1. business and industrial
2. agriculture and forestry
3. crops and seed

Century scale environmental reconstruction using stable carbon
isotopes: just one method from the big bag of tricks
Bradd Witt
School of Natural and Rural Systems Management
The University of Queensland Gatton Campus, Queensland 4343 Australia
e-mail: bwitt@uqg.uq.edu.au
Abstract/summary
The field of ‘historical ecology’ is coming to maturity at a time when we, in Australia,
are reflecting on our relationship with, and place in, the land. After an essentially
ahistorical approach to land use we are now attempting to place land management into
the context of environmental change since, and immediately preceding Western
European settlement. This volume reflects an emerging concern that collectively, nonindigenous Australians have no ‘environmental history’.
One component of ‘living in’ rather than ‘battling against’ the land is developing a
sense of our history. Without an oral narrative that is commonly shared, attempts to
develop the story of environmental change have to be based on retrospective and
reconstructive research. This volume captures part of this movement to develop an
environmental narrative and context for our future relationship with the land.
There are many methodological approaches to reconstructing a story of the past, from
local knowledge and oral history to the high-tech and hard sciences. This paper reviews
methods that apply stable carbon isotope techniques to reconstruct environmental
change. Although well suited to environmental history, carbon isotope techniques
remain under-utilised in the Australian context. Here I review applications to highlight
the strengths, and limitations, of carbon isotope techniques in the reconstruction of
century scale vegetation change. There have been two dominant applications of carbon
1
isotope techniques to environmental reconstruction. These applications fall broadly
under either stable carbon isotope analysis of organic matter in soils and sediments, or
alternatively, inferences of environmental states have been drawn from carbon isotope
analysis of animal tissues and residues. The main strength of stable carbon isotope
techniques is that they can be spatially precise while integrating a range environmental
information into one isotopic signal. This integrative strength is at the same time one of
the major limitations of carbon isotope techniques because floristic (taxonomic)
resolution is low.
Introduction
Imagine being able to reconstruct the diet of long extinct megafauna and attributes of
their environment by analysing a few milligrams of fossil tooth; to track the global
mid-Miocene expansion of C4 grasslands and savannas from soil carbon; or to
determine the metabolic turnover of animals from a few breath samples. Such insights
into biological processes spanning minutes to millions of years have been made
possible through the application of stable carbon isotope methods. Reconstruction of
landscape change over periods beyond the range of direct observational records, from
decades to millennia, has to be based on proxy records. What the researcher must
search for is an ‘indicator’ and or ‘integrator’ of past environmental states or processes.
This review focuses on the use of carbon isotope ratios as the indicator and integrator
of such processes.
However, it is not my intent to champion the ‘cause’ of stable isotopic methods as the
principal tool for environmental reconstruction. Rather I will emphasise the need for
critical use (choice) of methods appropriate to the issue under investigation. Also, that
environmental reconstruction should use as many methodological approaches as
possible. Triangulation using multiple techniques brings about both an openness to the
framing of the research problem and to the interpretation of data.
2
There is no natural scale for studying ecosystems. This issue is central to ecological
research and highlights the need to keep our own perspective (our own scale of
thinking) in mind when studying environmental phenomena (Levin 1992). Just as there
is no natural scale, there is also no natural method for studying ecosystems. The
methodological framework and methods we use to examine, analyse, interpret and
understand ecosystems are more often driven by our training or expertise in a specific
field. Often, the tools that we instinctively reach for (both instrumental and statistical)
when confronted with ecological problems are those with which we are already
familiar. This means our initial outlook and scales of investigation largely frame our
interpretation. Historical and palaeo-ecology often suffer from the limitations imposed
by each researcher’s expertise and experience.
The papers in this volume represent a wonderful review of many methods available to
reconstruct environmental change. But rather than thinking of them as a metaphorical
smorgasbord laid out for us to select our favourite method, we should step back from
the table and recognise the diversity of methods on offer. When combined rather than
used in isolation, this range of methods has the potential to revolutionise our
understanding of environmental change.
Scope and outline of the review
This paper is aimed primarily at those not familiar with the application of stable
isotopic methods in historical and palaeo-ecological investigation. For this reason it is
not intended to be an exhaustive review of the literature. There are many specific and
generalist reviews of isotopic methods and their application to a range of environmental
fields (Stout et al. 1975; van der Merwe 1982; Ehleringer et al. 1986; Rounick and
Winterbourn 1986; DeNiro 1987; Peterson and Fry 1987; O’Leary 1988; Rundel et al.
1989; Coleman and Fry 1991; Tieszen 1991; Tiezsen and Archer 1991; Lajtha and
Michener 1994; Wolf et al. 1994; Boutton 1996; Gannes et al. 1997; Pate 1997;
Hobson 1999). My purpose is to make the method more accessible by providing an
3
overall introduction to what carbon isotope techniques have to offer, and, as
importantly, highlight some limitations.
The remainder of this introductory section will review some fundamentals of isotopic
variability and the nomenclature used to report isotope ratios. In the broadest sense,
isotopic variation in carbon, whether from sediments and soil or from the tissue and byproducts of organisms, can provide clues to environmental change. Thus, the body of
the paper has two broad sections 1) a review of applications using carbon from soils
and sediment, and 2) the use of isotopes in animals and animal residues as
environmental integrators. The most widely applied technique used to date in century
scale vegetation reconstruction has been the study of carbon in soil organic matter, and
this will form the bulk of the review. The other techniques have been included because
of their potential in historical ecology.
Some background to stable isotopic techniques
Carbon is an essential ingredient for life on this planet. Like most elements, carbon
comes in more than one isotopic form. All carbon nuclei have twelve protons but the
number of neutrons can vary, giving rise to the three isotopes 14C, 13C, and 12C. It is the
subtle difference in mass between these three isotopes of carbon that permit
environmental processes to be traced. The heaviest of the carbon isotope family is the
'unstable' radiocarbon (14C). Because radiocarbon decays at a known rate, dating of
organic matter up to about 50 000 years old is possible. Carbon dating is a technique
familiar to scientists and non-scientist alike. However, it is the stable isotopes (13C and
12
C) that are relevant to this review.
Almost all carbon in terrestrial organisms is derived from atmospheric CO2. Carbon
from the atmosphere is utilised by plants during photosynthesis to build life. Plants
store the carbon for some time and then it may be consumed by animals, incorporated
into soil organic matter or released back into the atmosphere. In time most carbon
returns to the atmosphere although some carbon may be stored for many thousands or
4
millions of years in sedimentary sequences. Each step of this carbon cycle, from
photosynthesis to oxidation back into the atmosphere, has some impact on the ratio of
13
C to 12C in the resulting material. The isotopic composition of substances can
therefore provide insights into 1) the processes and source materials forming those
substances, 2) the rates of those processes, and/or 3) the prevailing environmental
conditions during formation (Boutton 1996).
Units of measure and reporting in isotopic studies
The absolute isotopic variation between substances of interest to the biological sciences
is very small, usually less than one percent and high precision mass spectrometers are
required for analysis (see Boutton 1991). Ratios of the rare (‘heavy’) 13C to the
common (‘lighter’) 12C in samples are reported using “δ” (delta) notation in units of
parts per thousand (‰). The δ13C symbol indicates the ratio of the isotopes (13C:12C)
relative to the ratio of a standard that has a defined value of 0‰ (see equation below).
For carbon the international standard is the Vienna Pee Dee Belemnite (V-PDB).
⎛ ( 13 C/ 12 C) sample - ( 13 C/ 12 C) standard
δ C (‰) = ⎜⎜
( 13 C/ 12 C) standard
⎝
13
⎞
⎟ × 1000
⎟
⎠
Isotopic variability within and between plants
The carbon isotope composition of plants ranges from around -9‰ to -32‰ (O’Leary
1988). Because plants discriminate against 13C during photosynthesis they are
isotopically more depleted in 13C than atmospheric CO2 which has a δ13C of around 8‰ (for a detailed review see Tieszen 1991). Smaller changes to the δ13C of plants can
result from abiotic conditions such as water availability/stress, light intensity, and the
concentration of CO2, and biotic factors such as growth form and genetic variation.
