARRISON, R.B., A.B. ADAMS,C.LICATA,B.FLAMING, G.L.
WAGONER,P.CARPENTER,AND E.D. VANCE. 2003. Quantify-
ing deep-soil and coarse-soil fractions: Avoiding sampling bias.
Soil Sci. Soc. Am. J. 67:1602–1606.
HARRISON, R.B., X. DONGSEN, C.L. HENRY,AND D.W. COLE.
1994. Long-term effects of heavy applications of biosolids to
coarse-textured soils: Nutrients and general soil properties.
For. Ecol. Manag. 66:165–177.
HOMANN, P.S., B.T. BORMANN,AND J.R. BOYLE. 2001. Detecting
treatment differences in soil carbon and nitrogen resulting from
forest manipulations. Soil Sci. Soc. Am. J. 65:463– 469.
HOMANN, P.S., M. HARMON,S.REMILLARD,AND E.A.H. SMITH-
WICK. 2005. What the soil reveals: Potential total ecosystem C
stores of the Pacific Northwest region, USA. For. Ecol. Manag.
220:270 –283.
HUNTINGTON, T.G., C.E. JOHNSON, A.H. JOHNSON, T.G. SICCAMA,
AND D.F. RYAN. 1989. Carbon, organic matter, and bulk density
relationships in a forested Spodosol. Soil Sci. 148:380 –386.
HUNTINGTON, T.G., D.F. RYAN,AND S.P. HAMBERG. 1988. Esti-
mating soil nitrogen and carbon pools in a northern hardwood
forest ecosystem. Soil Sci. Soc. Am. J. 52:1162–1167.
JENKINSON, D.S., D.D. HARKNESS, E.D. VANCE, D.E. ADAMS,AND
A.F. HARRISON. 1992. Calculating net primary production and
annual input of organic matter to soil from the amount and
radiocarbon content of soil organic matter. Soil Biol. Biochem.
24:295–308.
JOBB ´
AGY, E.G., AND R.B. JACKSON. 2000. The vertical distribution
of soil organic carbon and its relation to climate and vegetation.
Ecol. Appl. 10:423– 436.
JOHNSON, D.W. 1992. Effects of forest management on soil carbon
storage. Water Air Soil Pollut. 64:83–120.
JURGENSEN, M.F., M.J. LARSEN,AND A.E. HARVEY. 1977. A soil
sampler for steep, rocky sites. Publ. INT-217. Intermountain
Forest and Range Experimental Station, Ogden, UT. 5 p.
KAISER, K., K. EUSTERHUES,C.RUMPEL,G.GUGGENBERGER,AND
I. K ¨
OGEL-KNABNER. 2002. Stabilisation of organic matter by
soil minerals: Investigations of density and particle-size frac-
tions of two acid forest soils. J. Soil Sci. Plant Nutr. 165:451– 459.
KAISER, K., AND G. GUGGENBERGER. 2000. The role of DOM
sorption to mineral surfaces in the preservation of organic
matter in soils. Org. Geochem. 31:711–725.
KAISER, K., AND W. ZECH. 1998. Dissolved organic matter sorp-
tion by mineral constituents of subsoil clay fractions. Plant
Nutr. Soil Sci. 163:531–535.
KALBITZ, K., AND K. KAISER. 2008. Contribution of dissolved
organic matter to carbon storage in forest mineral soils. J. Plant
Nutr. Soil Sci. 171:52– 60.
KARLEN, D.L., M.J. MAUSBACH, J.W. DORAN, R.G. CLINE, R.F.
HARRIS,AND G.E. SCHUMAN. 1996. Soil quality: A concept,
definition, and framework for evaluation (a guest editorial).
Soil Sci. Soc. Am. J. 61:4 –10.
KHAN, S.A., R.L. MULVANEY, T.R. ELLSWORTH,AND C.W.
BOAST. 2007. The myth of nitrogen fertilization for soil carbon
sequestration. J. Environ. Qual. 36:1821–1832.
KIEM, R., AND I. K ¨
OGEL-KNABNER. 2003. Contribution of lignin
and polysaccharides to the refractory carbon pool in C-depleted
arable soils. Soil. Biol. Biochem. 35:101–118.
