- Email: [email protected]
Investigation of the ‘canopy effect’ in the isotope ecology of temperate woodlands
Journal of Archaeological Science 40 (2013) 3926e3935
Contents lists available at SciVerse ScienceDirect
Journal of Archaeological Science journal homepage: http://www.elsevier.com/locate/jas
Investigation of the ‘canopy effect’ in the isotope ecology of temperate woodlands M. Bonafini a, b, M. Pellegrini a, *, P. Ditchfield a, A.M. Pollard a a b
Research Laboratory for Archaeology and the History of Art, School of Archaeology, University of Oxford, South Parks Road, Oxford OX1 3QY, UK Department of Environmental Sciences, Università Ca’ Foscari, Calle Larga Santa Marta 2137, 30123 Venezia, Italy
a r t i c l e i n f o
a b s t r a c t
Article history: Received 14 August 2012 Received in revised form 8 March 2013 Accepted 31 March 2013
Anomalously 13C-depleted carbon stable isotope values from closed forest ecosystems have been termed the ‘canopy effect’. Originally this was ascribed to recycling of depleted carbon from forest floor decomposition of organic material, although others have suggested that it is equally likely to be due to variations in leaf-level processes in response to increased shade. This depletion in the heavier carbon isotope is passed on to woodland herbivores feeding within the forest environments. A similar isotopic depletion has also been reported in the archaeological literature from more open temperate woodland settings, but no measurements have been made on the plants at the base of the food chain in order to quantify the effect. In this study we attempt this by examining the carbon and nitrogen stable isotopic values of different species of grasses from a range of open to closed habitat settings within Wytham Wood, Oxfordshire, UK. We find a strong relationship between carbon isotopic depletion of plant tissue and lowered light intensity with an up to 5& shift between grass grown in open and closed locations. In order to follow this up the food chain, we also report data on wool from sheep grazing in open pastures near the Wood, and on fallow deer living within the woodland, but which turn out not to show a strong canopy effect, probably related to their feeding strategies. We conclude that there is indeed a strong ‘canopy effect’ in temperate woodland, probably related to differential light levels, but that not all apparently woodland-dwelling mammals show such an effect. We also show considerable isotopic variation at the base of the food chain, which should counsel caution when attempting to interpret dietary isotopes using mixing models. Ó 2013 Published by Elsevier Ltd.
Keywords: d13C d15N ‘Canopy effect’ Temperate woodlands Bone collagen Wool
1. Introduction The effect of recycling carbon dioxide on the carbon isotopic composition of leaves in a forest environment was first systematically investigated by Vogel (1978). In a forest in Upper Bavaria, Germany, he noted that leaves of Asperula and Oxalis growing near the ground had a values of d13C ¼ 31.4& and 31.5& respectively, whereas leaves of Larix and Fagus gave values of d13C increasing from at 31.2& for those growing at 2 m to 27.9& at 19 m. He attributed this phenomenon to the presence near the ground of depleted carbon dioxide emanating from the decay of plant material in the leaf litter. Measurements of CO2 collected from beneath an upturned barrel in a forest near Heidelberg gave d13C values of between 15.0& and 22.4& from June to August. This is substantially below the average atmospheric CO2 value of c. 7&, and suggests that the biogenic CO2 might have a value approaching that * Corresponding author. Tel.: þ44 1865 285204; fax: þ44 1865 285220. E-mail addresses: [email protected], [email protected] (M. Pellegrini). 0305-4403/$ e see front matter Ó 2013 Published by Elsevier Ltd. http://dx.doi.org/10.1016/j.jas.2013.03.028
of the humus itself at d13C ¼ 27.0&, since the collected sample was a mixture of air and exhaled CO2. He estimated that 15% of the CO2 incorporated into leaves at a height of 2 m above the ground is from the soil in this particular forest environment. An alternative explanation for this isotopic depletion has been proposed based on the work of Farquhar et al. (1982), who produced an equation to relate the carbon isotopic value of leaf tissue to the ratio of the CO2 concentration in intercellular leaf spaces (Ci) and the atmospheric CO2 concentration. Since Ci is inversely related to light intensity, this model predicts that more shaded locations should produce vegetation with more negative d13C values (van der Merwe and Medina, 1991). In their seminal paper on the canopy effect in the Amazonian forest, van der Merwe and Medina (1991, 251) concluded that ‘many observers have simply assumed that both alternatives may contribute to the canopy effect’. There is therefore a general assumption that carbon isotope depletion occurs in plants from forested environments, the degree of which is in some way proportional to forest density. Since plants are at the base of the food chain, any such depletion is expected to be transferred up the food chain.
M. Bonafini et al. / Journal of Archaeological Science 40 (2013) 3926e3935
The ‘canopy effect’ has now been widely invoked in archaeology to explain the observation of depleted d13C values in some animal bone collagen, attributed to depletion of vegetation d13C values in wooded or partially-wooded environments (e.g., Bocherens et al. 1995; Fizet et al. 1995; Rodière et al. 1996; Cerling and Harris, 1999; Bocherens et al. 1999; Iacumin et al. 2000; Drucker et al. 2000; Stewart and Lister, 2001; Krigbaum, 2003). The most extensive such consideration is that of Drucker et al. (2008), who studied isotopic values (from collagen or hair) in modern large herbivores from the boreal forest-steppe of Siberia, temperate forests of Dourdan (France), and the boreal forests of western Canada. This paper confirmed that ‘the canopy effect observed in plants is passed on to their consumers’, and concluded that the 3& shift in faunal collagen d13C values observed in the LateGlacialeEarly Holocene transition from a number of French sites was related to the differences in habitat. Although such depletion effects have been directly demonstrated in modern plant material from tropical Amazonia (van Der Merwe and Medina, 1989) and in other low latitude regions, they have been less well-studied in higher latitude temperate woodlands. Although convincing, this and other papers have inferred an isotopic shift at the base of the food chain from measurements of faunal collagen values in open compared to forested environments, rather than by direct observation of the flora. The aim of this research is to directly assess the extent of isotopic variation in grass growing at different locations in temperate woodlands (Wytham Woods, near Oxford, England) as a function of time through the growing season, degree of shade, and climatic conditions in general. At the same time we have analyzed biological tissues (bone collagen from fallow deer and wool from sheep) from animals reared in the same area. We deliberately chose to analyze whole grass samples for carbon and nitrogen isotopes (rather than selecting specific components such as cellulose, lignin, etc.), as this is the dietary basis of the grazing animals. The results of this study provide a quantitative measure of the ‘canopy effect’ in the flora at mid to high latitudes, and the extent of isotopic shifts which might be expected in the food chain at these latitudes.
3927
(Fig. 1). The wood covers 415 ha at an altitude from 90 to 165 m A.M.S.L. and is part of the Wytham Estate, which is c. 980 ha in extent. The remaining part of the estate is farmland. The woods are now exceptional in lowland England since they encapsulate the range of both woodland and non-woodland habitats that were common prior to agricultural intensification. Approximately one-third of the area is ancient semi-natural woodland, which was historically managed as coppice with standards. Over the course of the last 100 years coppice management has been largely abandoned. Another third of the area is recent seminatural woodland that has regenerated on arable, pasture or wood pasture sites in the last two hundred years. The remaining woodland area consists of a variety of plantations, some on ancient woodland sites, some on sites which were formerly open. The woods lies on neutral clay soils, ranging from thin, freelydraining rendzinas over Corallian limestone at the higher altitudes to poorly-drained deep clay soils at lower altitudes (Avery, 1980). 3. Materials and methods A series of experiments were carried out to quantify the canopy effect on flora and fauna in temperate conditions: 1) Four native grass species were placed in canopy and open contexts, following a period in an unheated greenhouse for germination. They were harvested at set time intervals during the growing season (March to September), and measured for their carbon and nitrogen isotopic compositions. 2) Bone collagen from fallow deer Dama dama living within Wytham Woods was extracted and measured for its carbon and nitrogen isotopic compositions. 3) Wool from sheep reared in the pasture adjacent to Wytham Woods was collected and measured for its carbon and nitrogen isotopic compositions.
3.1. Measurement of d13C and d15N in four native grass species 2. Wytham woods The research area was Wytham Wood (51.460 N 1.200 W; UK National Grid SP 460 080), which is located to the west of Oxford
The grass seed was bought from Emorsgate Seeds (http:// wildseed.co.uk), a nursery specializing in autochthonous flower seeds and wild herbs from Great Britain, and grown in individual
Fig. 1. Map of Wytham Woods, and location of placement of grass samples.
3928
M. Bonafini et al. / Journal of Archaeological Science 40 (2013) 3926e3935
pots as monocultures. The species used were (descriptions from Hubbard, 1984):
2nd August (3rd sampling) 4th September (4th sampling)
Bromus erectus (Upright Brome, hereafter BE in data tables). A densely tufted perennial, 40e120 cm high. A coarse fibrous grass of well-drained calcareous soils, abundant on chalk and limestone downs, especially in southern England. Lolium perenne (Perennial Rye-Grass, hereafter LP). A loosely to densely tufted perennial, 10e90 cm high. It is a valuable grazing and hay grass, prominent in old pastures and meadows, especially on the rich heavy soils of the lowlands. It has been cultivated in England for about 300 years. Poa nemoralis (Wood Meadow-Grass, hereafter PN). A loosely tufted perennial, 15e90 cm high. It is a rather delicate grass typical of shady places. In most parts of the British Isles it is locally abundant in woods and hedgerows on sandy to heavy soils. Poa trivialis (Rough Meadow-Grass, hereafter PT). A loosely tufted perennial plant, 20e100 cm high. Very common in meadows and pastures of the lowlands, especially on rich moist soils, and it occurs sometimes in partial shade areas.
The samples were taken to the Research Laboratory for Archaeology and the History of Art and freeze-dried at 45 C under vacuum for 48e72 h. After freeze-drying the samples were ground into powder using a Retsch MM 300 grinder at a vibration frequency of 25 Hz for 3e5 min. For each sample 1e2 mg of powder was weighed into tin capsules for isotopic analysis. All samples were measured for d13C and d15N in RLAHA with the SERCON Geo20-20 IRMS mass spectrometer system operating in continuous flow mode coupled to a Carlo Erba elemental analyser. Samples were run against an in-house alanine standard which is regularly checked against international standards (IAEA N1, N2, CH6 and USGS 40, 41), thus the isotopic results of the samples analysed in this study are referable to the international V-PDB and AIR standards for carbon and nitrogen respectively.
The preparation and sowing of these grasses took place on the 11th of April 2011 in an unheated greenhouse at the Wytham Woods field laboratory station run by the Department of Plant Sciences, University of Oxford. Each species was sown in five medium-sized (L60 cm, W25 cm, H6 cm) vessels, all filled with the same soil, in order to have the same conditions for each species and plant. The vessels were watered regularly with fresh tap water every two days for the first month (AprileMay). In early May, the seedlings of 2e3 cm were re-potted into smaller pots (12 cm diameter, height 14 cm). During the transplant the plants were distributed with the same density in each pot. Sixty pots were prepared for each species, giving a total of 240 pots. During the period 11the13th May 2011 the 240 pots were placed in 12 locations (5 pots for every species for each location) around Wytham Woods, distributed as follows (Fig. 1):
The wooded area of Wytham Woods supports an ungulate population that includes roe deer, fallow deer and muntjac. A metal fence (2.5 m in height) now surrounds the woodland, ensuring that the animals live and feed only within the woodlands. The bones of twenty one specimens of fallow deer (D. dama) were analyzed for their collagen carbon and nitrogen isotopes to evaluate the extent of the canopy effect on this species. The bone samples were taken from a collection of 50 antlers and skulls belonging to roe deer, fallow deer and muntjac. These remains came from animals culled by the gamekeepers of Wytham over the past 20 years, to maintain the populations at sustainable levels. Bone chunks (4e5 g each) were taken (by cutting with snippers) from the front plate between the horns of 21 skulls of fallow deer (D. dama) and prepared for collagen extraction. Collagen was extracted from between 3.0 and 6.0 g of bone per sample using a modified Longin method, according to a standard protocol (O’Connell and Hedges, 1999). Any superficial material was removed from the bone by shotblasting. Lipids were removed from the bone samples by soaking them in a 2:1 chloroform (CHCl3)e methanol (CH3OH) solution, first in ultrasonic bath for 30 min and then overnight. The cleaning was repeated until the lipid components were completely eliminated. Samples were then demineralised in 0.5 M HCl at 4 C for several days, changing the solution every other day, rinsed with deionised water, and gelatinised in a pH 3 solution for 48 h at 75 C. The solution was filtered, frozen and freeze-dried. Between 1.0 and 2.0 mg of dried collagen was loaded into a tin capsule for continuous flow combustion and isotopic analysis using an automated Carlo Erba carbon and nitrogen elemental analyser coupled with a continuous flow isotope ratio monitoring mass spectrometer (PDZ Europa Geo 20/20). Results from samples with collagen yield <1% or C:N ratio >3.5 (Ambrose, 1990) should be considered unreliable and are excluded from the analysis. However, none of the samples analysed in this project were outside these parameters.
three in open pasture (locations No. 1e3) two in the open pasture but close to the edge of the woods (location No. 4, 30 m from the edge of the wood, and location No. 5, 10 m from the edge of the wood) two on the border between the pasture and the woods (locations No. 6 and 7) two in the woods but close to the open pasture (location No. 8, 10 m from the pasture, and location No. 9, 30 m from the pasture) three in the deep woods (locations No 10e12) Temperature was continuously recorded at each location at halfhour intervals from midday on 19/5/2011 until 8 am on 12/9/2011 using an iButton Data Logger DS1921. Rainfall was recorded by collection at each location starting on 13 May 2011 until 12 September 2011. Light intensity levels (lumens per square foot) were continuously measured at half-hourly intervals at two locations (location 1 in the open pasture and location 12 in the deep woodland) from 11 August 2011 until 11 September 2011 using a HOBO Pendant Temperature/Light Data Logger. Grass samples were taken from each pot at approximately monthly intervals, by clipping blades to give a mass of approximately 3 g from each pot. The sampling dates were (all in 2011): 9th May (done when the pots were still in the greenhouse, referred to as sampling period 0) 2nd June (referred to as 1st sampling) 27th June (2nd sampling)
3.2. Measurement of d13C and d15N in the bones of fallow deer from Wytham Woods
3.3. Measurement of d13C and d15N in the wool of sheep from the pastures adjacent to Wytham Woods The sheep sampled for wool were reared in the immediate area of Wytham Woods, where there are numerous pastures for sheep and cattle managed by FAI Farms Ltd. It was decided to collect some samples of wool from these sheep, since, in contrast to the ungulates, these sheep should not show the canopy effect. The sampling was performed in a pastured area 250 m north of Wytham
M. Bonafini et al. / Journal of Archaeological Science 40 (2013) 3926e3935
3929
Woods by simply picking up strands of wool from the area where the sheep were grazing. Twenty six samples of wool were collected in the pasture, from sheep that (according to the manager at FAI Farms Ltd.) fed purely by grazing, without any other kind of fodder. The samples of wool were cleaned prior to isotopic analysis using a washing cycle of distilled water and chloroformemethanol in an ultrasonic bath. After cleaning, the samples were freeze-dried and weighed into tin capsules before isotopic analysis, as described above.