These modifying factors have permitted a wide range of eco-physiological studies such
5
as water use efficiency in plants (Boutton 1991 and Tieszen 1991). However, it is
photosynthetic pathway that is primarily responsible for determining the isotopic
composition of most organic carbon in materials of interest to the environmental
sciences.
Plants that use the C3 photosynthetic pathway have δ13C values in the range of -22 to 32‰ (usually -25 to -29‰). Plants that use the C4 photosynthetic pathway have δ13C
values in the range of -9 to -20‰ (usually -12 to -14‰). CAM (Crassulacean Acid
Metabolism) plants tend to have intermediate δ13C values (O’Leary 1988). It is the
major difference in δ13C value between C3 and C4 plants and the subsequent impact that
this difference has on the isotopic composition of microorganisms, herbivores and soil
organic matter that form the basis of most carbon-isotope studies of environmental
change. CAM photosynthesis is limited to several succulent species such as cacti and
is not a significant component of Australasian ecosystems. For this reason CAM
photosynthesis is not considered further in this review.
Most plants use the C3 photosynthetic pathway. This pathway is believed to be the first
photosynthetic system to have evolved, and for this reason is widely represented in
plant families. C3 plants include most trees, shrubs, and herbaceous plants of temperate
and non-water stressed environments. C4 photosynthesis is best represented by the
grasses of tropical and semi-arid environments. However, there are some
dicotyledonous families that include C4 species, such as Amaranthaceae and
Chenopodeaceae (for details see Moore 1982; Ehleringer et al. 1997). Where mean
annual temperature is at or above 21°c, C4 becomes increasingly dominant in the grass
flora (Bird and Pousai 1997). It is temperature during the growing season that is
primarily responsible for the distribution and dominance of C4 photosynthesis (see
Terri and Stowe 1976; Hattersley 1983; Ehleringer et al. 1997), although the degree of
water stress is also a factor. This means a relatively predictable pattern of C3 and C4
species distribution exists with latitude and altitude. Hattersley (1983) demonstrated
that the distribution of C4 grasses in Australia corresponds very well with temperature,
6
and therefore latitude. Where most rainfall occurs in summer, C4 photosynthesis
becomes increasingly dominant in the grass flora, and by about the Tropic of
Capricorn, 90% or more of the grasses are C4. (Fig. 1). There is very good agreement
between the percentage of C4 taxa and their contribution to herbaceous biomass.
[Fig. 1 near here]
The isotopic distinction between C3 and C4 plants, and its impact on the isotopic
composition of organic carbon in the soil has now been used widely in many ecological
studies aimed at determining dynamic shifts between grassy and woody vegetation at
time scales relevant to this volume. The reconstruction of vegetation change using
carbon isotope techniques is rare in Australia (Quade et al. 1995). This lack of
application in the Australian setting is surprising because one of the earliest studies was
carried out on the Darling Downs (south east Queensland) by a group of New Zealand
researchers for the purpose of reconstructing the environmental context and age of a
human skull (Hendy et al. 1972). Several applications of carbon isotope techniques to
soil and animal residues/tissues will be discussed below to highlight the principles as
used in environmental reconstruction.
Two broad applications: Organic matter as a tracer of biomass change
and animal residues and tissues as bio-incorporators
δ13C of soil organic matter to detect in situ vegetation change
Soil organic carbon is principally derived from the vegetation that the soil has
supported over time. The carbon isotope ratio of soil organic matter and vegetation are
similar under stable conditions. However, shifts in isotopic composition of the
vegetation (eg. forest invasion of a tropical savanna) are recorded in the soil in a
reasonably predictable way. There is a general trend of increasing age of organic
7
carbon with soil depth (Boutton 1996). Surface soil has a characteristically rapid
turnover of carbon relative to soil at depth (discussed later). This simple model has
been applied widely in carbon isotope analysis of soil to determine in situ changes in
vegetation over time, especially where changes in tree/grass dynamics are of interest.
Figure 2 shows conceptually the factors that govern the stable isotope composition of
soil organic carbon.
[Fig. 2 near here]
Application of this principle has increased rapidly since the mid-1970s. Because of the
power to detect C3 / C4 change it has been applied widely in tropical and semi-arid
areas (Schwartz et al. 1986; Tieszen and Archer 1991; McPherson et al. 1993; Bond et
al. 1994; Nordt et al. 1994; Archer 1995; Victoria et al. 1995; Desjardins et al. 1996;
Boutton et al. 1999; Bowman and Cook 2001).
Rather than attempt to review the now burgeoning literature on stable carbon isotope
studies of soil organic matter in environmental reconstruction, I have selected two case
studies that indicate the application of the technique to specific issues in environmental
change.
Shrub encroachment in semi-arid rangelands
The issue of anthropogenic versus natural change in semi-arid rangelands is of global
concern. It is widely held that the introduction of domestic stock, artificial waters and
altered fire regimes has converted many once grass-dominated semi-arid subtropical
and tropical rangelands into woody ecosystems (Archer 1994, 1995). Although it is
often asserted that this form of degradation/desertification has been human induced
(Ludwig and Tongway 1995), it is difficult to determine the extent to which broad
fluctuations in woody versus grassy dominance are driven by natural processes or by
human use (Archer 1996). The same issues of uncertainty about past ecosystem
structure is echoed in the debate over the extent to which recent ‘tree thickening’ has
occurred in many parts of eastern Australian. This issue is aggravated not only by
8
debate over acceptable levels of ‘tree clearing’ for the purposes of agricultural
production but also by the increasing focus on the nexus between CO2 emissions and
the agricultural sector (Fensham and Holman 1999). In semi-arid rangelands it is
argued by some that although grazing is a factor in vegetation change, natural climatic
variation has driven fluctuations in woody dominance, and that there is no reliable
evidence for the grass dominated systems that are believed to have existed in the past.
Fundamentally, we have little evidence one way or the other, due to poor records, and
little detailed reconstruction over the last few centuries (Witt 1998).
South African case study: the eastern Karoo
On the southern African sub-continent, and for well over a century, there has been
concern over apparent desertification, increasing ‘shrubbiness’ and loss of perennial
grass in the eastern Karoo (Shaw 1875; Schwarz 1919; Acocks 1953, 1979; Roux and
Vorster 1983; Dean and MacDonald 1994). Isotopic assessment of soil organic matter
from locations across the eastern Karoo has been conducted in an attempt to test the
hypothesis that there has been an increase in shrubby vegetation (Bond et al. 1994;
Hoffman et al. 1995). Soil was sampled from single pits (ranging in depth from 0.2 to
0.6 m) at 11 sites across the region where most change is believed to have occurred.
Changes in the δ13C of soil organic matter indicated that there had been some ‘recent’
trend to increased C3 vegetation at some sites (greater than 6‰ difference between
‘old’ soil and ‘recent’ surface soil organic matter). However, Bond et al. (1994)
concluded that the region has most likely been a broad transition zone between the C3
dominated shrubby vegetation and C4 savanna systems. The mean difference between
surface δ13C and older soil organic matter from depth was less than 2‰ across all sites.
In essence, the transition zone is a reflection of the gradient from arid winter-dominant
rain in the south west to more mesic and summer dominant conditions in the north east.
It was considered that only in the north had “true invasion” by Karoo shrubs occurred.
These authors point out that an interest in historical trends, and their conclusions based
on isotopic assessment, is not solely an academic debate (Hoffman et al. 1995). In
South Africa (and other countries) significant resources are channelled into policy and
9
management aimed at ‘reversing’ desertification and degradation. Restoration by
definition requires a ‘preferred’ landscape or state, and this desired state may be
explicit or implied. If a desired landscape, supported by legislation and policy is
fictional and based on preconceived notions of naturalness, then there is the real chance
of attempting to re-create an unattainable (and unsustainable) landscape.
