The stable light isotopes of principal interest as environmental proxies are the isotopes of hydrogen, carbon, nitrogen, and oxygen. In all cases, the chemistry of the different isotopes of an element remains the same as chemical properties are controlled by electron configuration. Among these light isotopes, the mass difference owing to one or two extra neutrons is sufficient to cause a small, but predictable, difference in the rates of chemical reactions or physical processes. If the starting materials and the products of a reaction are partitioned in some way, observable isotopic fractionation occurs.
184.108.40.206 Hydrogen (D/H) and oxygen (18O/16O) isotope fractionation
Fractionation of hydrogen (D/H) and oxygen (18O/16O) isotopes1 in nature follows similar patterns because their isotope effects are dominated by the processes undergone by water evaporation, condensation, and freezing. The magnitude of fractionation is controlled largely by temperature in both cases, but isotope effects are much higher for hydrogen because of the greater relative mass difference between deuterium (D or 2H) and hydrogen (1H) compared to oxygen (18O and 16O). Water vapor evaporates largely from low- to midlatitude ocean surfaces, and sea surface temperature (or latitude as a proxy) of the oceanic source influences isotopic ratios of moisture-laden weather systems (Dansgaard 1964). Further influences on the isotopic composition of rainfall occur during subsequent
1Isotope ratios are expressed by convention in the d notation, in parts per thousand (%o) relative to a standard: §X (%o) = (Rsampie-Rref)/Rref x 1000, where R is the isotopic ratio. V-SMOW (Standard Mean Ocean Water) is used as the reference for D/H and 18O/16O in water, V-PDB (Peedee Belemnite) for 18O/16O and 13C/12C in carbonates and organic materials, and atmospheric N2 (AIR) for 15N/14N.
transport of these weather systems across continents, related to factors such as the distance traveled, mountains crossed, and height and temperature of rainclouds (Dansgaard 1964; Rozanski et al. 1993). A further effect in arid landscapes is due to evaporation. Soil- and groundwater isotope values reflect this history, as do carbonates precipitated from these waters, with the additional influence of local temperature as fractionation during carbonate precipitation is temperature dependent. This is the basis of the familiar "temperature equation'' applied to many carbonates (e.g., forams, corals, and speleothems) (McCrea 1950). In addition, isotope effects also occur in plants. A large isotopic enrichment occurs during daytime evapotranspiration in plants under hot, dry conditions (Yakir 1992).
220.127.116.11 Carbon (13C/12C) isotope fractionation
Carbon (13C/12C) isotope fractionation provides fundamental information about pathways in the terrestrial carbon cycle (see also Sponheimer and Lee-Thorp Vol. 1, Chapter 17). A large depletion in 13C occurs during assimilation of atmospheric CO2 in photosynthesis, but to different degrees depending on the photosynthetic pathway (Smith and Epstein 1971). C3 plants (trees, shrubs and forbs, and temperate or shade-adapted grasses) are more depleted in 13C and have low 813C values, while C4 plants (mainly tropical grasses) are less depleted (O Figure 9.1). A smaller group of succulent plants follow the Crassulacean acid metabolism (CAM) pathway, which essentially alternates use of these two pathways by night and day, and depending on conditions, so that their isotopic compositions vary considerably. These plant values are archived in slightly different ways in a range of sample materials (O Figure 9.1).
18.104.22.168 Nitrogen (15N/14N) isotope ratios
Nitrogen (15N/14N) isotope ratios reveal pathways in the nitrogen cycle, but since the pathways are complex, the values observed between different ecosystems are highly variable. Nitrogen enters the terrestrial foodweb via N2-fixing bacteria in soils or plants to form nitrates or ammonium ions that are utilized by plants. The net effect of this biological nitrogen fixing and subsequent nitrogen loss is a slight enrichment in 15N, but the balance and mode of fixing and loss is strongly affected by environmental conditions (Heaton 1987). In a global survey, Handley et al. (1999) found a negative correlation between moisture availability and both leaf and soil 815N. Stepwise trophic enrichment of about 3-5%o occurs in animals because there is an ~3%o fractionation between diet and tissues. The trophic
A model depicting typical carbon isotope pathways in a modern savanna ecosystem. The values given in the boxes are typical mean values for each material, expressed relative to the international standard V PDB. We have used values for tooth enamel as this is the tissue most commonly analyzed in fossils;other tissues would show a similar separation, although different absolute values. In fossils, tooth enamel values tend to be slightly more positive, i.e., fossil enamel tends to be slightly enriched in 13C in comparison with modern specimens. Typical values for 813C in pedogenic carbonates or speleothems from pure C3 or C4 habitats are similar, but again slightly enriched
effect, however, can be surpassed by enrichment associated with environmental aridity (Heaton et al. 1986; Sealy et al. 1987; Johnson et al. 1997), effects that are certainly related to raised leaf and soil values in arid areas (Heaton 1987). The effect is very marked in animals living in areas with <400-mm rainfall/annum, so s15n can be an indicator of aridity.