(unless noted otherwise) of d13C and d15N for each grass species at each location. The values of d13C and d15N for the ungulate bone collagen (Table 2) and whole wool (Table 3) are the results of a single measurement for %C and %N, molar C/N ratio and values of d13C and d15N. The data were collected over 29 runs of the mass spectrometer from 18/7/2011 to 10/10/2011, and each batch contained (usually) five samples of our in-house alanine standard. As a result, we collected 215 replicates of the standard, which gave the following results:
4. Results
d13 C ¼ 26:88 0:24&; expected value 26:9&
The results are tabulated according to the experiments described above. Table 1 gives the average of a set of five replicates
d15 N ¼ 1:78 0:23&; expected value 1:6&
Table 1 Isotopic variation of d13C and d15N in four native herbaceous grass species according to sampling period and location. Location and species BE 1 BE 2 BE 3 BE 4 BE 5 BE 6 BE 7 BE 8 BE 9 BE 10 BE 11 BE 12 LP 1 LP 2 LP 3 LP 4 LP 5 LP 6 LP 7 LP 8 LP 9 LP 10 LP 11 LP 12 PN 1 PN 2 PN 3 PN 4 PN 5 PN 6 PN 7 PN 8 PN 9 PN 10 PN 11 PN 12 PT 1 PT 2 PT 3 PT 4 PT 5 PT 6 PT 7 PT 8 PT 9 PT 10 PT 11 PT 12
0 sampling (9th May)
1st sampling (2nd Jun)
2nd sampling (27th Jun)
3rd sampling (2nd Aug)
4th sampling (4th Sep)
d13C
d13C
d13C
d13C
d15N
d13C
d15N
28.9 0.1 29.2 0.5 29.7 0.7 28.9 0.4 29.6 0.4 30.0 0.4 34.4 0.2 34.6 0.6 n.d. 32.3 34.7 0.4 35.9 0.4 30.4 0.5 30.1 0.6 31.2 0.4 30.1 0.3 31.4 0.7 30.7 0.3 35.0 0.5 n.d. n.d. n.d. n.d. 34.6 28.3 0.7 28.3 1.0 30.0 0.3 28.3 0.8 28.6 0.5 28.2 0.5 33.7 0.3 34.5 0.3 n.d. n.d. n.d. 35.6 0.3 29.8 0.2 29.8 0.5 30.3 0.3 30.6 0.5 30.1 0.3 29.7 0.4 33.7 0.3 34.7 0.5 n.d. 34.6 0.5 35.54 0.2 35.8 0.4
0.3 0.1 0.3 0.1 0.8 0.8 0.6 0.4 n.d. 0.5 2.1 1.7 0.0 0.4 0.4 0.4 0.8 0.8 0.6 n.d. n.d. n.d. n.d. 0.4 0.7 1.0 0.3 0.2 0.3 1.2 0.5 1.7 n.d. n.d. n.d. 3.3 1.1 0.7 1.0 0.2 1.4 1.0 0.1 0.8 n.d. 1.3 1.8 1.8
29.1 0.3 29.7 0.5 30.0 0.4 29.3 0.1 29.8 0.3 29.4 0.5 33.6 0.2 n.d. n.d. n.d. n.d. n.d. 30.5 0.4 31.3 0.1 31.8 0.4 30.8 0.2 32.2 0.6 31.2 0.6 33.9 0.8 n.d. n.d. n.d. n.d. n.d. 27.1 0.4 28.9 0.8 28.9 0.9 28.3 0.6 29.2 0.7 29.4 0.7 32.8 0.9 34.0 0.1 n.d. n.d. n.d. n.d. 30.4 0.5 31.0 0.3 31.0 1.0 30.6 0.3 30.6 0.5 30.6 0.3 33.2 0.2 34.7 0.1 n.d. n.d. n.d. n.d.
1.16 0.7 1.75 0.8 0.73 1.4 0.97 1.3 0.30 2.0 0.04 0.2 3.41 1.2 n.d. n.d. n.d. n.d. n.d. 1.24 0.9 1.65 0.5 0.84 2.0 1.31 1.8 1.29 1.9 0.42 2.0 1.38 1.8 n.d. n.d. n.d. n.d. n.d. 1.08 1.3 2.20 0.7 0.74 2.2 2.14 1.2 1.38 1.8 0.37 0.3 0.24 2.2 1.75 0.8 n.d. n.d. n.d. n.d. 0.4 0.7 1.8 0.6 2.0 2.1 2.4 1.3 2.0 1.2 0.4 2.3 1.6 0.6 2.9 1.1 n.d. n.d. n.d. n.d.
28.6 28.1 28.7 28.7 28.8 28.2 28.8 28.5 29.0 29.2 29.6 29.2 30.6 30.6 30.9 30.7 31.0 31.3 31.2 31.0 30.9 31.0 31.3 31.4 30.4 30.1 29.8 29.7 29.7 30.3 29.9 15.2 30.2 30.5 30.4 30.6 30.9 30.8 30.8 30.9 30.6 31.1 30.2 31.1 30.7 30.9 30.8 30.5
d15N 0.3 0.3 0.4 0.5 0.4 0.6 0.4 0.3 0.5 0.6 0.3 0.5 0.5 0.4 0.1 0.4 0.2 0.3 0.2 0.3 0.7 0.6 0.3 0.2 0.1 0.5 0.3 0.6 0.4 0.3 0.4 0.2 0.3 0.5 0.1 0.2 0.1 0.3 0.1 0.1 0.2 0.2 0.3 0.3 0.2 0.2 0.4 0.2
1.3 2.0 1.6 0.9 1.5 0.7 0.8 1.0 1.6 0.4 0.8 1.7 1.1 1.7 1.1 0.6 1.0 1.2 1.5 0.9 1.2 0.1 0.4 0.3 3.4 2.2 3.9 3.9 3.9 4.2 4.3 4.7 2.0 1.2 1.4 2.7 0.5 0.5 0.7 0.2 0.2 0.1 0.3 0.1 0.3 1.0 0.4 3.2
0.3 0.3 0.3 0.6 0.6 0.6 0.5 0.5 0.5 1.3 0.9 0.8 0.7 0.9 0.3 0.8 0.5 0.4 0.3 0.8 0.8 1.3 0.7 0.8 0.5 1.5 1.1 0.8 0.8 0.6 0.4 0.7 1.0 0.7 0.8 0.6 0.8 0.9 0.6 0.4 0.5 0.3 0.5 0.4 1.2 0.5 0.8 2.7
27.1 26.9 28.8 27.7 26.4 27.7 31.7 31.4 31.0 30.9 31.1 31.7 27.6 28.0 28.8 29.3 28.3 27.4 32.9 33.7 32.3 33.1 33.1 33.6 27.7 27.6 28.4 27.1 27.0 27.3 33.3 33.9 32.7 32.9 32.8 34.3 27.6 28.7 29.2 28.5 28.5 27.6 33.1 33.4 31.3 32.7 32.2 32.6
d15N 0.5 0.4 0.7 0.6 0.4 0.2 0.6 0.4 0.3 0.5 0.4 0.6 0.3 0.5 0.7 0.6 0.5 0.3 0.2 0.9 0.7 0.7 0.2 0.5 0.5 0.3 1.0 0.5 0.8 0.7 0.1 0.3 0.3 0.7 0.3 0.3 0.7 0.2 0.5 0.7 0.2 0.6 0.6 0.5 0.8 0.7 0.7 0.9
0.1 1.2 1.5 0.6 0.9 0.8 0.2 0.7 2.0 1.3 1.3 0.6 0.9 0.2 1.0 0.3 0.9 0.5 0.2 1.5 0.5 0.6 0.5 0.0 0.3 0.1 0.9 0.3 0.7 0.5 2.2 2.8 2.4 2.1 2.3 2.5 0.3 0.1 0.2 0.2 0.7 0.4 0.3 0.3 0.3 0.3 0.8 0.5
0.8 1.1 0.8 0.6 0.4 0.4 0.6 0.7 0.5 0.6 0.5 0.5 0.6 0.8 0.9 0.6 0.4 0.4 0.5 0.3 0.2 0.3 0.4 0.9 0.8 0.5 1.9 0.5 0.7 0.4 1.2 0.6 0.6 1.0 0.9 0.4 1.0 0.6 0.3 0.8 0.5 0.6 04. 0.4 0.3 0.4 0.6 0.5
27.9 27.7 28.5 28.4 28.5 28.5 32.5 33.5 32.2 33.1 33.7 34.2 28.9 28.4 29.7 29.7 30.3 28.8 32.9 35.3 34.2 33.9 33.9 35.6 26.2 26.3 27.1 27.0 25.8 27.0 33.3 35.0 35.3 34.1 34.6 35.4 28.6 29.0 30.1 29.0 29.2 28.2 32.9 34.7 33.6 33.4 33.7 35.4
d15N 0.3 0.5 0.4 0.4 0.4 0.2 0.6 0.4 1.2 0.7 0.6 0.8 0.7 0.7 0.3 0.7 0.3 0.5 0.3 0.5 0.4 0.8 0.5 0.4 0.8 0.6 0.6 0.6 0.3 0.9 0.4 0.1 0.2 0.7 0.2 0.2 0.8 0.5 0.4 0.5 0.3 0.4 0.3 0.4 0.5 0.5 0.4 0.4
0.2 0.0 0.6 0.3 0.1 1.1 1.3 0.2 1.6 1.0 1.4 0.9 0.7 0.7 0.1 0.2 0.44 1.7 1.4 0.4 1.1 0.1 0.5 0.6 0.1 1.6 1.0 0.6 0.2 1.1 1.0 1.7 3.6 2.8 3.0 3.1 0.6 0.3 0.9 0.4 0.1 0.8 1.3 0.1 0.7 0.5 1.2 1.3
0.8 0.6 1.2 1.0 0.9 0.3 0.9 0.5 0.2 1.4 0.4 0.2 1.0 0.4 0.5 0.2 0.6 0.5 0.3 0.5 0.4 0.7 0.4 0.3 0.7 2.4 1.7 0.8 1.7 0.7 1.0 0.8 0.6 1.2 0.4 0.5 0.9 0.8 1.3 1.2 0.9 0.7 0.5 0.4 0.6 0.3 0.6 0.5
All values quoted are the mean and one standard deviation calculated from five measurements, unless noted as follows. Values in italics are from four measurements. Values in bold italics are from three measurements. Values in bold are from two measurements. Values with no associated standard deviation are from single measurements. n.d. ¼ not determined, usually because no grass was collectable.
0.3 0.9 0.9 0.3 0.9 0.4 1.3 0.6
0.4 0.6 1.0 1.3 0.8 0.4 1.3 0.6 0.5
1.3 1.3 0.8 1.4 0.9 0.3 0.9 0.6
1.0. 1.8 0.6 0.5 1.0 0.8 1.1 0.5 1.2
0.7 0.5 0.3
3930
M. Bonafini et al. / Journal of Archaeological Science 40 (2013) 3926e3935
Sample
Collagen yield (%)
%C
%N
d13C
d15N
C/N
d13C corra
C1 C2 C3 C4 C5 C6 C7 C8 C9 C10 C11 C12 C13 C14 C15 C16 C17 C18 C19 C20 C21
n.d 19.3 28.2 12.9 11.8 11.6 11.3 13.5 10.9 13.5 11.9 28.3 15.4 12.6 14.7 13.0 13.0 14.3 10.7 n.d. n.d.
36.3 33.7 31.6 32.0 37.8 34.5 41.2 40.8 31.4 40.2 34.1 28.9 32.0 35.5 41.7 33.6 33.9 42.2 35.9 41.1 38.6
13.4 12.5 11.7 12.0 13.3 12.9 15.1 14.8 11.6 14.6 12.4 10.7 11.8 12.9 15.2 12.2 12.3 15.3 13.1 14.6 13.8
23.6 23.8 23.5 23.2 24.1 23.6 24.2 24.0 23.8 23.9 24.1 23.8 23.8 24.0 24.1 24.1 20.6 22.7 24.4 24.6 24.2
3.6 5.9 4.9 4.8 4.9 7.8 5.6 5.0 5.0 4.5 5.2 5.0 4.9 5.5 4.7 5.0 7.3 5.6 4.8 6.7 6.9
3.14 3.15 3.15 3.10 3.31 3.12 3.20 3.21 3.15 3.21 3.21 3.15 3.17 3.20 3.20 3.22 3.21 3.21 3.20 3.29 3.26
22.4 22.6 22.3 22 22.9 22.4 23.0 22.8 22.6 22.7 22.9 22.6 22.6 22.8 22.9 22.9 19.4 21.5 23.2 23.4 23.0
a These values of d13C have been corrected for the modern fossil fuel effect by adding 1.2& (Friedli et al. 1986) for comparison with archaeological data.