It is important to stress that this type of research, although a fascinating ecological
application of technology, is not a goal in itself. The results of historical ecology
should feed into a wider forum of environmental change with the aim of contributing to
a popular understanding of environmental history. Popular ‘myths’ can be powerful
tools to bring about change, but if misguided, significant resources can be wasted.
Clearly, grazing and changed land use have altered vegetation in the eastern Karoo, but
recent carbon isotope evidence casts doubt on the reality of the popular notion of a
“shrub invaded Karoo”.
Grassland/ woodland dynamics: Texas USA
In many grasslands of the USA, shrub and tree increase has been observed for more
than a century (eg. Humphrey and Mehroff 1958; Johnsen 1962; Nelson and Beres
1987; Archer 1989, 1994). One specific example of this process is the expansion of
shrubs and trees from ‘shrub clusters’ and riparian zones in the grasslands of Rio
Grande Plains of Southern Texas (Archer 1995). A significant amount of work has been
carried out on the vegetation dynamics and successional processes associated with this
trend based on historical air photography, and by modelling plant growth rates and
rates of colonisation (Archer et al. 1988; Archer 1989, 1990, 1994). This research has
lead to a detailed knowledge of the processes involved in woodland colonisation, rates
of conversion and the resulting successional trends. Most of the initial research
focussed on using indirect indicators of processes of change, which although
informative, still left the question “was it grassland” or has there been a dynamic
expansion and contraction of woodland vegetation over the past few centuries?
10
In an attempt to overcome this remaining question, stable carbon isotopes have been
utilised in perhaps one of the most extensive applications of the technique to a
particular issue of century scale vegetation change yet attempted (see Boutton et al.
1998, 1999). The present subtropical thorn woodland in many areas of southern Texas
and northern New Mexico are believed to have been C4 dominated grasslands in the
recent past. The typical patterning of the landscape is understood to have comprised
open areas of grassland intersected by woody dominated drainage lines and dotted with
shrub clusters. As succession from grassland to woodland progresses the shrub clusters
coalesce until woodland dominates the landscape. Once the vegetation has reached this
state conversion back to grassland is very difficult. Boutton et al. (1998) combined
stable carbon isotope analysis of above and below ground biomass (foliage, litter and
roots) and soil organic matter with radiocarbon dating and dendrochronology in an
attempt to resolve the issue of whether the woody component of the landscape was
once grassland. Soil was intensively sampled from open grassland, shrub clusters,
drainage lines and transition zones to a depth of 1.5 m. Stable carbon isotope analyses
revealed that 50-90% of soil organic carbon below current C3 vegetation was C4 in
origin. Furthermore, the radiocarbon evidence indicated that the C3 dominant carbon in
the upper 0.15 m of these sites was recent (post 1960’s). The conclusions are in
agreement with previous studies and the anecdotal information that the current, C3
dominant, woody vegetation is an artefact of plant succession over the past 50-100
years. Debate still exists as to whether this now well-demonstrated change is entirely
anthropogenic (via grazing practices) or a response to slight climate change in the
region over the late Holocene (Boutton 1996).
Both the eastern Karoo and Rio Grande Plans case studies demonstrate the potential of
stable carbon isotope techniques to reconstruct, or at least test assumptions regarding,
century scale vegetation change. Both have applied mixed methods and compared
these lines of evidence to develop an improved understanding of the past.
11
Assumptions and caveats
The outcomes of the case studies discussed above indicate the power of stable carbon
isotopic assessment of soil in detecting site specific changes in vegetation over time.
There are several assumptions and limitations that need to be considered when planning
or interpreting δ13C values from in situ soil organic matter studies such as those
outlined above.
Age = depth assumption
There is a general and well-accepted relationship between the depth of soil carbon and
its relative age. However, soil contains a complex mixture carbon compounds that turn
over at different rates. Thus, the age of different carbon compounds may be
significantly different even if collected from the same depth. Although the mean age of
soil carbon may be, for example 500 years, there may also be carbon ranging in age
from recent to thousands of years. This situation creates some difficulties for carbon
isotope studies of vegetation change from soil because temporal resolution becomes
increasingly blurred with increasing depth. In addition, the age of carbon varies with
soil texture. Generally, soil carbon associated with the clay fraction (where most carbon
is held) will be older than that of the coarser fraction. The carbon isotope composition
clay fraction has been shown to respond more slowly to changes in above ground
biomass inputs. This slower carbon turnover time in the clay fraction has been used in
addition to changes with the isotope composition of soil with depth to infer changes in
vegetation (Hoffman et al. 1995; Boutton et al. 1998). This assumption has been used
widely, but is based principally on early work in cultivated and agricultural situations
where carbon turnover rates were being investigated (see review by Tieszen and Archer
1990).
Most data that I have seen published, particularly in semi-arid an non-forested
environments, have shown a relatively predictable increase in the δ13C of soil organic
matter with decreasing particle size. This effect has been attributed partly to the
12
decrease in atmospheric δ13CO2 over the past two centuries (discussed below) (Bird et
al. 1996). Similarly, differential decomposition of plant tissues and carbon compounds
via decay has been implicated to explain the effect (Boutton 1996).
[Fig. 3 near here]
My own data from a number of sites across semi-arid mulgalands of eastern Australia
has shown a characteristic difference between fine and course particles of
approximately 2.5 to 3‰ throughout the soil profile (Fig 3) (Witt 1998). Using the
assumption that the fine fraction is indicative of older carbon and that more recent
vegetation is reflected in the course fraction it would have been safe to assume that a
broad scale regional increase in C3, relative to C4 vegetation has occurred in the recent
past. This assumption was challenged at one site which is dominated by a mature and
dense stand of mulga (Acacia aneura), a C3 tree. Soil from beneath the undisturbed C3
tall mulga shrubland had typically C3 δ13C values for soil beneath semi arid vegetation.
All size fractions to a depth of 25 cm were within one per mil of each other (-23 to 24‰). A large area of this shrubland had recently (less than six years at the time of
sampling) been converted to C4 dominant grassland which provided an opportunity to
determine what impact a known shift from C3 to C4 vegetation would have on the δ13C
of soil organic matter (Fig. 4). In this case it was the fraction less than 10 μm that had
incorporated most of the new carbon from the C4 vegetation (an isotopic shift of
~2.5‰). The coarse fraction (>250 μm) had only shifted slightly (~1‰), and only in
the deepest layer sampled. No change had occurred in the fraction between 63 and 250
μm. These results ran contrary to the expected pattern of change.
[Fig 4 near here]
The complexity of carbon dynamics remains problematic in all carbon isotope studies
of soil organic matter. Although significant experimental work has been carried out, the
diversity of soil - climate - vegetation dynamics make it difficult to use general models
to determine the temporal resolution vegetation change from carbon isotope analyses
13
alone. Essentially the temporal resolution is always problematic with this application of
the δ13C technique. Still, the general rule that surface soils tend to reflect shorter time
frames of biomass turnover while deeper soil incorporates soil organic carbon over
much longer periods remains valid.
Differential fractionation of 13C with decomposition
As indicated above, the process of decomposition of organic matter has some impact on
the δ13C of soil carbon. This effect is attributed not only to the effect of microbial and
physical decay, but also to the diversity of δ13C values for different plant tissues and
compounds. Plant tissues and compounds vary in δ13C relative to whole plant values,
and they are differentially resistant to decomposition, some forming very resistant and
long-lived soil organic compounds while others rapidly oxidised. This process may
alter the δ13C of soil relative to bulk vegetation input, and an enrichment in soil δ13C of
approximately 1‰ can be expected. Although the precise mechanisms for this process
are not yet well understood the effect appears universal.