The sample materials most often used as environmental indicators in sites associated with hominin activities consist largely of bones and soils in close chronostratigraphic association with those sites. Fossil ostrich eggshell and cave speleothems represent promising but more rarely explored sources of paleoenvironmental and climate information.
Bones are abundant in most sites where conditions are conducive to preservation. They conserve information about many of the processes and conditions to which the animal was subjected when it was alive, which can be accessed via their stable light isotope compositions. Bones and dentine consist of about 25% (by weight)
of a fibrous protein, collagen, and about 75% embedded bioapatite mineral. Collagen can be relatively easily purified and analyzed for 813C and 815N. However, it tends to decompose over time so that by about 10-20 Ka collagen has all but disappeared from bones and teeth in most African sites (preservation is better in cool Eurasian sites). The mineral is composed of calcium phosphate crystals with many other ions substituted into the structure (LeGeros 1991) that tend to increase their reactivity and solubility. One exception is fluoride, which enhances stability. Substitutions also affect how well bones and teeth are preserved as fossils. Enamel is far more stable than bone apatite (Lee-Thorp and van der Merwe 1987) and consequently most isotopic studies of fossil fauna have relied on enamel as sample material. The component ions of interest are phosphate (PO3~) and carbonate (CO3_), with the latter occurring in small amounts (3-6%) as a substitution. Therefore, the isotopes available for study in this system are carbon and oxygen isotopes, and the latter may be extracted from either PO^ or
Bioapatite phosphate 818O has been developed as a paleotemperature tool, based on the rationale that body water 818O (d18Obw) in mammals is related to environmental or drinking water (818Ow), which can in turn be correlated with latitude and temperature effects on rainfall (Longinelli 1984; Luz and Kolodny 1985, 1989). In low- to midlatitudes, temperature effects are far less important than storm-track and amount effects on d18O of environmental water plus isotope effects on plant waters. We now know that animal behavior, related to drinking patterns and/or thermophysiology, can modulate these environmental signals considerably (Bocherens et al. 1996; Kohn et al. 1996; Sponheimer and Lee-Thorp 1999b, 2001) and that these patterns are conserved in fossil assemblages. Ayliffe and Chivas (1990) found a correlation between 818OPO4 and relative humidity in non-obligate drinking animals—in their case, kangaroos, so that these behaviors might still provide useful environmental data. In order to circumvent the problems of variable responses to climate factors, many authors have relied upon "well-behaved," obligate drinking species such as equids for extracting paleoclimate proxies (Sanchez-Chillon et al. 1994; Bryant et al. 1996).
Stable carbon isotope (d13C) analysis of herbivore tooth enamel (or collagen) indicates the relative amounts of C3 plants and C4 grasses consumed. The calculations are based on our understanding of typical plant values today and how they might be affected by climate variables, and by the difference between diet and enamel 813C values (or Adiet_en). There is a little "play" here, since C3 plants may be slightly enriched in 13C under hot, arid conditions, and vice versa. C3 plants in dense forests, where a "canopy effect'' with recycling of CO2 and low light prevails, are typically depleted in 13C (van der Merwe and Medina 1989, 1991). Alterations in the species composition of C4 grasses, in response to conditions, may shift their average d13C value slightly (by perhaps 1%o).
In addition, Adiet_en is not precisely established; some authors observe + 12%o (Lee-Thorp and van der Merwe 1987), while others have observed up to +14 %o (Cerling and Harris 1999).