5. Discussion 5.1. Isotopic variation in four native grass species growing in canopy and open conditions 5.1.1. Inherent variation of d13C and d15N in grasses growing under the same conditions Because all four grass species were initially sown in pots containing the same growing medium, and subjected to the same watering regime in the same environment before they were moved to the specified locations, the data from sampling period 0 (9th May) affords an opportunity to look at the inherent isotopic Table 3 Isotopic measurements on wool from the sheep present in a pasture adjacent to Wytham Woods. Sample
%C
%N
d13C
d15N
C/N
W1 W2 W3 W4 W5 W6 W7 W8 W9 W10 W11 W12 W13 W14 W15 W16 W17 W18 W19 W20 W21 W22 W23 W24 W25 W26
39.6 42.8 48.4 44.8 46.2 45.1 46.9 45.2 44.6 50.9 48.0 50.6 49.4 45.8 47.9 48.0 48.7 49.5 51.0 44.8 44.7 46.1 46.1 44.8 46.3 45.8
13.4 14.3 15.9 15.0 15.3 15.2 15.6 15.0 15.2 17.2 16.2 17.1 16.6 15.7 16.3 16.2 16.5 16.7 17.1 14.9 14.9 15.4 15.5 15.0 15.8 15.2
26.7 26.1 26.8 26.2 26.1 26.2 26.0 26.9 26.9 26.0 26.5 25.6 26.4 25.8 25.8 25.9 25.0 26.0 26.2 25.7 25.8 25.6 25.1 26.0 25.4 26.0
9.0 7.4 8.8 8.3 8.9 9.3 8.8 9.1 8.6 9.3 8.5 8.5 8.8 9.0 8.9 8.4 9.1 8.7 8.8 9.3 9.1 10.0 9.6 9.6 9.3 8.4
3.44 3.49 3.54 3.49 3.52 3.46 3.51 3.51 3.43 3.43 3.46 3.44 3.46 3.42 3.43 3.46 3.44 3.45 3.48 3.50 3.50 3.48 3.46 3.47 3.44 3.52
δ15N AIR (‰)
Table 2 Isotopic measurements on the collagen from fallow deer present in Wytham Woods.
4 3 2 1 0 -1 -2 -3 -4 -5 -6
BE LP PN PT
-32
-31
-30 -29 δ13C V-PDB (‰)
-28
-27
Fig. 2. Isotopic plot of four grass species grown under the same conditions.
variation within these species. Fig. 2 shows an isotopic plot of all values recorded on these four species at sampling time 0 (note that each point plotted is the average of five measurements). The averages and standard deviation of all plants for each species are shown in Table 4. There is clearly significant variation between the species, with a maximum difference between the species average of 2.0& in d13C and 4.1& in d15N. The variation within each species is also significant, ranging from 0.3& to 0.4& in d13C (one standard deviation) and 0.5& to 1.2& in d15N. Since these are from grasses growing under the same conditions, these figures must represent the minimum isotopic variation to be expected at the base of the food chain. Larger variation should be expected if different grass species are growing in the same area. 5.1.2. Comparison of d13C and d15N in grasses grown in open and shade conditions Fig. 3 shows the average isotopic value of d13C in B. erectus at each of the 12 sampling locations for each of the five sampling episodes. The other three species show very similar patterns. From the relatively homogeneous values at sampling period 0 (in the greenhouse), there are two basic patterns of behaviour, one followed by grass at sample locations 1e6, and one for locations 7e12. Fig. 1 shows that locations 1e5 are either in fully open conditions or are more marginal but still in the pasture, and 8e12 are either in deep woodland or more marginal but still in the woodland. Locations 6 and 7 are both on the border, but with different aspects. The isotopic data suggest that location 6 should be included in the open samples, and 7 the shaded. Further evidence for this grouping is presented below from an analysis of the environmental data collected. Fig. 4 shows the average isotopic value of d13C in all species grouped into either locations 1e6 (termed ‘open’) or 7e12 (closed) for each of the five sampling episodes. The values of d13C found for the plants growing in woodland, regardless of the species, are depleted in 13C compared to the normal range of values in C3 plants reported in the literature (van der Merwe and Medina, 1991), and also compared to the equivalent species growing in an open environment. The literature describes the depletion of 13C due to a canopy effect as a reduction between 2 and 5& compared to the values of d13C for the same plant in a non-woodland environment
Table 4 Average values of d13C and d15N for each species of grass at sampling period 0. Species
d13C (&)
BE LP PN PT
28.8 31.0 30.2 30.8
d15N (&) 0.4 0.3 0.3 0.2
1.2 0.9 3.1 0.2
0.5 0.6 1.2 1.1
M. Bonafini et al. / Journal of Archaeological Science 40 (2013) 3926e3935
3931
Table 5 Average differences in d13C and d15N values (&) between open (1e6) and shaded (7e12) locations for each grass species. Species
Fig. 3. Isotopic variation in d13C in Bromus erectus at each sampling location for each sampling date.
(van Der Merwe and Medina, 1989; Broadmeadow and Griffiths, 1993; France, 1996). The differences between the average values of d13C and d15N between grasses placed under the canopy (locations 7e12) compared to those located in open areas (locations 1e6) for the four species are shown in Table 5. Ignoring the sampling on the 9th of May (carried out before the setting out of the seedlings), the overall average of the difference in d13C (4.8&) is towards the maximum of the values discussed in the literature. The maximum difference observed for any one species (P. nemoralis on 27th June) has a value of 8.1&. This high difference is primarily due to the very high (i.e., less negative) average value of the grasses located in the pasture (26.6&), the explanation for which could be the observed production of inflorescences (ears) that occurred in the plants of P. nemoralis placed in the pasture locations during June. To prevent the setting of seeds (more likely in the open settings) influencing the isotopic values in the blades, it was decided to prune the seed heads, and to perform the pruning in every location for all the samples. It is shown in the literature that the production of seeds (ears) may lead to an enrichment in d13C because of the production of greater amounts of protein in the seeds (Tieszen, 1991). This hypothesis could explain the value reported on 27th June for the plants of P. nemoralis grown in the pasture, being further confirmed by the subsequent trend towards more depleted values after pruning (the plants were no longer in the maturation phase).
1st sampling 2nd Jun
2nd sampling 27th Jun
3rd sampling 2nd Aug
4th sampling 4th Sep
Average
Bromus erectus Lolium perenne Poa nemoralis Poa trivialis
d13C
3.9 4.9 5.8 4.2
5.0 5.1 8.1 5.0
5.0 4.2 6.0 4.8
4.0 2.6 4.8 3.3
4.5 4.2 6.2 4.3
Bromus erectus Lolium perenne Poa nemoralis Poa trivialis
d15N
1.8 1.1 2.7 0.4
0.9 1.0 2.4 0.5
0.5 0.5 2.1 1.0
2.7 0.4 0.3 0.9
1.5 0.7 1.7 0.7
The data for d15N variation through the growing season show a less marked patterning than those of d13C, and the divergence between individual locations is greater, but nevertheless some general trends can be observed. The open locations (1e6) show an enrichment of d15N (typically þ2&) in BE, LP and PN immediately following the relocation of the plants from the greenhouse, followed by either a stable value (PN) or a decline (BE, LP) through the rest of the season. d15N in PT in open environments remains approximately constant throughout the season. The closed environments (7e12) remain approximately constant through the season for all species. As shown in Table 5, the average difference in d15N between the open and shaded locations (average value locations 1e6 minus average value locations 7e12) is positive for all species, ranging from þ0.7& to þ1.7&. It is likely that the use of the same compost for each pot in each location has affected the observed differences in the d15N values, which are mainly influenced by the type of soil and the abundances of nitrogen compounds present (Michener and Lajtha, 2007). Nevertheless there is some slight evidence here for a ‘canopy effect’ in the nitrogen values. 5.2. Correlation of climatic parameters with d13C isotopic measurements in grass species 5.2.1. Temperature and Growing-degree days (GDD) In order to correlate the isotopic values measured in the grass with the recorded air temperature at each location, we have transformed the recorded temperature data into Growing-degree days (GDD). This construct is used by farmers and horticulturists to calculate the progress of a plant towards maturity. Under normal conditions, plants grow in response to cumulative daily air temperature, but only if the temperature is above a minimum cut-off for the particular species. GDD for one day is calculated as:
GDD ¼ ðTmax þ Tmin Þ=2 Tcrit
Fig. 4. Average value of d13C in all species grouped into either locations 1e6 (open environment) or 7e12 (closed environment) on each of the five sampling dates.
where Tmax and Tmin are the maximum and minimum recorded daily temperatures and Tcrit is the cut-off temperature for that particular species. The total GDD is accumulated over the growing season. Tcrit for most cases is taken as 10 C, and it is usual to cap the maximum temperature for temperate crops at 30 C (i.e., to record Tmax as 30 C if Tmax is 30 C or above), since plant growth response falls off at higher temperatures. For example, wheat (Triticum aestivum) is said to emerge at 143e178 GDD, and to reach maturity at 1550e1680 GDD (Wittwer, 1995) if the baseline is taken as 10 C. The temperatures recorded at each location (1e12) were inspected to give Tmax and Tmin for each day from 19/5/2011 to 12/9/ 2011, and the cumulative GDD totals calculated at each location on 2nd June, 27th June, 2nd August and 4th September, corresponding to the dates at which the grass was harvested for analysis. Fig. 5
3932
M. Bonafini et al. / Journal of Archaeological Science 40 (2013) 3926e3935
Fig. 5. GDD calculated at each location for each sampling date. Fig. 7. Cumulative rainfall at each location for each sampling date.
shows the accumulated GDD at each location. This shows that the locations can be clustered into two broad groups with similar characteristics: a low-GDD group (the ‘closed’ environments average GDD on 4th September ¼ 396) consisting of locations 7, 8, 9, 10, 11 and 12 (the temperature logger at location 11 failed on 27th July, but the pattern prior to this date is very similar to locations 10 and 12), and a high GDD group (the ‘open’ environments), consisting of locations 1e6, with location 2 being exceptional having a GDD value of 885 compared to the average of the other five of 659 on 4th September. Clearly these groupings are primarily controlled by the degree of open or shade (see location descriptions above), and it does support the suggestion given above that location 6 is best considered to be part of the open group, and location 7 the shaded group. However, a plot of GDD calculated at each of the sampling days against measured d13C values for each species at each sampling (Fig. 6) shows no relationship between GDD and d13C, suggesting that GDD and therefore temperature is not the prime factor in determining the carbon isotopic value in the plant tissue. 5.2.2. Rainfall The plot of cumulative rainfall for each location is similar to the GDD curves shown in Fig. 7, except that in this case location 6 (one of the marginal locations) is not grouped with locations 1e5 (the open locations), as in the GDD figure, but actually lies mostly below the group of closed location values (7e12). The rainfall data shows that location 2 received the most rainfall (690 ml over the measurement period), with locations 1e5 having a total amount greater than 90% of this value. Locations 6e12 had between 63 and 78% of this value. We are not in a position to calculate water use efficiency (WUE) or estimate the soil moisture content at each location but simply plotting cumulative rainfall (approximately at the date of
Fig. 6. d13C plotted against GDD for all species of grass.
sampling) against d13C value by species shows no relationship between d13C and rainfall (not shown). 5.2.3. Light intensity Light intensity was continuously measured at half-hourly intervals for Location 1 (open situation) and Location 12 (closed situation) from 11th August to 11th September. The average ratio of the areal light intensity at location 1 to that at location 12 during daylight hours is 17 (median value of ratio 15), so we can conclude that the light is on average approximately 15 times more intense in the open areas. It has been shown in tropical environments (Zimmerman and Ehleringer, 1990) that the d13C in plant tissue (expressed as D, the carbon isotope discrimination) is related to the daily photon flux, with d13C becoming less negative as the total photon flux increases. Table 5 and Fig. 4 (above) show a clear difference in the isotopic values of the grasses grown in the shady environments compared to the same species in open situations, in accord with these observations. On average, the grasses growing in shady environments are more depleted by between 4.2 and 6.2& in d13C, whilst the d15N values are enriched by between 0.7 and 1.7&. Although we are not in a position to definitively demonstrate this relationship, it does indicate that differences in light intensity have a stronger influence on the value of d13C in the grass tissue than temperature or rainfall, and suggests that further work is necessary to quantify this relationship. 5.3. Values of d13C and d15N in the bones of fallow deer present in Wytham Woods Table 2 shows that the bones analyzed have an average C/N ratio of 3.19 (within the acceptance range of 2.9e3.6) and high levels of carbon and nitrogen in the collagen (averages 36.0% and 13.2% respectively), which defines good preservation for the collagen (DeNiro, 1985). The collagen yields (where measured) range from 10.7 to 28.3% (average 14.8%, where the exception would be c. 20%) which is slightly surprising since these bones are recent samples that should not have undergone any alteration process. The isotopic values are plotted in Fig. 8. The majority have d13C values in a range from 22.5& to 24.5& (overall average of these 23.9 0.4&), with sample C17 (20.6&) showing a slight enrichment. The average value for d13C including C17 is 23.7 0.8&. The majority of the d15N values are clustered between 3.6 and 5.9& (average of these 5.0 0.5&), with four samples having values above 6& (C6 7.8&; C17 7.3&; C20 6.7&; C21 6.9&). The overall average of d15N is 5.4 1.0&. The purpose of analysing collagen from the bone of ungulates was to use these data as indicators for species expected to be living
M. Bonafini et al. / Journal of Archaeological Science 40 (2013) 3926e3935
Fig. 8. Isotopic values of d13C and d15N from fallow deer living within Wytham Woods.
in shaded habitats, and therefore expected to show a ‘canopy effect’. However, the samples available were from fallow deer which are classically a forest edge species that typically feeds in the open on grass, either in open fields, clearings or rides. The fact that these animals are most frequently seen in the woods, or may have been culled there, is because this where they go when disturbed or are not feeding, thereby giving the impression that they are mainly woodland animals. Fallow deer are generalist feeders and may browse forest plants as they move out to grass, and if hungry they will also eat leaves fallen from the canopy. At the time these animals died the population density was very high and there had been very little thinning of the woods and so the ground flora within the woodland would have been shaded out, meaning there was little to eat within the woodland. It is likely therefore that these deer were probably eating out in the open on unshaded plants. The limited isotopic bone collagen data on modern and archaeological fallow deer bone has recently been compiled by Sykes et al. (2011). This shows that only eight recent (Roman) archaeological samples have values for both d13C and d15N e the other archaeological values (Sykes et al. 2011; Table 1) have only d13C measurements, and these authors were not able to locate any modern values. Bocherens et al. (1999) have published data on 10 fallow deer from Scladina Cave, Belgium, dating to c. 100 ka BP, but these are not discussed here. The Roman and modern data are shown in Fig. 9. Because of the well-known fossil fuel effect (the reduction of d13C in atmospheric CO2 values over the last 200 years as a result of the increased burning of fossil fuel), our values on recent deer are not directly comparable with those from archaeological samples. To account for this, we have added 1.2& to the d13C values for the deer from Wytham Woods plotted in Fig. 9, as shown in Table 2 (Friedli et al. 1986).