Where vegetation change from grassland to woodland, and visa versa, is of interest the
large difference in structure and function of grasses and trees must be taken into
account. The root zone of grasses and trees alter the hydrology and bio-physical
processes throughout the soil column in different ways (Boutton et al. 1999). Trees
sequester vast amounts of carbon and usually produce more litter and below ground
tissue than do grasses. Above ground grass tissues are prone to ultraviolet breakdown,
grazing and removal by combustion, thus there may be a bias towards C3 incorporation
into the soil relative to C4. Therefore, relatively small increases in woody vegetation in
formerly open grassland may yield an exaggerated isotopic signal relative to the
situation of grassland replacing woodland.
Atmospheric 13CO2 composition is not constant
The δ13C of atmospheric CO2 has not been stable over geologic time. For studies
confined to the late Holocene atmospheric δ13CO2 can be considered relatively stable
14
until the onset of industrial emissions of CO2. Fossil fuels have low δ13C values
(averaging around -27‰) and their contribution to atmospheric δ13C has meant a drift
from -6.5‰ (pre-industrial) to a current value of –8.0‰ (Leuenberger et al. 1992 and
Keeling et al. 1995). This effect can be exaggerated near urban and other centres of
concentrated CO2 emissions. Such a change over the last 200+ years has been
transferred to plant tissue and then into the soil creating an apparent ‘shift’ in δ13C
values in the soil that has nothing to do with a change in proportion of C3 biomass. The
magnitude of this effect in the soil will be different depending on the carbon turnover
rates of particular soils. At a global scale it has been shown that surface soils at low
latitudes are close to current δ13CO2 trends and that higher latitude soils lag behind in
their sequestration of modern carbon inputs (Bird et al. 1996).
When the effect of the recent anthropogenic change to the carbon isotope composition
of the atmosphere is added to fractionation due to decomposition, unambiguous
interpretation of δ13C of soil organic matter values from a site can be confounded.
Microbial activity and decomposition may enrich soil in the order of 1‰, added to this
a further enrichment of 1.5‰ in the last 200 years attributed to atmospheric change.
These two factors alone can explain an almost universal trend to δ13C enrichment of 1
to 2 ‰ with soil depth even under stable C3 vegetation.
Bioturbation and soil disturbance: an unresolved issue
Soil is not an inert assortment of minerals. The ecological “community” of the soil
includes all forms of life from single-celled bacteria to vertebrates. This issue has often
been ignored or dismissed in stable isotope studies of soil. In cases where it is
acknowledged as a potential complication, the authors can only do their best to avoid
obvious signs of disturbance. Termites, ants, reptiles and mammals will mix soil both
laterally and vertically. Ensuring adequate numbers of samples is perhaps the best
technique to either mask this effect or determine which samples may have
‘unrepresentative’ characteristics. Abiotic factors, particularly erosion and deposition,
can be very difficult to gauge in some soils and this can be especially true of many of
15
the relatively homogeneous soils of semi-arid and arid environments. In this situation
significant deflation from or addition of material to the soil profile may have occurred
that can not be easily recognised in the field.
Spatial Heterogeneity of soil
A single core sample of soil will represent only the dynamics of carbon for that site and
thus, regional interpretation based on data from a single pit is not appropriate. Soil and
the history of vegetation on the soil will vary over meters and for this reason the
technique is well suited to ecotonal and community changes at a location. However, a
regional signal will require a regional approach to sampling. For example, Bond et al.
(1994) in their Karoo study sampled across a broad region although each site was based
on one sample. As with all ecological studies the balance between generality and
precision will alway need to be considered in such situations.
Synthesis of in situ δ13C studies of soil
Stable carbon isotope analyses of soil provide an excellent tool for determining
vegetation dynamics over period of decades to millennia. Although there are
complicating issues (as discussed above) the technique is robust and can provide a
good, if coarse, signal of vegetation change as shown in the case studies of the Rio
Grande Plains and the Karoo. When combined with other methods and compared with
alternative lines of evidence carbon isotopes can be a powerful tool, and indeed can
indicate vegetation change where there may be few other suitable methods capable of
detecting change. The technique is particularly suited to grassland/woodland (forest)
dynamics (i.e. where changing primary productivity of C3 and C4 plants are of interest).
For this reason many questions related to historical vegetation dynamics on the
Australia continent are well suited to stable carbon isotope studies (for example, closed
forest/open forest and woodland dynamics of the tropics, shrub encroachment of the
arid and semi arid zone, and issues of anthropogenically induced tree ‘thickening’
across much of eastern Australia).
16
As yet there are few published Australian examples using soil δ13C to trace historical
(century scale) vegetation change (Witt 1998; Bowman and Cook 2001). Although
some preliminary work has been done (W.H. Burrows, pers. comm.; Witt 1998), much
of the stable isotope analyses of soil carbon in Australia has been conducted for the
purposes of understanding carbon dynamics in relation to agricultural land use (for
example the National Greenhouse Gas Inventory, M.I. Bird pers. comm.) and not to
reconstruct vegetation change.
In the discussion above I have sketched out some significant issues and limitations
associated with the use of stable carbon isotopes. In the Australian context, a lack of
work to date is not necessarily a negative, in fact it should mean that these issues can be
taken into account in the planning stage of any future research.
Ex situ landscape change using δ13C
So far I have focused on the use of stable isotopes where the biomass at the site has
contributed to a record of change in the soil. This is one of the unique features of the
stable isotope technique. Many other palaeoecological techniques depend on finding
deposits that accumulate environmental information over a wide geographic area (e.g.
palynology of lake and marine sediments and caves). The potential of this type of
application of δ13C techniques, although not widely utilised to date will be discussed
below using the issue of biomass burning as an illustration.
The history and role of fire in Australia has prompted debate since settlement,
especially the relationship between fire, vegetation and humans (Flannery 1994, Bird
1995, Bowman 1998). The extent to which Aboriginal people modified the Australian
biota through fire can at times become a heated debate. Regardless of our preferred
understanding of history (Bowman 2001), there is a need to explore as many lines of
investigation as possible to develop a more sophisticated historical record.
17
Fire releases vast quantities carbon in various forms. Elemental carbon (soot) from
terrestrial burning settles out into marine and other aquatic sediments. The δ13C of this
elemental carbon should reflect the relative proportion of C3 and C4 biomass being
burned (Bird and Gröcke 1997; Bird and Cali 1998). This has now been applied in
several studies off northern Australia to determine the photosynthetic source of
terrestrial elemental carbon in marine sediments (Wang et al. 1999, van der Kaars et al.
2000). The technique provides a story of fire frequency as well as the relative C3 and
C4 contribution to biomass that has been burned. Thus it is possible to determine if the
‘catchment’ areas being burned were grassland or woodland/forest. One of the first
such studies was carried out by Bird and Cali (1998) from a one million year long
marine sedimentary sequence “down wind” from sub-Saharan Africa. The record
showed the onset of significant biomass burning some 400 000 years ago, and that an
anomalous peak in the abundance of elemental carbon in the current interglacial period
may be anthropogenic in origin.
Application of this technique has typically been over long time scales. Because of this
very long time frame, detecting vegetation change at a temporal resolution of decades
to centuries is not possible. However, the technique of using sediments in this way
means that unlike soil cores, where carbon dynamics at a site specific scale are being
detected, a large regional signal of fire and vegetation can be traced. The only major
limitation to the application of the use of elemental carbon from sediments will be
finding suitable depositional environments that have experienced rapid sediment
accumulation over the Holocene.
It has also been shown that the δ13C of river sediments can capture information of
catchment vegetation patterns and change (Bird et al. 1994; Martinelli et al. 1999).
There is the potential for lakes, reservoirs, dams and near shore marine sediments to
hold evidence of terrestrial vegetation change in the form of elemental carbon and other
terrestrially derived carbon compounds. Applications of using ex-situ δ13C analysis of
deposited carbon remains untested for detecting recent environmental change.
18
Animals and animal residues as environmental integrators
Analysis of fossils and fossil assemblages as indicators of past environments is one of
the oldest techniques in palaeoecology, but is usually applied to temporal scales beyond
those of interest in this volume. However, assemblages and the morphological
characteristics of fossil flora and fauna are just one way to infer environmental change.