KOELE, N., AND E.E. HILDEBRAND. 2008. The ecological signifi-
cance of the coarse soil fraction for Picea abies (L.) Karst.
seedling nutrition. Plant Soil 312:163–174.
LAL, R. 2005. Forest soils and carbon sequestration. For. Ecol.
Manag. 220:242–258.
LIEBIG, M.A., H.A. JOHNSON, J.D. HANSON,AND A.B. FRANK.
2005. Soil carbon under switchgrass stands and cultivated
cropland. Biomass Bioenergy 28:347–354.
LORENZ, K., AND R. LAL. 2005. The depth distribution of soil
organic carbon in relation to land use and management and the
potential of carbon sequestration in subsoil horizons. Adv.
Agron. 88:35– 66.
MANLEY, J.T., G.E. SCHUMAN, J.D. REEDER,AND R.H. HART.
1995. Rangeland and soil carbon and nitrogen responses to
grazing. J. Soil Water Conserv. 50:294 –298.
NATIONAL RESEARCH COUNCIL. 1993. Soil and water quality: An
agenda for agriculture. National Academy Press, Washington,
DC. 516 pp.
NICOLOSO, R.S., C.W. RICE, T.J.C. AMADO,AND J.E. FIORIN.
2009. Deep soil carbon sequestration under no-tillage crop-
ping systems in tropical and temperate climates. IOP Confer-
ence Series: Earth and Environmental Science 6:242030.
NIEROP, K.G.J., AND P. BUURMAN. 1999. Insoluble organic matter
Forest Science 57(1) 2011 75

fractions in incipient podzol B horizons: Preservation of ali-
phatic biopolymers from roots. Humic Subst. Environ. 1:29 –37.
OADES, J.M. 1988. The retention of organic matter in soils. Bio-
geochemistry 5:35–70.
O’NEILL, K.P., M.C. AMACHER,AND C.H. PERRY. 2005. Soils as
an indicator of forest health: A guide to the collection, analysis,
and interpretation of soil indicator data in the Forest Inventory
and Analysis Program. US For. Serv. Gen. Tech. Rep. NC-258.
St. Paul, MN. 54 p.
PAGE-DUMROESE, D.S., M.F. JURGENSEN, R.E. BROWN,AND G.D.
MROZ. 1999. Comparison of methods for determining bulk
densities of rocky forest soils. Soil Sci. Soc. Am. J. 63:379 –383.
PAUL, E.A., R.F. FOLLETT, S.W. LEAVITT,A.HALVORSON, G.A.
PETERSON,AND D.J. LYON. 1997. Radiocarbon dating for de-
termination of soil organic matter pool sizes and dynamics. Soil
Sci. Soc. Am. J. 61:1058 –1067.
PEREZ-GARCIA, J., B. LIPPKE,J.COMNICK,AND C. MANRIQUEZ.
2005. An assessment of carbon pools, storage, and wood prod-
ucts market substitution using life-cycle analysis results. Wood
Fiber Sci. 37:140 –148.
PIKUL, J.L., AND J.K. AASE. 1995. Infiltration and soil properties as
affected by annual cropping in the northern great plains. Agron.
J. 87:656 – 662.
PLANTE, A.F., R.T. CONANT, C.E. STEWART,K.PAUSTIAN,AND J.
SIX. 2006. Impact of soil texture on the distribution of soil
organic matter in physical and chemical fractions. Soil Sci. Soc.
Am. J. 70:287–296.
PRICHARD, S.J., D.L. PETERSON,AND R.D. HAMMER. 2000. Carbon
distribution in subalpine forests and meadows of the Olympic
Mountains, Washington. Soil Sci. Soc. Am. J. 64:1834 –1845.
RIGHI, D., H. DINEL, H.-R. SCHULTEN,AND M. SCHNITZER. 1995.
Characterization of clay-organic-matter complexes resistant to
oxidation by peroxide. Eur. J. Soil Sci. 46:423– 429.
RISK, D., L. KELLMAN,HBELTRAMI,AND A. DIOCHON. 2008. In
situ incubations highlight the environmental constraints on soil
organic carbon decomposition. Environ. Res. Lett. 3:1–5.