Fig. 9. Isotopic values from modern and Roman fallow deer.
3933
The Wytham Woods deer are on average slightly more negative in d13C (22.6 0.8& when corrected for the fossil fuel effect) than the Roman fallow deer from Monkton, Kent (average d13C value from eight measurements on possibly three individuals 21.0 0.5&: Sykes et al. 2011, 160), whereas the d15N values are comparable (Monkton average 6.5 1.0&, compared to 5.4 1.0&). The difference in d13C between these two populations is, however, significant (t z 6, n ¼ 27, tcrit z 2). This might suggest that the deer from Wytham Woods are subject to a slight canopy effect, since the Roman samples are believed to come from a managed breeding herd, which implies a deer park (open) environment. Despite the fact that the fallow deer in Wytham Woods might be expected not to show a canopy effect because of their ecology, our data suggest that one might be at least partially present, although uncertainties introduced by the fossil fuel correction might also account for this difference. If this is the case, the explanation for this could be the fence along the border of the woods, with a height of 2.5 m. The fence was installed as a result of numerous complaints from farms in the neighbourhood about the frequent excursions of roe and fallow deer in search of food, thus preventing them from accessing fully open grazing. 5.4. Values of d13C and d15N in the wool of sheep from pasture adjacent to Wytham Woods Fig. 10 shows the values of d13C and d15N in the analyzed wool from the sheep grazing near Wytham Woods. All values are very close to each other, with a range of d13C from 25.5& to 26.5& (average value 26.0 0.5&) and d15N from 8.0& to 9.5& (average value 8.9& 0.5). For comparison, also shown are some unpublished data from the wool of rare breed sheep from southern England (Good, Ditchfield, Pollard, unpublished), which have an average value of d13C ¼ 25.5 0.5& and d15N 7.3& 0.6. These values make a useful comparison, because it is known that these sheep grazed only in open areas. It can be seen that the carbon isotope values overlap completely, whereas the d15N values in the Wytham flock are offset by approximately þ1.5& (presumably because of variations in underlying geology or rainfall). The overlap in d13C is sufficient to confirm that there is no discernible canopy effect in the d13C values of wool from the Wytham sheep. 5.5. Correlation between isotopic values in sheep wool and ungulate collagen with grass values In order to compare the faunal data with the grass values, we have divided the grass data into those relating to open environments (locations 1e6) and those from closed environments (locations 7e12), for all four species and at all sampling periods (except
Fig. 10. Isotopic values of d13C and d15N in wool from sheep living in pasture near Wytham Woods compared to unpublished data from other modern southern English sheep.
3934
M. Bonafini et al. / Journal of Archaeological Science 40 (2013) 3926e3935
variation within each species, ranging from 0.3& to 0.4& in
d13C (one standard deviation) and 0.5& to 1.2& in d15N. These
Fig. 11. Comparison of all isotopic data, showing d13C values for wool and collagen converted to a ‘dietary’ value.
sampling period 0, which relates to the period in the greenhouse). The average values for these two groups are d13Copen ¼ 29.0 1.4&, d15Nopen ¼ 0.1 0.9&, d13Cclosed ¼ 33.6 1.2&, d15Nclosed ¼ 1.1 1.1&. Since the average values for the wool are d13C ¼ 26.0 0.5& and d15N ¼ 8.9 0.5&, the offset D (wool open grass) between these two is therefore D13C ¼ þ3.0& and D15N ¼ þ9&. The offset between average fallow deer collagen (d13C ¼ 23.7 0.8&, d15N ¼ 5.4 1.0&) and closed grass D (collagen closed grass) is D13C ¼ þ5.9& and D15N ¼ þ6.5&. In order to test the hypothesis that the sheep have grown on a diet of grass from the open pasture, and to measure the extent to which the ungulates may have been reared in the closed canopy, we have applied correction values from the literature to convert d13C in the animal tissue values to an underlying diet d13C value. The correction factors used here are: For animal hair: d13C ¼ 3& (Sponheimer et al. 2003; Drucker et al. 2008) For bone collagen: d13C ¼ 5& (van der Merwe, 1989; Drucker et al. 2008) This gives an average dietary d13C value for the sheep of d13Ccor ¼ 29.0 0.5&, and for the ungulates d13Ccor ¼
28.7 0.8&. These values are to be compared with grass growing in an open pasture of d13Copen ¼ 29.0 1.4&, and d13Cclosed ¼ 33.0 1.4 Fig. 11 shows a comparison of all the isotopic data, but with the individual d13C values for wool and collagen converted to a ‘dietary’ value using the factors listed above. The results of d13Ccor of the sheep wool are seen to have the same d13C values as the grass growing in an open environment, but the ungulates’ d13Ccor values are similar to those of the sheep, and more consistent with the d13C values of the grass growing in an open environment than those of the woodland context. For the sheep, this is as expected, but for the fallow deer it supports the observation that these animals feed predominantly on pasture from open environments (as expected from the ecological observations), rather than from the closed environments. 6. Conclusions We have shown that, even for grasses growing in the same environmental conditions (sampling period 0, in the greenhouse), there is significant variation in the isotopic values recorded between grass species (maximum difference between averages of 2.0& in d13C and 4.1& in d15N). There is also significant isotopic
values must represent the minimum isotopic variation to be expected at the base of the food chain, and perhaps call in to question the precision which can be derived from some of the dietary reconstruction models now used, since these generally assume that dietary inputs are ‘point’ isotopic sources. The experiments showed a strong variation in d13C values for grass growing in a closed woodland environment compared to an open environment. This variation of the d13C values in the grass can be attributed to a canopy effect of approximately 5& in d13C present in the temperate woodland. Of the two phenomena postulated to cause a canopy effect (shading vs. respired CO2), we have been able to infer an inverse relationship between d13C in the leaves and the integrated light intensity received by the plants, suggesting that differences of levels of light irradiation are important, but we cannot rule out a respired CO2 effect from current data. We have also shown that the d13C values in the whole wool shed by sheep feeding in open pasture near the measurement site have values consistent with those to be expected from herbivores grazing on grass growing in a fully open environment. The d13C values derived from fallow deer bone collagen, although somewhat more depleted than the limited set of comparative data available, are also fully consistent with those expected from animals grazing in an open habitat. Although conventionally thought of as being woodland animals, which therefore ought to show a canopy effect, when looked at in more detail, they do not. This is consistent with the information we have for the feeding behaviour of these animals. We conclude that the canopy effect is a real effect in temperate woodland, and that the degree of shift in the d13C values is likely to be related to differential shade levels corresponding to the percentage of woodland cover, but we are not yet in a position to fully calibrate the effect. We have also shown that the expectation that fauna living within the woodland will show such an effect needs to be moderated by detailed ecological knowledge. Acknowledgements The authors would like to acknowledge the help of Dr Nick Brown, Department of Plant Sciences, University of Oxford in arranging access to Wytham Woods, and also the assistance of the technical staff at the research facility. We would also like to thank Dr Stephen Ellwood, WildCRU, University of Oxford for his assistance and for providing the deer samples analyzed in this study and two anonymous reviewers for constructive comments. References Ambrose, S.H., 1990. Preparation and characterization of bone and tooth collagen for isotopic analysis. Journal of Archaeological Science 17, 431e451. Avery, B.W., 1980. Soil Classification for England and Wales. In: Soil Survey Technical Monograph No.14. Harpenden. Bocherens, H., Fogel, M.L., Tuross, N., Zeder, M., 1995. Trophic structure and climatic information from isotopic signatures in Pleistocene cave fauna of southern England. Journal of Archaeological Science 22, 327e340. Bocherens, H., Billiou, D., Mariotti, A., Patou-Mathis, M., Otte, M., Bonjean, D., Toussaint, M., 1999. Palaeoenvirnonmental and palaeodietary implications of isotopic biogeochemistry of Last Interglacial Neanderthal and mammal bones in Scladina Cave (Belgium). Journal of Archaeological Science 26, 599e607. Broadmeadow, M.S.J., Griffiths, H., 1993. Carbon isotope discrimination and the coupling of CO2 fluxes within forest canopies. In: Ehleringer, J.R., Hall, A.E., Farquhar, G.D. (Eds.), Stable Isotopes and Plant Carbon-water Relations. Academic Press, San Diego, pp. 109e130. Cerling, T.E., Harris, J.M., 1999. Carbon isotope fractionation between diet and bioapatite in ungulate mammals and implications for ecological and paleoecological studies. Oecologia 120, 347e363. DeNiro, M.J., 1985. Postmortem preservation and alteration of in vivo bone collagen isotope ratios in relation to palaeodietary reconstruction. Nature 317, 806e809. Drucker, D., Bocherens, H., Marriotti, A., 2000. Contribution de la biogéochimie isotopique à l’étude de la paléobiologie des grands mammifères du Pléistocène
M. Bonafini et al. / Journal of Archaeological Science 40 (2013) 3926e3935 supérieur: application aux rennes et aux chevaux magdaléniens et solutréens du Sud-Ouest de la France. Mémoire de la Société Préhistorique Française 28, 13e27. Drucker, D.G., Bridault, A., Hobson, K.A., Szuma, E., Bocherens, H., 2008. Can carbon13 in large herbivores reflect the canopy effect in temperate and boreal ecosystems? Evidence from modern and ancient ungulates. Palaeogeography, Palaeoclimatology, Palaeoecology 268, 69e82. Farquhar, G.D., O’Leary, M.H., Berry, J.A., 1982. On the relationship between carbon isotope discrimination and the intercellular carbon dioxide concentration in leaves. Australian Journal of Plant Physiology 9, 121e137. Fizet, M., Mariotti, A., Bocherens, H., Lange-Badré, B., Vandermeersch, B., Borel, J.P., Bellon, G., 1995. Effect of diet, physiology and climate on carbon and nitrogen isotopes of collagen in a late Pleistocene anthropic palaeoecosytem: Marillac, Charente, France. Journal of Archaeological Science 22, 67e79. France, R., 1996. Carbon isotope ratios in logged and unlogged boreal forests: examination of the potential for determining wildlife habitat use. Environmental Management 20, 249e255. Friedli, H., Lötscher, H., Oeschger, H., Siegenthaler, U., Stauffer, B., 1986. Ice core record of 13C/12C ratio of atmospheric CO2 in the past two centuries. Nature 324, 237e238. Hubbard, C.E., 1984. Grasses. A Guide to Their Structure, Identification, Uses and Distribution in the British Isles, third ed. Penguin Books, Harmondsworth. Iacumin, P., Nikolaev, V., Ramigni, M., 2000. C and N stable isotope measurements on Eurasian fossil mammals, 40 000 to 10 000 years BP: herbivore physiologies and palaeoenvironmental reconstruction. Palaeogeography, Palaeoclimatology, Palaeoecology 163, 33e47. Krigbaum, J., 2003. Neolithic subsistence patterns in northern Borneo reconstructed with stable carbon isotopes of enamel. Journal of Anthropological Archaeology 22, 292e304. Michener, R.H., Lajtha, K., 2007. Stable Isotopes in Ecology and Environmental Science, second ed. Blackwell Publishing, Oxford. O’Connell, T.C., Hedges, R.E.M., 1999. Isotopic comparison of hair and bone: archaeological analyses. Journal of Archaeological Science 26, 661e665.
3935
Rodière, É., Bocherens, H., Angibault, J.-M., Mariotti, A., 1996. Particularités isotopiqes de l’azote chez le chevreuil (Capreolus capreolus L.): implications pour les reconstitutions paléoenvironnementales. Comptes Rendus de l’Academie des Sciences, Paris, Series IIA 323, 179e185. Sponheimer, M., Robinson, T., Ayliffe, L., Passey, B., Roeder, B., Shipley, L., Lopez, E., Cerling, T., Dearing, D., Ehleringer, J., 2003. An experimental study of carbonisotope fractionation between diet, hair and feces of mammalian herbivores. Canadian Journal of Zoology 81, 871e876. Stewart, J.R., Lister, A.M., 2001. Cryptic northern refugia and the origins of the modern biota. Trends in Ecology and Evolution 16, 608e613. Sykes, N.J., Baker, K.H., Carden, R.F., Higham, T.F.G., Hoellzel, A.R., Stevens, R.E., 2011. New evidence for the establishment and management of the European fallow deer (Dama dama dama) in Roman Britain. Journal of Archaeological Science 38, 156e165. Tieszen, L.L., 1991. Natural variations in the carbon isotope values of plants: implications for archaeology, ecology, and paleoecology. Journal of Archaeological Science 18, 227e248. van der Merwe, N.J., 1989. Natural variation in 13C concentration and its effect on environmental reconstruction using 13C/12C ratios in animal bones. In: Price, T.D. (Ed.), The Chemistry of Prehistoric Human Bone. Cambridge University Press, Cambridge, pp. 105e125. van Der Merwe, N.J., Medina, E., 1989. Photosynthesis and 13C/12C ratios in Amazonian rain forests. Geochimica et Cosmochimica Acta 53, 1091e1094. van der Merwe, N.J., Medina, E., 1991. The canopy effect, carbon isotope ratios and foodwebs in Amazonia. Journal of Archaeological Science 18, 249e259. Vogel, J.C., 1978. Recycling of CO2 in a forest environment. Oecologia Plantarum 13, 89e94. Wittwer, S.H., 1995. Food, Climate, and Carbon Dioxide: the Global Environment and World Food Production. Lewis, Boca Raton. Zimmerman, J.K., Ehleringer, J.R., 1990. Carbon isotope ratios are correlated with irradiance levels in the Panamanian orchid Catasetum viridiflavum. Oecologia 83, 247e249.