In the field of isotope techniques there is much truth to the expression “you are what
you eat”. A range of isotopes including carbon have been used widely in
ecophysiology, palaeoecology and ecology. The isotopic composition of animal tissues
residues and faeces records a wide range of environmental information including diet,
stress, trophic level, water balance, geology, migration (see Gannes et al. 1997; Hobson
1999). To date applications have been primarily in contemporary ecology (Bunn and
Boon 1993, Hobson 1999) and physiology (Metges et al. 1990), or alternatively,
palaeoecology (Wang et al. 1994; Johnson et al. 1999). What has been lacking is a
bridge between these two temporal scales of study. Historical ecology has received
little attention. Here I will illustrate the potential application of these concepts focusing
on carbon, however, it is very important to point out that the greatest power in isotopic
studies of animal tissue is achieved when a range of isotopic/elemental analyses are
combined.
Animal tissues and faeces reflect the carbon isotopic composition of the diet (Tieszen
and Imbamba 1980; Gearing 1991). Grazing animals such as kangaroos will carry
either C3 or C4 isotopic signature in their tissue depending of the photosynthetic
pathway used by the herbage they consume (Pate et al. 1998; Pate and Noble 2000;
Witt and Ayliffe 2001). Similarly, a koala will be isotopically similar to the C3 foliage
that it eats (Pate and Schoeninger 1993). Any change in diet will be recorded in the
animal at different rates depending on the turnover time of the tissue (e.g. Tieszen et
al. 1983). For example, when a mixed feeding animal moves to a new and isotopically
distinct source of carbon, this change will be first recorded in tissues that have high
rates of turnover, while other tissues such as bone will be buffered from short term
dietary shifts because of the longer residence time of carbon (Libby et al. 1964; Tieszen
19
et al. 1983). This apparently compounding situation can work to the advantage of the
ecologist or palaeo ecologist. By selecting the appropriate tissue, an appropriate
resolution emerges. Bulk bone and entire tooth samples for example, will provide long
term ‘averages’ of the diet or habitat of an animal. Tissue such as hair, wool (Witt et
al., 1998), horn, tusk, and teeth (Balasse et al. 1999, 2001), when sampled carefully,
can provide very high resolution environmental information.
What has been lacking to date is the application of these concepts to questions of
environmental change relevant to the last few centuries. Part of the reason may be that
those people with the expertise and facilities to undertake the research are normally
concerned with either environmental processes that operate on geologic time, or with
contemporary ecophysiology. Recently, carbon isotopic assessment animal residues
and tissue have been used to assess environmental change in the semi-arid zone of
Australia (Witt et al. 1997, 2000) and the Karoo in South Africa (Witt et al. 1998,
1999a).
In the Australian context the carbon isotope composition of sheep faeces from beneath
shearing sheds have been assessed as an indicator of changing availability of grasses
over a period of several decades (Witt et al. 1997, 2000). This work has shown that the
faeces of sheep that accumulate beneath shearing sheds can be used as an indicator of
the diet of sheep in the days leading up to shearing. Furthermore, carbon isotope
composition of faeces is within 1‰ of bulk diet (Jones et al. 1979). Application of this
approach has indicated a significant decline in grass since the 1950’s on a property in
the south west of Queensland (Witt et al. 1997) (Fig. 5). This change corresponded to
vegetation changes that are believed to have occurred across the region (Nobel 1997).
Further research on another property in the same region has extended this near annual
record back to span most of the 20th century (Witt et al. 1999b, 2000; Witt and
Berghammer unpublished data). The broad trend in the isotope record is for C3
vegetation to dominate the diet throughout the record with the exception of the period
that corresponds to the 1950’s when grasses constitute most of the diet. Stable carbon
isotope analyses of faeces was used as the initial tool to determine the broad trend in
20
the proportion of C3 and C4 vegetation in the diet of sheep. However, significantly
more information than that indicated by stable carbon isotope analyses alone has been
gained with the addition of radiocarbon determinations, microhistological analysis of
leaf cuticle material and pollen analyses of these faecal deposits (Fig. 6).
[Figs. 5 and 6 near here]
The concept of using animal tissue as an integrator of environmental change was also
applied to further investigate vegetation change in the eastern Karoo. Using archived
wool staples from a farm in South Africa Witt (1998) was able to develop a high
resolution record C3 and C4 vegetation change at a property scale (see Witt et al. 1999a
for a summary of the results). The wool of fibre-producing sheep (particularly
merinos), grows continuously and is usually shorn once a year. As the fibre grows it is
constantly recording and preserving an isotopic record of a sheep’s diet. Carbon isotope
analyses of wool collected every five years since 1915 indicated significant changes in
the relative availability of C3 and C4 vegetation for most of the 20th century (Fig. 7).
Witt et al. (1998, 1999a) suggested that this may indicate a change in pasture
composition following a collapse in grasses during the 1930’s, a period of very heavy
stock rates. Each staple of wool, when sectioned into 2 mm lengths was able to yield a
record of near-weekly dietary resolution for an individual sheep.
The examples outlined above demonstrate the possibilities of using deposits of animal
residues, such as faeces, or even archived animal tissue to aid in the reconstruction of
century scale vegetation change. Little work has been done in this area, and it is
important that deposits such as those beneath very old shearing sheds are explored
before they vanish. The greatest difficulty with this line of investigation is finding
suitable sample material. Sample material need not be limited to sheep. There are
numerous collections of native and exotic animals kept in museums and universities all
over the world. It could be that the tissue of these animals holds isotopic information,
that, if analysed, could shed light on environmental change over the last few centuries.
21
[Fig 7 near here]
Overview, Conclusions and summary
Stable carbon isotopes have been used widely in the biological and environmental
sciences for over four decades. Insights into vegetation and environmental change
spanning decades to millions of years has been made possible through the analysis
stable carbon isotopes in soils, fossils and other animal tissues. The use of soil carbon
as an indicator of past vegetation change has significant strengths over other
palaeoecological methods. This strength lays primarily in a capacity to detect site
specific changes over time and does not require special depositional sites such as lakes
and caves. The technique should be well suited to the Australian context because in
many areas of the continent there is a significant interest in the past dynamics, and
interactions between, ‘wooded’ (C3) vegetation, and grasslands. Because C4
photosynthesis is a major component of much of the Australian grass flora and the
difference between the carbon isotope values of C3 and C4 plants is incorporated into
soil organic matter over time it should be possible to address issues such as ‘tree
thickening’, shrub encroachment and grassland/forest boundary interactions. These
issues must be dealt with in order to best conserve Australia’s ‘natural’ biodiversity.
In the review I also highlighted the potential of using stable carbon isotope analyses of
elemental carbon from aquatic sediments and the use of animal tissues and residues
such as faecal deposits from beneath shearing sheds. These novel approaches have not
yet been widely applied in century scale vegetation reconstruction.
Although a powerful tool for detecting vegetation changes in the past, stable carbon
isotope analysis suffers from the major limitation of only being able to detect changes
in the contribution of C3 and C4 plant biomass to soil, or animal tissue, overtime. Thus,
the technique has very low floristic resolution. However, as has been emphasised
throughout the review, a combination of historical and palaeoecological technique
should be used where ever possible to develop a richer and more rigorous picture of
past environmental states and processes. I have indicated that although the techniques
22
have been widely applied there has been little use of stable carbon isotopes to detect
historical vegetation change in the Australian context. Given the relative maturity of
stable carbon isotopy as a field of environmental science there is now an opportunity to
apply these techniques to improve our understanding of century-scale vegetation
change in Australia.
References
Acocks JPH (1953) Veld types of South Africa. Memoirs Botanical Survey of South
Africa 28, 1-128.
Acocks JPH (1979) The flora that matched the fauna. Bothalia 12, 673-709.
Archer S (1989) Have southern Texas savannas been converted to woodlands in recent
history? The American Naturalist 134, 545-561.