Still, these small differences make no practical difference to distinguishing between C3 and C4 consumers since these S13C distinctions are large (© Figure 9.1). The presence of C4 grazers in an assemblage allows us to deduce presence of tropical C4 grasses in a landscape. Although this information alone may be useful, for instance, if a shift between cold- and warm-season rain patterns is suspected (Lee-Thorp and Beaumont 1995), in practice we need to take this process further. Usually we require information about how closed or open vegetation cover was. A C3/C4 index has been developed to reflect the relative openness of an environment from S13C data in a faunal assemblage (Sponheimer and Lee-Thorp 2003) (see below).
The eggshells of ostriches and earlier ratite taxa are durable and often abundant in paleontological and archeological sites. The shells consist of a protein matrix and inorganic calcium carbonate. Eggshells can yield S13C, S15N, and S18O proxy information reflecting conditions during the short egg-laying season. Ostriches are mixed feeders that prefer tender plants, which can include succulents as well as annual grasses and forbs in their diets; hence S13C data can indicate presence of both C4 grasses and CAM plants (Stern et al. 1994; Segalen et al. 2006). The S15N values are thought to reflect mean annual rainfall (MAP) (Johnson et al. 1997) in the same way as mammals in arid regions with under 400-mm rain/year (Sealy et al. 1987). S18O data also reflect aridity but from a slightly different perspective; since ostriches are drought-tolerant animals that do not need to drink free water, S18O primarily reflects leafwater enrichment due to evapotranspiration under hot, dry conditions (Lee-Thorp and Talma 2000). Because of the high variability in S18O and S15N data, large numbers of analyses are required (Lee-Thorp and Talma 2000). To date, only one study has produced a sequence, extending beyond the Holocene based on all three proxies (Johnson et al. 1997).
A history of overlying vegetation maybe preserved for long periods in remnant or-ganics and pedogenic carbonates in paleosols (Cerling et al. 1991). The principles are by now familiar: carbon isotopes in both the organic and inorganic components of soils systematically reflect differences between vegetation following the C3 and C4 pathways, allowing estimations of the relative mix of woody and grassy plants on the landscape. 813C of paleosol organic matter, where preserved, is a reasonably direct reflection of the mean isotopic composition of the vegetation, with a small enrichment in 13C owing to decomposition and associated processes (Cerling et al. 1991). Pedogenic carbonate nodules are formed about 0.5-1 m below the active soil horizon from soil-respired CO2 (Cerling and Quade 1993). There is a net enrichment in 13C of 14-17%o due to combined effects of diffusion and isotopic fractionation during carbonate precipitation, causing increases of 4.4-5%o, and 9.5-12.5%o, respectively (Cerling et al. 1988, 1991; Cerling and Quade 1993). Pedogenic carbonates have a relatively restricted distribution; for one, nodule formation is associated with semiarid to arid conditions. An important constraint is the identification of "true" pedogenic nodules formed well below the active, identifiable soil horizon, where diffusion enrichment is complete and there is no mixing with atmospheric CO2 (Cerling et al. 1988; Cerling and Quade 1993).
Cave flowstones and stalagmites are composed of calcium carbonate formed from CO2—rich seepage water dripping into cave systems, degassing, and precipitating as carbonate. Their 813C and 818O values reflect the proportions of C3 and C4 plants in overlying vegetation, and the isotopic composition of the water, in a similar manner to pedogenic carbonates. Speleothems have some decided advantages. For one, a closed cave system is a protected environment, where averaged annual temperatures are maintained year-round, along with high relative humidity. In a closed cave system equilibrium conditions are more likely, meaning that greater confidence can be placed in the isotopic data as sound indicators of environmental conditions. Very importantly, they are incremental structures and can be precisely dated using thorium-uranium disequilibrium or lead isotopes. Ages are more difficult to obtain for older Pliocene- or Miocene-age speleothems, but paleomagnetic and lead-uranium methods are being developed (Hopley 2004). So, where the right material can be located and the ages determined, speleothems can be sampled at small intervals to yield continuous, high-resolution records of a quality unlike any of the other proxies discussed here. Continuous sequences like this are invaluable because they can be compared with other dated continuous records, and they allow us to check trends and the scales of variability.
The following section describes some important applications of isotopic tools to paleoenvironmental and -climate problems.
Was this article helpful?