Contents lists available at SciVerse ScienceDirect
Journal of Archaeological Science journal homepage: http://www.elsevier.com/locate/jas
Investigation of the ‘canopy effect’ in the isotope ecology of temperate woodlands M. Bonafini a, b, M. Pellegrini a, *, P. Ditchfield a, A.M. Pollard a a b
Research Laboratory for Archaeology and the History of Art, School of Archaeology, University of Oxford, South Parks Road, Oxford OX1 3QY, UK Department of Environmental Sciences, Università Ca’ Foscari, Calle Larga Santa Marta 2137, 30123 Venezia, Italy
a r t i c l e i n f o
a b s t r a c t
Article history: Received 14 August 2012 Received in revised form 8 March 2013 Accepted 31 March 2013
Anomalously 13C-depleted carbon stable isotope values from closed forest ecosystems have been termed the ‘canopy effect’. Originally this was ascribed to recycling of depleted carbon from forest floor decomposition of organic material, although others have suggested that it is equally likely to be due to variations in leaf-level processes in response to increased shade. This depletion in the heavier carbon isotope is passed on to woodland herbivores feeding within the forest environments. A similar isotopic depletion has also been reported in the archaeological literature from more open temperate woodland settings, but no measurements have been made on the plants at the base of the food chain in order to quantify the effect. In this study we attempt this by examining the carbon and nitrogen stable isotopic values of different species of grasses from a range of open to closed habitat settings within Wytham Wood, Oxfordshire, UK. We find a strong relationship between carbon isotopic depletion of plant tissue and lowered light intensity with an up to 5& shift between grass grown in open and closed locations. In order to follow this up the food chain, we also report data on wool from sheep grazing in open pastures near the Wood, and on fallow deer living within the woodland, but which turn out not to show a strong canopy effect, probably related to their feeding strategies. We conclude that there is indeed a strong ‘canopy effect’ in temperate woodland, probably related to differential light levels, but that not all apparently woodland-dwelling mammals show such an effect. We also show considerable isotopic variation at the base of the food chain, which should counsel caution when attempting to interpret dietary isotopes using mixing models. Ó 2013 Published by Elsevier Ltd.
Keywords: d13C d15N ‘Canopy effect’ Temperate woodlands Bone collagen Wool
1. Introduction The effect of recycling carbon dioxide on the carbon isotopic composition of leaves in a forest environment was first systematically investigated by Vogel (1978). In a forest in Upper Bavaria, Germany, he noted that leaves of Asperula and Oxalis growing near the ground had a values of d13C ¼ 31.4& and 31.5& respectively, whereas leaves of Larix and Fagus gave values of d13C increasing from at 31.2& for those growing at 2 m to 27.9& at 19 m. He attributed this phenomenon to the presence near the ground of depleted carbon dioxide emanating from the decay of plant material in the leaf litter. Measurements of CO2 collected from beneath an upturned barrel in a forest near Heidelberg gave d13C values of between 15.0& and 22.4& from June to August. This is substantially below the average atmospheric CO2 value of c. 7&, and suggests that the biogenic CO2 might have a value approaching that * Corresponding author. Tel.: þ44 1865 285204; fax: þ44 1865 285220. E-mail addresses: [email protected], [email protected] (M. Pellegrini). 0305-4403/$ e see front matter Ó 2013 Published by Elsevier Ltd. http://dx.doi.org/10.1016/j.jas.2013.03.028
of the humus itself at d13C ¼ 27.0&, since the collected sample was a mixture of air and exhaled CO2. He estimated that 15% of the CO2 incorporated into leaves at a height of 2 m above the ground is from the soil in this particular forest environment. An alternative explanation for this isotopic depletion has been proposed based on the work of Farquhar et al. (1982), who produced an equation to relate the carbon isotopic value of leaf tissue to the ratio of the CO2 concentration in intercellular leaf spaces (Ci) and the atmospheric CO2 concentration. Since Ci is inversely related to light intensity, this model predicts that more shaded locations should produce vegetation with more negative d13C values (van der Merwe and Medina, 1991). In their seminal paper on the canopy effect in the Amazonian forest, van der Merwe and Medina (1991, 251) concluded that ‘many observers have simply assumed that both alternatives may contribute to the canopy effect’. There is therefore a general assumption that carbon isotope depletion occurs in plants from forested environments, the degree of which is in some way proportional to forest density. Since plants are at the base of the food chain, any such depletion is expected to be transferred up the food chain.
M. Bonafini et al. / Journal of Archaeological Science 40 (2013) 3926e3935
The ‘canopy effect’ has now been widely invoked in archaeology to explain the observation of depleted d13C values in some animal bone collagen, attributed to depletion of vegetation d13C values in wooded or partially-wooded environments (e.g., Bocherens et al. 1995; Fizet et al. 1995; Rodière et al. 1996; Cerling and Harris, 1999; Bocherens et al. 1999; Iacumin et al. 2000; Drucker et al. 2000; Stewart and Lister, 2001; Krigbaum, 2003). The most extensive such consideration is that of Drucker et al. (2008), who studied isotopic values (from collagen or hair) in modern large herbivores from the boreal forest-steppe of Siberia, temperate forests of Dourdan (France), and the boreal forests of western Canada. This paper confirmed that ‘the canopy effect observed in plants is passed on to their consumers’, and concluded that the 3& shift in faunal collagen d13C values observed in the LateGlacialeEarly Holocene transition from a number of French sites was related to the differences in habitat. Although such depletion effects have been directly demonstrated in modern plant material from tropical Amazonia (van Der Merwe and Medina, 1989) and in other low latitude regions, they have been less well-studied in higher latitude temperate woodlands. Although convincing, this and other papers have inferred an isotopic shift at the base of the food chain from measurements of faunal collagen values in open compared to forested environments, rather than by direct observation of the flora. The aim of this research is to directly assess the extent of isotopic variation in grass growing at different locations in temperate woodlands (Wytham Woods, near Oxford, England) as a function of time through the growing season, degree of shade, and climatic conditions in general. At the same time we have analyzed biological tissues (bone collagen from fallow deer and wool from sheep) from animals reared in the same area. We deliberately chose to analyze whole grass samples for carbon and nitrogen isotopes (rather than selecting specific components such as cellulose, lignin, etc.), as this is the dietary basis of the grazing animals. The results of this study provide a quantitative measure of the ‘canopy effect’ in the flora at mid to high latitudes, and the extent of isotopic shifts which might be expected in the food chain at these latitudes.
3927
(Fig. 1). The wood covers 415 ha at an altitude from 90 to 165 m A.M.S.L. and is part of the Wytham Estate, which is c. 980 ha in extent. The remaining part of the estate is farmland. The woods are now exceptional in lowland England since they encapsulate the range of both woodland and non-woodland habitats that were common prior to agricultural intensification. Approximately one-third of the area is ancient semi-natural woodland, which was historically managed as coppice with standards. Over the course of the last 100 years coppice management has been largely abandoned. Another third of the area is recent seminatural woodland that has regenerated on arable, pasture or wood pasture sites in the last two hundred years. The remaining woodland area consists of a variety of plantations, some on ancient woodland sites, some on sites which were formerly open. The woods lies on neutral clay soils, ranging from thin, freelydraining rendzinas over Corallian limestone at the higher altitudes to poorly-drained deep clay soils at lower altitudes (Avery, 1980). 3. Materials and methods A series of experiments were carried out to quantify the canopy effect on flora and fauna in temperate conditions: 1) Four native grass species were placed in canopy and open contexts, following a period in an unheated greenhouse for germination. They were harvested at set time intervals during the growing season (March to September), and measured for their carbon and nitrogen isotopic compositions. 2) Bone collagen from fallow deer Dama dama living within Wytham Woods was extracted and measured for its carbon and nitrogen isotopic compositions. 3) Wool from sheep reared in the pasture adjacent to Wytham Woods was collected and measured for its carbon and nitrogen isotopic compositions.
3.1. Measurement of d13C and d15N in four native grass species 2. Wytham woods The research area was Wytham Wood (51.460 N 1.200 W; UK National Grid SP 460 080), which is located to the west of Oxford
The grass seed was bought from Emorsgate Seeds (http:// wildseed.co.uk), a nursery specializing in autochthonous flower seeds and wild herbs from Great Britain, and grown in individual
Fig. 1. Map of Wytham Woods, and location of placement of grass samples.
3928
M. Bonafini et al. / Journal of Archaeological Science 40 (2013) 3926e3935
pots as monocultures. The species used were (descriptions from Hubbard, 1984):
2nd August (3rd sampling) 4th September (4th sampling)
Bromus erectus (Upright Brome, hereafter BE in data tables). A densely tufted perennial, 40e120 cm high. A coarse fibrous grass of well-drained calcareous soils, abundant on chalk and limestone downs, especially in southern England. Lolium perenne (Perennial Rye-Grass, hereafter LP). A loosely to densely tufted perennial, 10e90 cm high. It is a valuable grazing and hay grass, prominent in old pastures and meadows, especially on the rich heavy soils of the lowlands. It has been cultivated in England for about 300 years. Poa nemoralis (Wood Meadow-Grass, hereafter PN). A loosely tufted perennial, 15e90 cm high. It is a rather delicate grass typical of shady places. In most parts of the British Isles it is locally abundant in woods and hedgerows on sandy to heavy soils. Poa trivialis (Rough Meadow-Grass, hereafter PT). A loosely tufted perennial plant, 20e100 cm high. Very common in meadows and pastures of the lowlands, especially on rich moist soils, and it occurs sometimes in partial shade areas.
The samples were taken to the Research Laboratory for Archaeology and the History of Art and freeze-dried at 45 C under vacuum for 48e72 h. After freeze-drying the samples were ground into powder using a Retsch MM 300 grinder at a vibration frequency of 25 Hz for 3e5 min. For each sample 1e2 mg of powder was weighed into tin capsules for isotopic analysis. All samples were measured for d13C and d15N in RLAHA with the SERCON Geo20-20 IRMS mass spectrometer system operating in continuous flow mode coupled to a Carlo Erba elemental analyser. Samples were run against an in-house alanine standard which is regularly checked against international standards (IAEA N1, N2, CH6 and USGS 40, 41), thus the isotopic results of the samples analysed in this study are referable to the international V-PDB and AIR standards for carbon and nitrogen respectively.
The preparation and sowing of these grasses took place on the 11th of April 2011 in an unheated greenhouse at the Wytham Woods field laboratory station run by the Department of Plant Sciences, University of Oxford. Each species was sown in five medium-sized (L60 cm, W25 cm, H6 cm) vessels, all filled with the same soil, in order to have the same conditions for each species and plant. The vessels were watered regularly with fresh tap water every two days for the first month (AprileMay). In early May, the seedlings of 2e3 cm were re-potted into smaller pots (12 cm diameter, height 14 cm). During the transplant the plants were distributed with the same density in each pot. Sixty pots were prepared for each species, giving a total of 240 pots. During the period 11the13th May 2011 the 240 pots were placed in 12 locations (5 pots for every species for each location) around Wytham Woods, distributed as follows (Fig. 1):
The wooded area of Wytham Woods supports an ungulate population that includes roe deer, fallow deer and muntjac. A metal fence (2.5 m in height) now surrounds the woodland, ensuring that the animals live and feed only within the woodlands. The bones of twenty one specimens of fallow deer (D. dama) were analyzed for their collagen carbon and nitrogen isotopes to evaluate the extent of the canopy effect on this species. The bone samples were taken from a collection of 50 antlers and skulls belonging to roe deer, fallow deer and muntjac. These remains came from animals culled by the gamekeepers of Wytham over the past 20 years, to maintain the populations at sustainable levels. Bone chunks (4e5 g each) were taken (by cutting with snippers) from the front plate between the horns of 21 skulls of fallow deer (D. dama) and prepared for collagen extraction. Collagen was extracted from between 3.0 and 6.0 g of bone per sample using a modified Longin method, according to a standard protocol (O’Connell and Hedges, 1999). Any superficial material was removed from the bone by shotblasting. Lipids were removed from the bone samples by soaking them in a 2:1 chloroform (CHCl3)e methanol (CH3OH) solution, first in ultrasonic bath for 30 min and then overnight. The cleaning was repeated until the lipid components were completely eliminated. Samples were then demineralised in 0.5 M HCl at 4 C for several days, changing the solution every other day, rinsed with deionised water, and gelatinised in a pH 3 solution for 48 h at 75 C. The solution was filtered, frozen and freeze-dried. Between 1.0 and 2.0 mg of dried collagen was loaded into a tin capsule for continuous flow combustion and isotopic analysis using an automated Carlo Erba carbon and nitrogen elemental analyser coupled with a continuous flow isotope ratio monitoring mass spectrometer (PDZ Europa Geo 20/20). Results from samples with collagen yield <1% or C:N ratio >3.5 (Ambrose, 1990) should be considered unreliable and are excluded from the analysis. However, none of the samples analysed in this project were outside these parameters.
three in open pasture (locations No. 1e3) two in the open pasture but close to the edge of the woods (location No. 4, 30 m from the edge of the wood, and location No. 5, 10 m from the edge of the wood) two on the border between the pasture and the woods (locations No. 6 and 7) two in the woods but close to the open pasture (location No. 8, 10 m from the pasture, and location No. 9, 30 m from the pasture) three in the deep woods (locations No 10e12) Temperature was continuously recorded at each location at halfhour intervals from midday on 19/5/2011 until 8 am on 12/9/2011 using an iButton Data Logger DS1921. Rainfall was recorded by collection at each location starting on 13 May 2011 until 12 September 2011. Light intensity levels (lumens per square foot) were continuously measured at half-hourly intervals at two locations (location 1 in the open pasture and location 12 in the deep woodland) from 11 August 2011 until 11 September 2011 using a HOBO Pendant Temperature/Light Data Logger. Grass samples were taken from each pot at approximately monthly intervals, by clipping blades to give a mass of approximately 3 g from each pot. The sampling dates were (all in 2011): 9th May (done when the pots were still in the greenhouse, referred to as sampling period 0) 2nd June (referred to as 1st sampling) 27th June (2nd sampling)
3.2. Measurement of d13C and d15N in the bones of fallow deer from Wytham Woods
3.3. Measurement of d13C and d15N in the wool of sheep from the pastures adjacent to Wytham Woods The sheep sampled for wool were reared in the immediate area of Wytham Woods, where there are numerous pastures for sheep and cattle managed by FAI Farms Ltd. It was decided to collect some samples of wool from these sheep, since, in contrast to the ungulates, these sheep should not show the canopy effect. The sampling was performed in a pastured area 250 m north of Wytham
M. Bonafini et al. / Journal of Archaeological Science 40 (2013) 3926e3935
3929
Woods by simply picking up strands of wool from the area where the sheep were grazing. Twenty six samples of wool were collected in the pasture, from sheep that (according to the manager at FAI Farms Ltd.) fed purely by grazing, without any other kind of fodder. The samples of wool were cleaned prior to isotopic analysis using a washing cycle of distilled water and chloroformemethanol in an ultrasonic bath. After cleaning, the samples were freeze-dried and weighed into tin capsules before isotopic analysis, as described above.