Archer S (1990) Development and stability of grass/woody mosaics in a subtropical ,
savanna parkland, Texas, U.S.A. Journal of Biogeography 17, 453-462.
Archer S (1994) Woody plant encroachment into southwestern grasslands and
savannas: rates, patterns and proximate causes In ‘Ecological Implications of
Livestock Herbivory in the West’. (Eds M Vavra, WA Laycock and RD Pieper)
Pp.13-68. (Society for Range Management: Denver)
Archer S (1995) Tree-grass dynamics in a Prosopis-thornscrub savanna parkland:
reconstructing the past and predicting the future. Écoscience 2, 83-99.
Archer S (1996) Assessing and interpreting grass-woody plant dynamics. In ‘The
Ecology and Management of Grazing Systems’. (Eds J Hodgson, AW Illius) pp.
101-134. (CAB International, Wallingford)
Archer S, Scifres C, Bassham CR, Maggio R (1988) Autogenic succession in a
subtropical savanna: conversion of grassland to thorn woodland. Ecological
Monographs 58,111-127.
23
Balasse M, Bocherens H, Mariotti A (1999) Intra-bone variability of collagen and
apatite isotopic composition used as evidence of a change in diet. Journal of
Archaeological Science 26, 593-598.
Balasse M, Bocherens H, Mariotti A, Ambrose SH (2001) Detection of dietary changes
by intra-tooth carbon and nitrogen isotopic analysis: an experimental study of
dentine collagen of cattle (Bos taurus). Journal of Archaeological Science 28,
235-245.
Bird MI (1995) Fire, prehistoric humanity, and the environment. Interdisciplinary
Science Reviews 20,141-153.
Bird MI, Cali JA (1998) A million-year record of fire in sub-Saharan Africa. Nature
394, 767-769.
Bird MI, Chivas AR, Head J (1996) A latitudinal gradient in carbon turnover times in
forest soils. Nature 381, 143-146.
Bird MI, Giresse P, Chivas AR (1994) The effect of forest and savanna vegetation on
the carbon-isotope composition of sediments of the Sanaga River (Cameroon).
Limnology and Oceanography 39, 1845-1854.
Bird MI, Gröcke DR (1997) Determination of the abundance and carbon isotope
composition of elemental carbon in sediments. Geochimica Cosmochimica et
Acta 16, 3413-3423.
Bird MI, Pousai P (1997) Variations of δ13C in the surface soil organic pool. Global
Biogeochem. Cycle 11, 313-322.
Bond WJ, Stock WD, Hoffman MT (1994) Has the Karoo spread? A test for
desertification using carbon isotopes from soils. South African Journal of
Science 90, 391-397.
Boutton TW (1991) Stable carbon isotope ratios of natural materials: I. Sample
preparation and mass spectrometric analysis. In ‘Carbon Isotopic Techniques’.
(Eds DC Coleman, B Fry) pp. 155-171. (Academic Press, Inc: San Diego)
24
Boutton TW (1996) Stable carbon isotope ratios of soil organic matter and their use as
indicators of vegetation and climate change. In ‘Mass Spectrometry of Soils’.
(Eds TW Boutton, S Yamasaki). pp 47-82 (Marcel Dekker, Inc: New York)
Boutton TW, Archer SR, Midwood AJ (1999) Stable isotopes in ecosystem science:
structure, function and dynamics of a subtropical savanna. Rapid
Communications in Mass Spectrometry 13,1263-1277.
Boutton TW, ArcherSR, Midwood AJ, Zitzer SF, Bol R (1998) δ13C values of soil
organic carbon and their use in documenting vegetation change in a subtropical
savanna ecosystem. Geoderma 82, 5-41.
Bowman DMJS (1998) The impact of Aboriginal landscape burning on the Australian
biota .New Phytologist 140, 385-410.
Bowman DMJS (2001) Future eating and country keeping: what role has environmental
history in the management of biodiversity. Journal of Biogeography 28, 1-16.
Bowman DMJ, Cook GD (2001) Can carbon isotopes (δ13C) in soil carbon be used to
describe the dynamics of Eucalyptus savanna-rainforest boundaries in the
Australian monsoon tropics? Austral Ecology 27, 94-102.
Bunn SE, Boon PI (1993) What sources of organic carbon drive food webs in a
billabong? a study based on stable carbon isotope analysis. Oecologia 96, 8594.
Coleman DC, Fry B (Eds) (1991) ‘Carbon Isotope Techniques’. (Academic Press Inc:
San Diego)
Dean WRJ, MacDonald AW (1994) Historical changes in the stocking rates of
domestic livestock as a measure of semi-arid and arid rangeland degradation in
the Cape Province, South Africa. Journal of Arid Environments 26, 281-298.
DeNiro MJ (1987) Stable isotopy and archaeology. American Scientist 75, 182-191.
Desjardins T, Carneiro Fliho A, Mariotti A, Chauvel A, Girardin C (1996) Changes of
the forest-savanna boundary in Brazilian Amazonia during the Holocene
revealed by stable isotope ratios of soil organic carbon. Oecologia 108, 749-
25
756.
Ehleringer JR, Rundel PW, Nagy KA (1986) Stable isotopes in physiological ecology
and food web research. Trends in Ecology and Evolution 1, 42-45.
Ehleringer JR, Cerling TE, Helliker BR (1997) C4 photosynthesis, atmospheric CO2,
and climate. Oecologia 112, 285-299.
Fensham RJ, Holman JE (1999) Temporal and spatial patterns in drought-related tree
dieback in Australian savanna. Journal of Applied Ecology 36, 1035-1050.
Flannery TF (1994) ‘The future eaters: an ecological history of the Australasian lands
and people’. (Reed Books: Kew)
Gannes LZ, O’Brien DM, Martinez del Rio C (1997) Stable isotopes in animal ecology:
assumptions, caveats, and a call for more laboratory experiments. Ecology 78,
1271-1276.
Gearing JN (1991) The study of diet and trophic relationships through natural
abundance 13C. In ‘Carbon isotope techniques’ (Eds DC Coleman, B Fry). pp.
201-218. (Academic Press, Inc: San Diego)
Hattersley PW (1983) The distribution of C3 and C4 grasses in Australia in relation to
climate. Oecologia 57, 113-128.
Hendy CH, Rafter TA, MacIntosh NWG (1972) The formation of carbonate nodules in
the soils of the Darling Downs, Queensland, Australia, and the dating of the
Talgai cranium. In ‘Proceedings of the 8th Radiocarbon Dating Conference,
Lower Hutt, New Zealand’. (Eds TA Rafter, T Grant-Taylor). pp D 106-126.
(Royal Society of New Zealand: Wellington)
Hobson KA (1999) Tracing origins and migration of wildlife using stable isotopes: a
review. Oecologia 120, 314-326.
Hoffman MT, Bond WJ, Stock WD (1995) Desertification of the eastern Karoo, South
Africa: conflicting paleoecological, historical and soil isotope evidence.
Environmental Monitoring and Assessment 37, 159-177.
26
Humphrey RP, Mehroff LA (1958) Vegetation changes on a southern Arizona
grassland range. Ecology 39, 720-726.
Johnson BJ, Miller GH, Fogel ML, Magee JW, Gagan MK, Chivas AR (1999) 65,000
Years of vegetation change in Central Australia and the Australian summer
monsoon. Science 284, 1150-1152.
Johnsen TN (1962) One-seed Juniper invasion of northern Arizona grasslands.
Ecological Monographs 32, 187-207.
Jones RJ, Ludlow MM, Troughton JH, Blunt CG (1979) Estimation of the proportion of
C3 and C4 plant species in the diet of animals from the ratio of natural 12C and
13
C isotopes in faeces. Journal of Agricultural Science, Cambridge 92, 91-100.
Keeling CD, Whorf TP, Whalen M, van der Plicht J (1995) Interannual extremes in the
rate of rise of atmospheric carbon dioxide since 1980. Nature 375, 666-670.