ROTHE, A., K. CROMACK,JR., S.C. RESH,E.MAKINECI,AND Y.
SON. 2002. Soil carbon and nitrogen changes under Douglas-fir
with and without red alder. Soil Sci. Soc. Am. J. 66:1988 –1995.
RUMPEL, C., K. EUSTERHUES,AND I. K ¨
OGEL-KNABNER. 2004.
Location and chemical composition of stabilized organic car-
bon in topsoil and subsoil horizons of two acid forest soils. Soil
Biol. Biochem. 36:177–190.
RUMPEL, C., I. K ¨
OGEL-KNABNER,AND F. BRUHN. 2002. Vertical
distribution, age, and chemical composition of organic carbon
in two forest soils of different pedogenesis. Org. Geochem.
33:1131–1142.
SCH ¨
ONING, I., AND I. K ¨
OGEL-KNABNER. 2006. Chemical composi-
tion of young and old carbon pools throughout Cambisol and
Luvisol profiles under forests. Soil Biol. Biochem. 38:2411–2424.
SCHUMAN, G.E., J.D. REEDER,JTMANLEY,RHHART,AND W.A.
MANLEY. 1999. Impact of grazing management on the carbon
and nitrogen balance of a mixed-grass Rangeland. Ecol. Appl.
9:65–71.
SOLLINS, P., P. HOMANN,AND B.A. CALDWELL. 1996. Stabilization
and destabilization of soil organic matter: Mechanisms and
controls. Geoderma 74:65–105.
STRAHM, B.D., R.B. HARRISON, T.A. TERRY, T.B. HARRINGTON,
A.B. ADAMS,AND P.W. FOOTEN. 2009. Changes in dissolved
organic matter with depth suggest the potential for postharvest
organic matter retention to increase subsurface soil carbon
pools. For. Ecol. Manag. 258:2347–2352.
SWANSTON, C.W., B.A. CALDWELL, P.S. HOMANN,L.GANIO,AND
P. SOLLINS. 2002. Carbon dynamics during a long-term incu-
bation of separate and recombined density fractions from seven
forest soils. Soil Biol. Biochem. 34:1121–1130.
TOLBERT, V.R., D.E. TODD,JR., L.K. MANN, C.M. JAWDY, D.A.
MAYS,R.MALIK, W., BANDARANAYAKE,A.HOUSTON,D.
TYLER,AND D.E. PETTRY. 2002. Changes in soil quality and
below-ground carbon storage with conversion of traditional
agricultural crop lands to bioenergy crop production. Environ.
Pollut. 116:S97–S106.
TORN, M.S., P.M. VITOUSEK,AND S.E. TRUMBORE. 2005. The
influence of nutrient availability on SOM turnover estimated by
incubations and radiocarbon modeling. Ecosystems 8:352–372.
TURNER, J., M.J. LAMBERT,AND D.W. JOHNSON. 2005. Experience
with patterns of change in soil carbon resulting from forest
plantation establishment in eastern Australia. For. Ecol.
Manag. 220:259 –269.
UGOLINI, F.C., G. CORTI,A.AGNELLI,AND F. PICCARDI. 1996.
Mineralogical, physical and chemical properties of rock frag-
ments in soil. Soil Sci. 161:521–542.
US ENVIRONMENTAL PROTECTION AGENCY. 2009. 2009 U.S.
Greenhouse Gas Inventory Report 1990 –2007. Available on-
line at www.epa.gov; last accessed Dec. 10, 2010.
US FOREST SERVICE. Forest Inventory and Analysis National Pro-
gram. Available online at www.fia.fs.fed.us; last accessed Dec.
10, 2010.
VAN MIEGROET, H., J.L. BOETTINGER, M.A. BAKER,J.NIELSEN,
D. EVANS,AND A. STUM. 2005. Soil carbon distribution and
quality in a montane rangeland-forest mosaic in northern Utah.
For. Ecol. Manag. 220:284 –299.
WHITNEY, N., AND D. ZABOWSKI. 2004. Total soil nitrogen in the
coarse fraction and at depth. Soil Sci. Soc. Am. J. 68:612– 619.
76 Forest Science 57(1) 2011