(unless noted otherwise) of d13C and d15N for each grass species at each location. The values of d13C and d15N for the ungulate bone collagen (Table 2) and whole wool (Table 3) are the results of a single measurement for %C and %N, molar C/N ratio and values of d13C and d15N. The data were collected over 29 runs of the mass spectrometer from 18/7/2011 to 10/10/2011, and each batch contained (usually) five samples of our in-house alanine standard. As a result, we collected 215 replicates of the standard, which gave the following results:
4. Results
d13 C ¼ 26:88 0:24&; expected value 26:9&
The results are tabulated according to the experiments described above. Table 1 gives the average of a set of five replicates
d15 N ¼ 1:78 0:23&; expected value 1:6&
Table 1 Isotopic variation of d13C and d15N in four native herbaceous grass species according to sampling period and location. Location and species BE 1 BE 2 BE 3 BE 4 BE 5 BE 6 BE 7 BE 8 BE 9 BE 10 BE 11 BE 12 LP 1 LP 2 LP 3 LP 4 LP 5 LP 6 LP 7 LP 8 LP 9 LP 10 LP 11 LP 12 PN 1 PN 2 PN 3 PN 4 PN 5 PN 6 PN 7 PN 8 PN 9 PN 10 PN 11 PN 12 PT 1 PT 2 PT 3 PT 4 PT 5 PT 6 PT 7 PT 8 PT 9 PT 10 PT 11 PT 12
0 sampling (9th May)
1st sampling (2nd Jun)
2nd sampling (27th Jun)
3rd sampling (2nd Aug)
4th sampling (4th Sep)
d13C
d13C
d13C
d13C
d15N
d13C
d15N
28.9 0.1 29.2 0.5 29.7 0.7 28.9 0.4 29.6 0.4 30.0 0.4 34.4 0.2 34.6 0.6 n.d. 32.3 34.7 0.4 35.9 0.4 30.4 0.5 30.1 0.6 31.2 0.4 30.1 0.3 31.4 0.7 30.7 0.3 35.0 0.5 n.d. n.d. n.d. n.d. 34.6 28.3 0.7 28.3 1.0 30.0 0.3 28.3 0.8 28.6 0.5 28.2 0.5 33.7 0.3 34.5 0.3 n.d. n.d. n.d. 35.6 0.3 29.8 0.2 29.8 0.5 30.3 0.3 30.6 0.5 30.1 0.3 29.7 0.4 33.7 0.3 34.7 0.5 n.d. 34.6 0.5 35.54 0.2 35.8 0.4
0.3 0.1 0.3 0.1 0.8 0.8 0.6 0.4 n.d. 0.5 2.1 1.7 0.0 0.4 0.4 0.4 0.8 0.8 0.6 n.d. n.d. n.d. n.d. 0.4 0.7 1.0 0.3 0.2 0.3 1.2 0.5 1.7 n.d. n.d. n.d. 3.3 1.1 0.7 1.0 0.2 1.4 1.0 0.1 0.8 n.d. 1.3 1.8 1.8
29.1 0.3 29.7 0.5 30.0 0.4 29.3 0.1 29.8 0.3 29.4 0.5 33.6 0.2 n.d. n.d. n.d. n.d. n.d. 30.5 0.4 31.3 0.1 31.8 0.4 30.8 0.2 32.2 0.6 31.2 0.6 33.9 0.8 n.d. n.d. n.d. n.d. n.d. 27.1 0.4 28.9 0.8 28.9 0.9 28.3 0.6 29.2 0.7 29.4 0.7 32.8 0.9 34.0 0.1 n.d. n.d. n.d. n.d. 30.4 0.5 31.0 0.3 31.0 1.0 30.6 0.3 30.6 0.5 30.6 0.3 33.2 0.2 34.7 0.1 n.d. n.d. n.d. n.d.
1.16 0.7 1.75 0.8 0.73 1.4 0.97 1.3 0.30 2.0 0.04 0.2 3.41 1.2 n.d. n.d. n.d. n.d. n.d. 1.24 0.9 1.65 0.5 0.84 2.0 1.31 1.8 1.29 1.9 0.42 2.0 1.38 1.8 n.d. n.d. n.d. n.d. n.d. 1.08 1.3 2.20 0.7 0.74 2.2 2.14 1.2 1.38 1.8 0.37 0.3 0.24 2.2 1.75 0.8 n.d. n.d. n.d. n.d. 0.4 0.7 1.8 0.6 2.0 2.1 2.4 1.3 2.0 1.2 0.4 2.3 1.6 0.6 2.9 1.1 n.d. n.d. n.d. n.d.
28.6 28.1 28.7 28.7 28.8 28.2 28.8 28.5 29.0 29.2 29.6 29.2 30.6 30.6 30.9 30.7 31.0 31.3 31.2 31.0 30.9 31.0 31.3 31.4 30.4 30.1 29.8 29.7 29.7 30.3 29.9 15.2 30.2 30.5 30.4 30.6 30.9 30.8 30.8 30.9 30.6 31.1 30.2 31.1 30.7 30.9 30.8 30.5
d15N 0.3 0.3 0.4 0.5 0.4 0.6 0.4 0.3 0.5 0.6 0.3 0.5 0.5 0.4 0.1 0.4 0.2 0.3 0.2 0.3 0.7 0.6 0.3 0.2 0.1 0.5 0.3 0.6 0.4 0.3 0.4 0.2 0.3 0.5 0.1 0.2 0.1 0.3 0.1 0.1 0.2 0.2 0.3 0.3 0.2 0.2 0.4 0.2
1.3 2.0 1.6 0.9 1.5 0.7 0.8 1.0 1.6 0.4 0.8 1.7 1.1 1.7 1.1 0.6 1.0 1.2 1.5 0.9 1.2 0.1 0.4 0.3 3.4 2.2 3.9 3.9 3.9 4.2 4.3 4.7 2.0 1.2 1.4 2.7 0.5 0.5 0.7 0.2 0.2 0.1 0.3 0.1 0.3 1.0 0.4 3.2
0.3 0.3 0.3 0.6 0.6 0.6 0.5 0.5 0.5 1.3 0.9 0.8 0.7 0.9 0.3 0.8 0.5 0.4 0.3 0.8 0.8 1.3 0.7 0.8 0.5 1.5 1.1 0.8 0.8 0.6 0.4 0.7 1.0 0.7 0.8 0.6 0.8 0.9 0.6 0.4 0.5 0.3 0.5 0.4 1.2 0.5 0.8 2.7
27.1 26.9 28.8 27.7 26.4 27.7 31.7 31.4 31.0 30.9 31.1 31.7 27.6 28.0 28.8 29.3 28.3 27.4 32.9 33.7 32.3 33.1 33.1 33.6 27.7 27.6 28.4 27.1 27.0 27.3 33.3 33.9 32.7 32.9 32.8 34.3 27.6 28.7 29.2 28.5 28.5 27.6 33.1 33.4 31.3 32.7 32.2 32.6
d15N 0.5 0.4 0.7 0.6 0.4 0.2 0.6 0.4 0.3 0.5 0.4 0.6 0.3 0.5 0.7 0.6 0.5 0.3 0.2 0.9 0.7 0.7 0.2 0.5 0.5 0.3 1.0 0.5 0.8 0.7 0.1 0.3 0.3 0.7 0.3 0.3 0.7 0.2 0.5 0.7 0.2 0.6 0.6 0.5 0.8 0.7 0.7 0.9
0.1 1.2 1.5 0.6 0.9 0.8 0.2 0.7 2.0 1.3 1.3 0.6 0.9 0.2 1.0 0.3 0.9 0.5 0.2 1.5 0.5 0.6 0.5 0.0 0.3 0.1 0.9 0.3 0.7 0.5 2.2 2.8 2.4 2.1 2.3 2.5 0.3 0.1 0.2 0.2 0.7 0.4 0.3 0.3 0.3 0.3 0.8 0.5
0.8 1.1 0.8 0.6 0.4 0.4 0.6 0.7 0.5 0.6 0.5 0.5 0.6 0.8 0.9 0.6 0.4 0.4 0.5 0.3 0.2 0.3 0.4 0.9 0.8 0.5 1.9 0.5 0.7 0.4 1.2 0.6 0.6 1.0 0.9 0.4 1.0 0.6 0.3 0.8 0.5 0.6 04. 0.4 0.3 0.4 0.6 0.5
27.9 27.7 28.5 28.4 28.5 28.5 32.5 33.5 32.2 33.1 33.7 34.2 28.9 28.4 29.7 29.7 30.3 28.8 32.9 35.3 34.2 33.9 33.9 35.6 26.2 26.3 27.1 27.0 25.8 27.0 33.3 35.0 35.3 34.1 34.6 35.4 28.6 29.0 30.1 29.0 29.2 28.2 32.9 34.7 33.6 33.4 33.7 35.4
d15N 0.3 0.5 0.4 0.4 0.4 0.2 0.6 0.4 1.2 0.7 0.6 0.8 0.7 0.7 0.3 0.7 0.3 0.5 0.3 0.5 0.4 0.8 0.5 0.4 0.8 0.6 0.6 0.6 0.3 0.9 0.4 0.1 0.2 0.7 0.2 0.2 0.8 0.5 0.4 0.5 0.3 0.4 0.3 0.4 0.5 0.5 0.4 0.4
0.2 0.0 0.6 0.3 0.1 1.1 1.3 0.2 1.6 1.0 1.4 0.9 0.7 0.7 0.1 0.2 0.44 1.7 1.4 0.4 1.1 0.1 0.5 0.6 0.1 1.6 1.0 0.6 0.2 1.1 1.0 1.7 3.6 2.8 3.0 3.1 0.6 0.3 0.9 0.4 0.1 0.8 1.3 0.1 0.7 0.5 1.2 1.3
0.8 0.6 1.2 1.0 0.9 0.3 0.9 0.5 0.2 1.4 0.4 0.2 1.0 0.4 0.5 0.2 0.6 0.5 0.3 0.5 0.4 0.7 0.4 0.3 0.7 2.4 1.7 0.8 1.7 0.7 1.0 0.8 0.6 1.2 0.4 0.5 0.9 0.8 1.3 1.2 0.9 0.7 0.5 0.4 0.6 0.3 0.6 0.5
All values quoted are the mean and one standard deviation calculated from five measurements, unless noted as follows. Values in italics are from four measurements. Values in bold italics are from three measurements. Values in bold are from two measurements. Values with no associated standard deviation are from single measurements. n.d. ¼ not determined, usually because no grass was collectable.
0.3 0.9 0.9 0.3 0.9 0.4 1.3 0.6
0.4 0.6 1.0 1.3 0.8 0.4 1.3 0.6 0.5
1.3 1.3 0.8 1.4 0.9 0.3 0.9 0.6
1.0. 1.8 0.6 0.5 1.0 0.8 1.1 0.5 1.2
0.7 0.5 0.3
3930
M. Bonafini et al. / Journal of Archaeological Science 40 (2013) 3926e3935
Sample
Collagen yield (%)
%C
%N
d13C
d15N
C/N
d13C corra
C1 C2 C3 C4 C5 C6 C7 C8 C9 C10 C11 C12 C13 C14 C15 C16 C17 C18 C19 C20 C21
n.d 19.3 28.2 12.9 11.8 11.6 11.3 13.5 10.9 13.5 11.9 28.3 15.4 12.6 14.7 13.0 13.0 14.3 10.7 n.d. n.d.
36.3 33.7 31.6 32.0 37.8 34.5 41.2 40.8 31.4 40.2 34.1 28.9 32.0 35.5 41.7 33.6 33.9 42.2 35.9 41.1 38.6
13.4 12.5 11.7 12.0 13.3 12.9 15.1 14.8 11.6 14.6 12.4 10.7 11.8 12.9 15.2 12.2 12.3 15.3 13.1 14.6 13.8
23.6 23.8 23.5 23.2 24.1 23.6 24.2 24.0 23.8 23.9 24.1 23.8 23.8 24.0 24.1 24.1 20.6 22.7 24.4 24.6 24.2
3.6 5.9 4.9 4.8 4.9 7.8 5.6 5.0 5.0 4.5 5.2 5.0 4.9 5.5 4.7 5.0 7.3 5.6 4.8 6.7 6.9
3.14 3.15 3.15 3.10 3.31 3.12 3.20 3.21 3.15 3.21 3.21 3.15 3.17 3.20 3.20 3.22 3.21 3.21 3.20 3.29 3.26
22.4 22.6 22.3 22 22.9 22.4 23.0 22.8 22.6 22.7 22.9 22.6 22.6 22.8 22.9 22.9 19.4 21.5 23.2 23.4 23.0
a These values of d13C have been corrected for the modern fossil fuel effect by adding 1.2& (Friedli et al. 1986) for comparison with archaeological data.