Lajtha K, Michener RH (1994) ‘Stable Isotopes in Ecology and Environmental
Science’. (Blackwell Scientific Publications: Oxford)
Leuenberger M, Siegenthaler U, Langway CC (1992) Carbon isotope composition of
atmospheric CO2 during the last ice age from an Antarctic ice core. Nature 357,
488-490.
Levin SA (1992) The problem of pattern and scale in ecology. Ecology 73, 1943-1967.
Libby WF, Berger R, Mead JF, Alexander GV, Ross JF (1964) Replacement rates for
human tissue from atmospheric radiocarbon. Science 146, 1170-1172.
Ludwig JA, Tongway DJ (1995) Desertification in Australia: an eye to grass roots and
landscapes. Environmental Monitoring and Assessment 37, 231-237.
Martinelli LA, Ballester MV, Krusche AV, Victoria RL, de Camargo PB, Bernardes,
M, Ometto JPHB (1999) Landcover changes and δ13C composition of riverine
particulate organic matter in the Piracicaba River Basin (southeast region of
Brazil). Limnology and Oceanography 44, 1826-1833.
McPherson GR, Boutton TW, Midwood AJ (1993) Stable carbon isotope analysis of
27
soil organic matter illustrates vegetation change at the grassland/woodland
boundary in southeastern Arizona, USA. Oecologia 93, 95-101.
Metges C, Kempe K, Schmidt HL (1990) Dependence of the carbon-isotope contents of
breath carbon dioxide, milk, serum, and rumen fermentation products on the
δ13C value of food in dairy cows. British Journal of Nutrition 63, 187-196.
Moore PD (1982) Evolution of photosynthetic pathways in flowering plants. Nature
295, 647-648.
Nelson JT, Beres PL (1987) Was it grassland? A look at vegetation in the Brewster
County, Texas through the eyes of a photographer in 1899. Texas Journal of
Agriculture and Natural Resources 1, 34-37.
Noble JC (1997) ‘The Delicate and Noxious Scrub’. (CSIRO: Canberra)
Nordt LC, Boutton TW, Hallmark CT, Waters MR (1994) Late Quaternary vegetation
and climate changes in central Texas based on isotopic composition of organic
carbon. Quaternary Research 41, 109-120.
O’Leary MH (1988) Carbon isotopes in photosynthesis. BioScience 38,328-336.
Pate FD (1997) Bone chemistry and paleodiet: reconstructing prehistoric subsistencesettlement systems in Australia. Journal of Anthropological Archaeology
16,103-120.
Pate FD, Anson TJ, Schoeninger MJ, Noble AH (1998) Bone collagen stable carbon
and nitrogen isotope variability in modern South Australian mammals: a
baseline for palaeoecological inferences. Quaternary Australasia 16, 43-51.
Pate FD, Noble AH (2000) Geographic distribution of C3 and C4 grasses recorded in
bone collagen stable carbon isotope values of south Australian herbivores.
Australian Journal of Botany 48, 203-207.
Pate FD, Schoeninger MJ (1993) Stable carbon isotope ratios in bone collagen as
indicators of marine and terrestrial dietary composition in southeastern
Australia: a preliminary report. In ‘Archaeometry: Current Australian Research’
(eds BL Frankhauser, JR Bird) pp. 38-44. Department of Prehistory, Research
28
School of Pacific Studies, Australian National University, Canberra. Occasional
Papers in Prehistory No. 22.
Peterson BJ, Fry B (1987) Stable isotopes in ecosystem studies. Annual Review of
Ecology and Systematics 18, 293-320.
Quade J, Chivas AR, McCulloch MT (1995) Strontium and carbon isotope tracers and
the origins of soil carbonate in South Australia and Victoria. Palaeogeography,
Palaeoclimatology, Palaeoecology 113, 103-117.
Rounick JS, Winterbourn MJ (1986) Stable carbon isotopes and carbon flow in
ecosystems. BioScience 36, 171-177.
Roux PW, Vorster M (1983) Vegetation change in the Karoo. Proceedings of the
Grassland Society of Southern Africa 18, 25-29.
Rundel PW, Ehleringer JR, Nagy KA (Eds) (1989) ‘Stable isotopes in ecological
research’ Ecological Studies No. 68 (Springer-Verlag: New York)
Schwartz D, Mariotti A, Lanfranchi R, Guillet B (1986) 13C/12C ratios of soil organic
matter as indicators of vegetation changes in the Congo. Geoderma 39, 97-103.
Schwarz EHL (1919) The progressive desiccation of Africa: the cause and the remedy.
South African Journal of Science 15, 139-190.
Shaw J (1875) On the changes going on in the vegetation of South Africa through the
introduction of merino sheep. Journal of the Linnean Society of Botany 14, 202209.
Stout JD, Rafter TA, Troughton JH (1975) The possible significance of isotope ratios in
palaeoecology. In ‘Quaternary studies’. (Eds RP Suggate, MM Creswell). pp.
279-286. (Royal Society of New Zealand: Wellington)
Terri JA, Stowe LG (1976) Climatic patterns and the distribution of C4 grasses in North
America. Oecologia 23,1-12.
29
Tieszen LL (1991) Natural variations in the carbon isotope values of plants:
implications for archaeology, ecology, and paleoecology. Journal of
Archaeological Science 18, 227-248.
Tieszen LL, Archer S (1990) Isotopic assessment of vegetation changes in grassland
and woodland systems. In ‘Plant Biology of the Basin and Range’. Ecological
Studies Series, vol. 80 (Eds CB Osmond, LF Pitelka, GM Hidy). pp293-321.
(Springer-Verlag: Heidelberg)
Tieszen LL, Boutton TW, Tesdahl KG, Slade NH (1983) Fractionation and turnover of
carbon isotopes in animal tissues: implications for δ13C analysis of diet.
Oecologia 57, 32-37.
Tieszen LL, Imbamba SK (1980) Photosynthetic systems, carbon isotope
discrimination and herbivore selectivity in Kenya. African Journal of Ecology
18, 237-242.
van der Kaars S, Wang X, Kershaw P, Guichard F, Setiabudi DA (2000) A late
Quaternary palaeoecological record from the Banda Sea, Indonesia: patterns of
vegetation, climate and biomass burning in Indonesia and northern Australia.
Palaeogeography, Palaeoclimatology, Palaeoecology 155, 135-153.
van der Merwe NJ (1982) Carbon isotopes, photosynthesis, and archaeology. American
Scientist 70, 596-606.
Victoria RL, Fernandes F, Piccolo MdeC, De Camargo PB, Trumbore S (1995) Past
vegetation changes in the Brazilian Pantanal arboreal-grassy savanna ecotone
by using carbon isotopes in the soil organic matter. Global Change Biology 1,
165-171.
Wang X., van der Kaars S, Kershaw P, Bird MI, Jansen F (1999) A reord of fire,
vegetation and climate through the last three glacial cycles from Lombock
Ridge core G6-4, eastern Indian Ocean, Indonesia. Palaeogeography,
Palaeoclimatology, Palaeoecology 147, 241-256.
Wang Y, Cerling TE, MacFadden BJ (1994) Fossil horses and carbon isotopes: new
30
evidence for Cenozoic dietery, habitat, and ecosystem changes in North
America. Palaeogeography, Palaeoclimatology, Palaeoecology 107, 269-279
Witt GB (1998) ‘How the west was once: reconstructing historical vegetation change
and monitoring the present using carbon isotope techniques’. PhD thesis (The
University of Queensland: Brisbane)
Witt GB, Ayliffe LK (2001) Carbon isotope variability in the bone collagen of red
kangaroos (Macropus rufus) is age dependent: implications for palaeoecological
studies. Journal of Archaeological Science 28, 247-252.
Witt GB, Berghammer LJ, Beeton RJS, Moll EJ (2000) Retrospective monitoring of
rangeland vegetation change: ecohistory from deposits of sheep dung associated
with shearing sheds. Austral Ecology 25, 260-267.