5. Discussion 5.1. Isotopic variation in four native grass species growing in canopy and open conditions 5.1.1. Inherent variation of d13C and d15N in grasses growing under the same conditions Because all four grass species were initially sown in pots containing the same growing medium, and subjected to the same watering regime in the same environment before they were moved to the specified locations, the data from sampling period 0 (9th May) affords an opportunity to look at the inherent isotopic Table 3 Isotopic measurements on wool from the sheep present in a pasture adjacent to Wytham Woods. Sample
%C
%N
d13C
d15N
C/N
W1 W2 W3 W4 W5 W6 W7 W8 W9 W10 W11 W12 W13 W14 W15 W16 W17 W18 W19 W20 W21 W22 W23 W24 W25 W26
39.6 42.8 48.4 44.8 46.2 45.1 46.9 45.2 44.6 50.9 48.0 50.6 49.4 45.8 47.9 48.0 48.7 49.5 51.0 44.8 44.7 46.1 46.1 44.8 46.3 45.8
13.4 14.3 15.9 15.0 15.3 15.2 15.6 15.0 15.2 17.2 16.2 17.1 16.6 15.7 16.3 16.2 16.5 16.7 17.1 14.9 14.9 15.4 15.5 15.0 15.8 15.2
26.7 26.1 26.8 26.2 26.1 26.2 26.0 26.9 26.9 26.0 26.5 25.6 26.4 25.8 25.8 25.9 25.0 26.0 26.2 25.7 25.8 25.6 25.1 26.0 25.4 26.0
9.0 7.4 8.8 8.3 8.9 9.3 8.8 9.1 8.6 9.3 8.5 8.5 8.8 9.0 8.9 8.4 9.1 8.7 8.8 9.3 9.1 10.0 9.6 9.6 9.3 8.4
3.44 3.49 3.54 3.49 3.52 3.46 3.51 3.51 3.43 3.43 3.46 3.44 3.46 3.42 3.43 3.46 3.44 3.45 3.48 3.50 3.50 3.48 3.46 3.47 3.44 3.52
δ15N AIR (‰)
Table 2 Isotopic measurements on the collagen from fallow deer present in Wytham Woods.
4 3 2 1 0 -1 -2 -3 -4 -5 -6
BE LP PN PT
-32
-31
-30 -29 δ13C V-PDB (‰)
-28
-27
Fig. 2. Isotopic plot of four grass species grown under the same conditions.
variation within these species. Fig. 2 shows an isotopic plot of all values recorded on these four species at sampling time 0 (note that each point plotted is the average of five measurements). The averages and standard deviation of all plants for each species are shown in Table 4. There is clearly significant variation between the species, with a maximum difference between the species average of 2.0& in d13C and 4.1& in d15N. The variation within each species is also significant, ranging from 0.3& to 0.4& in d13C (one standard deviation) and 0.5& to 1.2& in d15N. Since these are from grasses growing under the same conditions, these figures must represent the minimum isotopic variation to be expected at the base of the food chain. Larger variation should be expected if different grass species are growing in the same area. 5.1.2. Comparison of d13C and d15N in grasses grown in open and shade conditions Fig. 3 shows the average isotopic value of d13C in B. erectus at each of the 12 sampling locations for each of the five sampling episodes. The other three species show very similar patterns. From the relatively homogeneous values at sampling period 0 (in the greenhouse), there are two basic patterns of behaviour, one followed by grass at sample locations 1e6, and one for locations 7e12. Fig. 1 shows that locations 1e5 are either in fully open conditions or are more marginal but still in the pasture, and 8e12 are either in deep woodland or more marginal but still in the woodland. Locations 6 and 7 are both on the border, but with different aspects. The isotopic data suggest that location 6 should be included in the open samples, and 7 the shaded. Further evidence for this grouping is presented below from an analysis of the environmental data collected. Fig. 4 shows the average isotopic value of d13C in all species grouped into either locations 1e6 (termed ‘open’) or 7e12 (closed) for each of the five sampling episodes. The values of d13C found for the plants growing in woodland, regardless of the species, are depleted in 13C compared to the normal range of values in C3 plants reported in the literature (van der Merwe and Medina, 1991), and also compared to the equivalent species growing in an open environment. The literature describes the depletion of 13C due to a canopy effect as a reduction between 2 and 5& compared to the values of d13C for the same plant in a non-woodland environment
Table 4 Average values of d13C and d15N for each species of grass at sampling period 0. Species
d13C (&)
BE LP PN PT
28.8 31.0 30.2 30.8
d15N (&) 0.4 0.3 0.3 0.2
1.2 0.9 3.1 0.2
0.5 0.6 1.2 1.1
M. Bonafini et al. / Journal of Archaeological Science 40 (2013) 3926e3935
3931
Table 5 Average differences in d13C and d15N values (&) between open (1e6) and shaded (7e12) locations for each grass species. Species
Fig. 3. Isotopic variation in d13C in Bromus erectus at each sampling location for each sampling date.
(van Der Merwe and Medina, 1989; Broadmeadow and Griffiths, 1993; France, 1996). The differences between the average values of d13C and d15N between grasses placed under the canopy (locations 7e12) compared to those located in open areas (locations 1e6) for the four species are shown in Table 5. Ignoring the sampling on the 9th of May (carried out before the setting out of the seedlings), the overall average of the difference in d13C (4.8&) is towards the maximum of the values discussed in the literature. The maximum difference observed for any one species (P. nemoralis on 27th June) has a value of 8.1&. This high difference is primarily due to the very high (i.e., less negative) average value of the grasses located in the pasture (26.6&), the explanation for which could be the observed production of inflorescences (ears) that occurred in the plants of P. nemoralis placed in the pasture locations during June. To prevent the setting of seeds (more likely in the open settings) influencing the isotopic values in the blades, it was decided to prune the seed heads, and to perform the pruning in every location for all the samples. It is shown in the literature that the production of seeds (ears) may lead to an enrichment in d13C because of the production of greater amounts of protein in the seeds (Tieszen, 1991). This hypothesis could explain the value reported on 27th June for the plants of P. nemoralis grown in the pasture, being further confirmed by the subsequent trend towards more depleted values after pruning (the plants were no longer in the maturation phase).
1st sampling 2nd Jun
2nd sampling 27th Jun
3rd sampling 2nd Aug
4th sampling 4th Sep
Average
Bromus erectus Lolium perenne Poa nemoralis Poa trivialis
d13C
3.9 4.9 5.8 4.2
5.0 5.1 8.1 5.0
5.0 4.2 6.0 4.8
4.0 2.6 4.8 3.3
4.5 4.2 6.2 4.3
Bromus erectus Lolium perenne Poa nemoralis Poa trivialis
d15N
1.8 1.1 2.7 0.4
0.9 1.0 2.4 0.5
0.5 0.5 2.1 1.0
2.7 0.4 0.3 0.9
1.5 0.7 1.7 0.7
The data for d15N variation through the growing season show a less marked patterning than those of d13C, and the divergence between individual locations is greater, but nevertheless some general trends can be observed. The open locations (1e6) show an enrichment of d15N (typically þ2&) in BE, LP and PN immediately following the relocation of the plants from the greenhouse, followed by either a stable value (PN) or a decline (BE, LP) through the rest of the season. d15N in PT in open environments remains approximately constant throughout the season. The closed environments (7e12) remain approximately constant through the season for all species. As shown in Table 5, the average difference in d15N between the open and shaded locations (average value locations 1e6 minus average value locations 7e12) is positive for all species, ranging from þ0.7& to þ1.7&. It is likely that the use of the same compost for each pot in each location has affected the observed differences in the d15N values, which are mainly influenced by the type of soil and the abundances of nitrogen compounds present (Michener and Lajtha, 2007). Nevertheless there is some slight evidence here for a ‘canopy effect’ in the nitrogen values. 5.2. Correlation of climatic parameters with d13C isotopic measurements in grass species 5.2.1. Temperature and Growing-degree days (GDD) In order to correlate the isotopic values measured in the grass with the recorded air temperature at each location, we have transformed the recorded temperature data into Growing-degree days (GDD). This construct is used by farmers and horticulturists to calculate the progress of a plant towards maturity. Under normal conditions, plants grow in response to cumulative daily air temperature, but only if the temperature is above a minimum cut-off for the particular species. GDD for one day is calculated as:
GDD ¼ ðTmax þ Tmin Þ=2 Tcrit
Fig. 4. Average value of d13C in all species grouped into either locations 1e6 (open environment) or 7e12 (closed environment) on each of the five sampling dates.
where Tmax and Tmin are the maximum and minimum recorded daily temperatures and Tcrit is the cut-off temperature for that particular species. The total GDD is accumulated over the growing season. Tcrit for most cases is taken as 10 C, and it is usual to cap the maximum temperature for temperate crops at 30 C (i.e., to record Tmax as 30 C if Tmax is 30 C or above), since plant growth response falls off at higher temperatures. For example, wheat (Triticum aestivum) is said to emerge at 143e178 GDD, and to reach maturity at 1550e1680 GDD (Wittwer, 1995) if the baseline is taken as 10 C. The temperatures recorded at each location (1e12) were inspected to give Tmax and Tmin for each day from 19/5/2011 to 12/9/ 2011, and the cumulative GDD totals calculated at each location on 2nd June, 27th June, 2nd August and 4th September, corresponding to the dates at which the grass was harvested for analysis. Fig. 5
3932
M. Bonafini et al. / Journal of Archaeological Science 40 (2013) 3926e3935
Fig. 5. GDD calculated at each location for each sampling date. Fig. 7. Cumulative rainfall at each location for each sampling date.
shows the accumulated GDD at each location. This shows that the locations can be clustered into two broad groups with similar characteristics: a low-GDD group (the ‘closed’ environments average GDD on 4th September ¼ 396) consisting of locations 7, 8, 9, 10, 11 and 12 (the temperature logger at location 11 failed on 27th July, but the pattern prior to this date is very similar to locations 10 and 12), and a high GDD group (the ‘open’ environments), consisting of locations 1e6, with location 2 being exceptional having a GDD value of 885 compared to the average of the other five of 659 on 4th September. Clearly these groupings are primarily controlled by the degree of open or shade (see location descriptions above), and it does support the suggestion given above that location 6 is best considered to be part of the open group, and location 7 the shaded group. However, a plot of GDD calculated at each of the sampling days against measured d13C values for each species at each sampling (Fig. 6) shows no relationship between GDD and d13C, suggesting that GDD and therefore temperature is not the prime factor in determining the carbon isotopic value in the plant tissue. 5.2.2. Rainfall The plot of cumulative rainfall for each location is similar to the GDD curves shown in Fig. 7, except that in this case location 6 (one of the marginal locations) is not grouped with locations 1e5 (the open locations), as in the GDD figure, but actually lies mostly below the group of closed location values (7e12). The rainfall data shows that location 2 received the most rainfall (690 ml over the measurement period), with locations 1e5 having a total amount greater than 90% of this value. Locations 6e12 had between 63 and 78% of this value. We are not in a position to calculate water use efficiency (WUE) or estimate the soil moisture content at each location but simply plotting cumulative rainfall (approximately at the date of
Fig. 6. d13C plotted against GDD for all species of grass.
sampling) against d13C value by species shows no relationship between d13C and rainfall (not shown). 5.2.3. Light intensity Light intensity was continuously measured at half-hourly intervals for Location 1 (open situation) and Location 12 (closed situation) from 11th August to 11th September. The average ratio of the areal light intensity at location 1 to that at location 12 during daylight hours is 17 (median value of ratio 15), so we can conclude that the light is on average approximately 15 times more intense in the open areas. It has been shown in tropical environments (Zimmerman and Ehleringer, 1990) that the d13C in plant tissue (expressed as D, the carbon isotope discrimination) is related to the daily photon flux, with d13C becoming less negative as the total photon flux increases. Table 5 and Fig. 4 (above) show a clear difference in the isotopic values of the grasses grown in the shady environments compared to the same species in open situations, in accord with these observations. On average, the grasses growing in shady environments are more depleted by between 4.2 and 6.2& in d13C, whilst the d15N values are enriched by between 0.7 and 1.7&. Although we are not in a position to definitively demonstrate this relationship, it does indicate that differences in light intensity have a stronger influence on the value of d13C in the grass tissue than temperature or rainfall, and suggests that further work is necessary to quantify this relationship. 5.3. Values of d13C and d15N in the bones of fallow deer present in Wytham Woods Table 2 shows that the bones analyzed have an average C/N ratio of 3.19 (within the acceptance range of 2.9e3.6) and high levels of carbon and nitrogen in the collagen (averages 36.0% and 13.2% respectively), which defines good preservation for the collagen (DeNiro, 1985). The collagen yields (where measured) range from 10.7 to 28.3% (average 14.8%, where the exception would be c. 20%) which is slightly surprising since these bones are recent samples that should not have undergone any alteration process. The isotopic values are plotted in Fig. 8. The majority have d13C values in a range from 22.5& to 24.5& (overall average of these 23.9 0.4&), with sample C17 (20.6&) showing a slight enrichment. The average value for d13C including C17 is 23.7 0.8&. The majority of the d15N values are clustered between 3.6 and 5.9& (average of these 5.0 0.5&), with four samples having values above 6& (C6 7.8&; C17 7.3&; C20 6.7&; C21 6.9&). The overall average of d15N is 5.4 1.0&. The purpose of analysing collagen from the bone of ungulates was to use these data as indicators for species expected to be living
M. Bonafini et al. / Journal of Archaeological Science 40 (2013) 3926e3935
Fig. 8. Isotopic values of d13C and d15N from fallow deer living within Wytham Woods.
in shaded habitats, and therefore expected to show a ‘canopy effect’. However, the samples available were from fallow deer which are classically a forest edge species that typically feeds in the open on grass, either in open fields, clearings or rides. The fact that these animals are most frequently seen in the woods, or may have been culled there, is because this where they go when disturbed or are not feeding, thereby giving the impression that they are mainly woodland animals. Fallow deer are generalist feeders and may browse forest plants as they move out to grass, and if hungry they will also eat leaves fallen from the canopy. At the time these animals died the population density was very high and there had been very little thinning of the woods and so the ground flora within the woodland would have been shaded out, meaning there was little to eat within the woodland. It is likely therefore that these deer were probably eating out in the open on unshaded plants. The limited isotopic bone collagen data on modern and archaeological fallow deer bone has recently been compiled by Sykes et al. (2011). This shows that only eight recent (Roman) archaeological samples have values for both d13C and d15N e the other archaeological values (Sykes et al. 2011; Table 1) have only d13C measurements, and these authors were not able to locate any modern values. Bocherens et al. (1999) have published data on 10 fallow deer from Scladina Cave, Belgium, dating to c. 100 ka BP, but these are not discussed here. The Roman and modern data are shown in Fig. 9. Because of the well-known fossil fuel effect (the reduction of d13C in atmospheric CO2 values over the last 200 years as a result of the increased burning of fossil fuel), our values on recent deer are not directly comparable with those from archaeological samples. To account for this, we have added 1.2& to the d13C values for the deer from Wytham Woods plotted in Fig. 9, as shown in Table 2 (Friedli et al. 1986).