Witt GB, Berghammer LJ, Bland S, Beeton RJS, Moll EJ (1999b) Ecological history
from faecal deposits beneath shearing sheds: a novel approach for
reconstructing vegetation change in Australian rangelands. In ‘People and
Rangelands: Building the Future’. (Eds D Eldridge, D Freudenberger) pp. 243244. Proceedings VI International Rangelands Congress, Vol 1.
Witt GB, Moll EJ, Beeton RJS (1997) Sheep faeces under shearing sheds; a
documentary of vegetation change using stable isotope analysis. Rangeland
Journal 19, 109-115.
Witt GB, Moll EJ, Beeton RJS, Hoffman MT (1999a) Vegetation in the eastern Karoo
(South Africa) from 1916 to 1992 using carbon isotope analysis of historical
wool staples. In ‘People and Rangelands Building the Future’. (Eds D Eldridge,
D Freudenberger) pp. 242-243. Proceedings of the VI International Rangelands
Congress, Vol 1.
Witt GB, Moll EJ, Beeton RJS Murray PJ (1998) Isotopes, wool and rangeland
monitoring: let the sheep do the sampling. Environmental Management 22, 145152.
Wolf DC, Legg JO, Boutton TW (1994) Isotopic methods for the study of soil organic
31
matter dynamics. In ‘Methods of soil analysis, Part 2. Microbiological and
biochemical properties’. (Eds RW Weaver, S Angle, P Bottomley) pp. 865-906.
Book Series, No. 5 (Soil Science Society of America: Madison)
32
Figure captions
Fig. 1. An approximate indication of the percentage of species using the C4
photosynthetic pathway in the grass flora across Australia (adapted from Hattersley
1983). Note that for much of mainland Australia three-quarters or more of all grass
species in any one area are C4, and the by the tropic of Capricorn (shown as the dashed
line) this value exceeds 90%.
Fig. 2. A model of the major stages and processes that determine the stable carbon
isotope composition of soil organic matter (adapted from Boutton 1996).
33
Atmospheric CO2
emissions
-8‰
Anthropogenic CO2
(fossil fuel combustion and
land use)
-27‰
Δ from -6.5‰ since pre-industrial
range -12‰ (urban) to -7.5‰
(rural and unpolluted atmos.)
return of CO2 to Atmosphere
C3 vegetation
-27‰
Decomposition and
respiration
(range -32 to -22‰)
Litter and nondecomposed
plant material
(δ13C dependant
on source)
Photosynthesis and
subsequent isotopic
discrimination
SOIL ORGANIC
MATTER
decay
C4 vegetation
(δ13C dependant on source
and function of preceding
stages)
-13‰
(range -20 to -9‰)
Environmental and biological factors affecting photosynthesis,
available moisture, decomposition etc.
34
Fig. 3. The typical relationship of soil particle size on the stable carbon isotope ratio of
-21
13
δ C in ‰
-22
-23
-24
-25
<1
0
10
-63
63
-12
5
12
5-2
50
25
0-1
00
0
bu
lk
so
il
-26
size fraction of soil (μm)
organic matter from surface soil (0-2 cm) in south west Queensland (Currawinya
National Park 28º38’ S’, 144º 50’ E) (Witt unpublished data).
35
stable Acacia aneura tall
shrubland
recent C4 grassland
cleared of C3 vegetation six years
earlier
C3 dominant vegetation
A.
B.
δ13C in ‰
-26
0
-25
-24
-23
δ13C in ‰
-22
-21
δ C >250μm
δ13C >125μm
δ13C >63μm
δ13C >10μm
δ13C <10μm
-26
0
5
10
-24
-23
-22
-21
δ C >250μm
δ13C >125μm
δ13C >63μm
δ13C >10μm
δ13C <10μm
13
depth in cm
depth in cm
13
-25
5
10
15
15
20
20
Fig. 4. The isotopic response in different soil particle size classes to the replacement of
a stable C3 vegetation type (A) with a C4 dominated grassland (B) (Witt 1998). The
grassland on Croxdale Research Station (26º30’S, 146º10’E) had been created by
clearing the Acacia aneura tall shrubland in 1991 five years prior to my sampling with
two prescribed fires in the intervening period. Soil was pooled from 6 pits in each
vegetation type so as to avoid small scale variation.
36
100
200
300
400
500 -25
-24
-23
-22
-21
-20
-19
-18
-17
1990
1985
years
1980
1975
1963/64
1965
winter
summer
1960
0
100
200
300
400
rainfall in mm
500 -25
(ANU-10491)
profile 1
profile 2
profile 3
-24
-23
-22
-21
-20
-19
-18
0
10
20
30
40
50
60
70
80
90
100
-17
δ13C(PDB) in ‰
Fig. 5. The stable carbon isotope composition of three profiles of sheep faeces buried
beneath the Currawinya shearing shed (28º51’S, 144º30’E) plotted against the
corresponding summer and winter rainfall. Because the depth of each profile varied
slightly they have been plotted as relative depths. With the exception of a short period
between the late 1960s and early 1970s, the deposit represents the diet of sheep (as
indicated by their faeces) from the mid-1950s to 1992 when shearing ceased. There was
a dramatic change in the diet of sheep over this period from an estimated 40% C4
derived plant material during the 1950s to well below 10% by the early 1990s (see Witt
et al. 1997 and Witt 1998).
37
relative depth
expressed as a percentage
0
early
1960's
pre-1950's
approx.
chronology
hiatus
~1965-'75
relative contribution of
plant group to total diet
1950's
1975-1995
browse
herbs, forbs &
chenopods
graminoids
13
δ C in ‰
-16
-18
-20
-22
-24
70
60
50
40
30
20
10
0
depth of deposit in cm
Fig. 6. Environmental change on Ambathala, south west Queensland (25º58’S,
145º20’E), over most of the 20th century as indicated by sheep faeces that have
accumulated beneath the property’s shearing shed. Both the δ13C and the proportion of
leaf cuticle remains of different plant groups in the faeces record a significant decline
in grass (shaded black) since the 1950s.
This supports the observed trend for
Currawinya, further south, over the same period (see Fig. 5). However, it is also clear
that prior to the 1950s grass was very rare in the diet (this record is based on
unpublished data of Witt and Berghammer, see also Witt et al. 2000 for details of
methods)
38
δ13C in ‰
-12
-12
-12
-14
-14
-14
-16
-16
-16
-18
-18
-18
-20
-20
-20
-22
-22
-22
-24
-24
δ13C in ‰
-26
Oct./Nov.
-26
1919 - 1920
Oct./Nov.
-24
-26
-12
-12
-12
-14
-14
-14
-16
-16
-16
-18
-18
-18
-20
-20
-20
-22
-22
-22
-24
-24
-26
δ13C in ‰
1915 - 1916
1929 - 1930
Sept./Oct.
-26
1934 - 1935
Sept./Oct.
-24
-26
-12
-12
-12
-14
-14
-14
-16
-16
-16
-18
-18
-18
-20
-20
-20
-22
-22
-22
-24
-24
-26
1950 - 1951
Aug.
-26
1967 - 1968
Aug.
-24
-26
1924 - 1925
Oct./Nov.
1936 - 1937
Sept.
1979 - 1980
Aug.
-12
δ13C in ‰
-14
-16
-18
-20
-22
-24
-26
1991 - 1992
Aug.
Fig. 7. The δ13C of ten wool staples spanning much of the 20th century from Fair View,
South Africa (32º05’S, 26º10E). Each staple of wool represents the contribution of C3
and C4 plants to diet over the year between shearing. The period of growth and the
approximate time of shearing are shown at the bottom left hand corner of each graph
(note that the time of shearing has changed from October and November early in the
series to August since 1950). Throughout the 1920s and 1930s C4 vegetation (grass) is
rare or absent from the diet for most of the year. This ‘collapse’ of grasses coincides
39
with very high stock numbers from the mid-1920s until the beginning of the second
world war.
40