Fig. 9. Isotopic values from modern and Roman fallow deer.
3933
The Wytham Woods deer are on average slightly more negative in d13C (22.6 0.8& when corrected for the fossil fuel effect) than the Roman fallow deer from Monkton, Kent (average d13C value from eight measurements on possibly three individuals 21.0 0.5&: Sykes et al. 2011, 160), whereas the d15N values are comparable (Monkton average 6.5 1.0&, compared to 5.4 1.0&). The difference in d13C between these two populations is, however, significant (t z 6, n ¼ 27, tcrit z 2). This might suggest that the deer from Wytham Woods are subject to a slight canopy effect, since the Roman samples are believed to come from a managed breeding herd, which implies a deer park (open) environment. Despite the fact that the fallow deer in Wytham Woods might be expected not to show a canopy effect because of their ecology, our data suggest that one might be at least partially present, although uncertainties introduced by the fossil fuel correction might also account for this difference. If this is the case, the explanation for this could be the fence along the border of the woods, with a height of 2.5 m. The fence was installed as a result of numerous complaints from farms in the neighbourhood about the frequent excursions of roe and fallow deer in search of food, thus preventing them from accessing fully open grazing. 5.4. Values of d13C and d15N in the wool of sheep from pasture adjacent to Wytham Woods Fig. 10 shows the values of d13C and d15N in the analyzed wool from the sheep grazing near Wytham Woods. All values are very close to each other, with a range of d13C from 25.5& to 26.5& (average value 26.0 0.5&) and d15N from 8.0& to 9.5& (average value 8.9& 0.5). For comparison, also shown are some unpublished data from the wool of rare breed sheep from southern England (Good, Ditchfield, Pollard, unpublished), which have an average value of d13C ¼ 25.5 0.5& and d15N 7.3& 0.6. These values make a useful comparison, because it is known that these sheep grazed only in open areas. It can be seen that the carbon isotope values overlap completely, whereas the d15N values in the Wytham flock are offset by approximately þ1.5& (presumably because of variations in underlying geology or rainfall). The overlap in d13C is sufficient to confirm that there is no discernible canopy effect in the d13C values of wool from the Wytham sheep. 5.5. Correlation between isotopic values in sheep wool and ungulate collagen with grass values In order to compare the faunal data with the grass values, we have divided the grass data into those relating to open environments (locations 1e6) and those from closed environments (locations 7e12), for all four species and at all sampling periods (except
Fig. 10. Isotopic values of d13C and d15N in wool from sheep living in pasture near Wytham Woods compared to unpublished data from other modern southern English sheep.
3934
M. Bonafini et al. / Journal of Archaeological Science 40 (2013) 3926e3935
variation within each species, ranging from 0.3& to 0.4& in
d13C (one standard deviation) and 0.5& to 1.2& in d15N. These
Fig. 11. Comparison of all isotopic data, showing d13C values for wool and collagen converted to a ‘dietary’ value.
sampling period 0, which relates to the period in the greenhouse). The average values for these two groups are d13Copen ¼ 29.0 1.4&, d15Nopen ¼ 0.1 0.9&, d13Cclosed ¼ 33.6 1.2&, d15Nclosed ¼ 1.1 1.1&. Since the average values for the wool are d13C ¼ 26.0 0.5& and d15N ¼ 8.9 0.5&, the offset D (wool open grass) between these two is therefore D13C ¼ þ3.0& and D15N ¼ þ9&. The offset between average fallow deer collagen (d13C ¼ 23.7 0.8&, d15N ¼ 5.4 1.0&) and closed grass D (collagen closed grass) is D13C ¼ þ5.9& and D15N ¼ þ6.5&. In order to test the hypothesis that the sheep have grown on a diet of grass from the open pasture, and to measure the extent to which the ungulates may have been reared in the closed canopy, we have applied correction values from the literature to convert d13C in the animal tissue values to an underlying diet d13C value. The correction factors used here are: For animal hair: d13C ¼ 3& (Sponheimer et al. 2003; Drucker et al. 2008) For bone collagen: d13C ¼ 5& (van der Merwe, 1989; Drucker et al. 2008) This gives an average dietary d13C value for the sheep of d13Ccor ¼ 29.0 0.5&, and for the ungulates d13Ccor ¼
28.7 0.8&. These values are to be compared with grass growing in an open pasture of d13Copen ¼ 29.0 1.4&, and d13Cclosed ¼ 33.0 1.4 Fig. 11 shows a comparison of all the isotopic data, but with the individual d13C values for wool and collagen converted to a ‘dietary’ value using the factors listed above. The results of d13Ccor of the sheep wool are seen to have the same d13C values as the grass growing in an open environment, but the ungulates’ d13Ccor values are similar to those of the sheep, and more consistent with the d13C values of the grass growing in an open environment than those of the woodland context. For the sheep, this is as expected, but for the fallow deer it supports the observation that these animals feed predominantly on pasture from open environments (as expected from the ecological observations), rather than from the closed environments. 6. Conclusions We have shown that, even for grasses growing in the same environmental conditions (sampling period 0, in the greenhouse), there is significant variation in the isotopic values recorded between grass species (maximum difference between averages of 2.0& in d13C and 4.1& in d15N). There is also significant isotopic
values must represent the minimum isotopic variation to be expected at the base of the food chain, and perhaps call in to question the precision which can be derived from some of the dietary reconstruction models now used, since these generally assume that dietary inputs are ‘point’ isotopic sources. The experiments showed a strong variation in d13C values for grass growing in a closed woodland environment compared to an open environment. This variation of the d13C values in the grass can be attributed to a canopy effect of approximately 5& in d13C present in the temperate woodland. Of the two phenomena postulated to cause a canopy effect (shading vs. respired CO2), we have been able to infer an inverse relationship between d13C in the leaves and the integrated light intensity received by the plants, suggesting that differences of levels of light irradiation are important, but we cannot rule out a respired CO2 effect from current data. We have also shown that the d13C values in the whole wool shed by sheep feeding in open pasture near the measurement site have values consistent with those to be expected from herbivores grazing on grass growing in a fully open environment. The d13C values derived from fallow deer bone collagen, although somewhat more depleted than the limited set of comparative data available, are also fully consistent with those expected from animals grazing in an open habitat. Although conventionally thought of as being woodland animals, which therefore ought to show a canopy effect, when looked at in more detail, they do not. This is consistent with the information we have for the feeding behaviour of these animals. We conclude that the canopy effect is a real effect in temperate woodland, and that the degree of shift in the d13C values is likely to be related to differential shade levels corresponding to the percentage of woodland cover, but we are not yet in a position to fully calibrate the effect. We have also shown that the expectation that fauna living within the woodland will show such an effect needs to be moderated by detailed ecological knowledge. Acknowledgements The authors would like to acknowledge the help of Dr Nick Brown, Department of Plant Sciences, University of Oxford in arranging access to Wytham Woods, and also the assistance of the technical staff at the research facility. We would also like to thank Dr Stephen Ellwood, WildCRU, University of Oxford for his assistance and for providing the deer samples analyzed in this study and two anonymous reviewers for constructive comments. References Ambrose, S.H., 1990. Preparation and characterization of bone and tooth collagen for isotopic analysis. Journal of Archaeological Science 17, 431e451. Avery, B.W., 1980. Soil Classification for England and Wales. In: Soil Survey Technical Monograph No.14. Harpenden. Bocherens, H., Fogel, M.L., Tuross, N., Zeder, M., 1995. Trophic structure and climatic information from isotopic signatures in Pleistocene cave fauna of southern England. Journal of Archaeological Science 22, 327e340. Bocherens, H., Billiou, D., Mariotti, A., Patou-Mathis, M., Otte, M., Bonjean, D., Toussaint, M., 1999. Palaeoenvirnonmental and palaeodietary implications of isotopic biogeochemistry of Last Interglacial Neanderthal and mammal bones in Scladina Cave (Belgium). Journal of Archaeological Science 26, 599e607. Broadmeadow, M.S.J., Griffiths, H., 1993. Carbon isotope discrimination and the coupling of CO2 fluxes within forest canopies. In: Ehleringer, J.R., Hall, A.E., Farquhar, G.D. (Eds.), Stable Isotopes and Plant Carbon-water Relations. Academic Press, San Diego, pp. 109e130. Cerling, T.E., Harris, J.M., 1999. Carbon isotope fractionation between diet and bioapatite in ungulate mammals and implications for ecological and paleoecological studies. Oecologia 120, 347e363. DeNiro, M.J., 1985. Postmortem preservation and alteration of in vivo bone collagen isotope ratios in relation to palaeodietary reconstruction. Nature 317, 806e809. Drucker, D., Bocherens, H., Marriotti, A., 2000. Contribution de la biogéochimie isotopique à l’étude de la paléobiologie des grands mammifères du Pléistocène
M. Bonafini et al. / Journal of Archaeological Science 40 (2013) 3926e3935 supérieur: application aux rennes et aux chevaux magdaléniens et solutréens du Sud-Ouest de la France. Mémoire de la Société Préhistorique Française 28, 13e27. Drucker, D.G., Bridault, A., Hobson, K.A., Szuma, E., Bocherens, H., 2008. Can carbon13 in large herbivores reflect the canopy effect in temperate and boreal ecosystems? Evidence from modern and ancient ungulates. Palaeogeography, Palaeoclimatology, Palaeoecology 268, 69e82. Farquhar, G.D., O’Leary, M.H., Berry, J.A., 1982. On the relationship between carbon isotope discrimination and the intercellular carbon dioxide concentration in leaves. Australian Journal of Plant Physiology 9, 121e137. Fizet, M., Mariotti, A., Bocherens, H., Lange-Badré, B., Vandermeersch, B., Borel, J.P., Bellon, G., 1995. Effect of diet, physiology and climate on carbon and nitrogen isotopes of collagen in a late Pleistocene anthropic palaeoecosytem: Marillac, Charente, France. Journal of Archaeological Science 22, 67e79. France, R., 1996. Carbon isotope ratios in logged and unlogged boreal forests: examination of the potential for determining wildlife habitat use. Environmental Management 20, 249e255. Friedli, H., Lötscher, H., Oeschger, H., Siegenthaler, U., Stauffer, B., 1986. Ice core record of 13C/12C ratio of atmospheric CO2 in the past two centuries. Nature 324, 237e238. Hubbard, C.E., 1984. Grasses. A Guide to Their Structure, Identification, Uses and Distribution in the British Isles, third ed. Penguin Books, Harmondsworth. Iacumin, P., Nikolaev, V., Ramigni, M., 2000. C and N stable isotope measurements on Eurasian fossil mammals, 40 000 to 10 000 years BP: herbivore physiologies and palaeoenvironmental reconstruction. Palaeogeography, Palaeoclimatology, Palaeoecology 163, 33e47. Krigbaum, J., 2003. Neolithic subsistence patterns in northern Borneo reconstructed with stable carbon isotopes of enamel. Journal of Anthropological Archaeology 22, 292e304. Michener, R.H., Lajtha, K., 2007. Stable Isotopes in Ecology and Environmental Science, second ed. Blackwell Publishing, Oxford. O’Connell, T.C., Hedges, R.E.M., 1999. Isotopic comparison of hair and bone: archaeological analyses. Journal of Archaeological Science 26, 661e665.
3935
Rodière, É., Bocherens, H., Angibault, J.-M., Mariotti, A., 1996. Particularités isotopiqes de l’azote chez le chevreuil (Capreolus capreolus L.): implications pour les reconstitutions paléoenvironnementales. Comptes Rendus de l’Academie des Sciences, Paris, Series IIA 323, 179e185. Sponheimer, M., Robinson, T., Ayliffe, L., Passey, B., Roeder, B., Shipley, L., Lopez, E., Cerling, T., Dearing, D., Ehleringer, J., 2003. An experimental study of carbonisotope fractionation between diet, hair and feces of mammalian herbivores. Canadian Journal of Zoology 81, 871e876. Stewart, J.R., Lister, A.M., 2001. Cryptic northern refugia and the origins of the modern biota. Trends in Ecology and Evolution 16, 608e613. Sykes, N.J., Baker, K.H., Carden, R.F., Higham, T.F.G., Hoellzel, A.R., Stevens, R.E., 2011. New evidence for the establishment and management of the European fallow deer (Dama dama dama) in Roman Britain. Journal of Archaeological Science 38, 156e165. Tieszen, L.L., 1991. Natural variations in the carbon isotope values of plants: implications for archaeology, ecology, and paleoecology. Journal of Archaeological Science 18, 227e248. van der Merwe, N.J., 1989. Natural variation in 13C concentration and its effect on environmental reconstruction using 13C/12C ratios in animal bones. In: Price, T.D. (Ed.), The Chemistry of Prehistoric Human Bone. Cambridge University Press, Cambridge, pp. 105e125. van Der Merwe, N.J., Medina, E., 1989. Photosynthesis and 13C/12C ratios in Amazonian rain forests. Geochimica et Cosmochimica Acta 53, 1091e1094. van der Merwe, N.J., Medina, E., 1991. The canopy effect, carbon isotope ratios and foodwebs in Amazonia. Journal of Archaeological Science 18, 249e259. Vogel, J.C., 1978. Recycling of CO2 in a forest environment. Oecologia Plantarum 13, 89e94. Wittwer, S.H., 1995. Food, Climate, and Carbon Dioxide: the Global Environment and World Food Production. Lewis, Boca Raton. Zimmerman, J.K., Ehleringer, J.R., 1990. Carbon isotope ratios are correlated with irradiance levels in the Panamanian orchid Catasetum viridiflavum. Oecologia 83, 247e249.