Stable isotope characteristics across narrow savanna

1 downloads 0 Views 631KB Size Report
mixing model to calculate %C4 and %C3 biomass from the d13C of surface soil ...... crimination and growth of red mangrove (Rhizophora-mangle. L). Aust J Plant ... Cresswell RG (2001) Early human occupation at Devil's Lair, southwestern ...

Oecologia (2005) DOI 10.1007/s00442-005-0105-5

ECOSYSTEM ECOLOGY

Matthew J. Wooller Æ Beverly J. Johnson Andrew Wilkie Æ Marilyn L. Fogel

Stable isotope characteristics across narrow savanna / woodland ecotones in Wolfe Creek Meteorite Crater, Western Australia

Received: 4 November 2004 / Accepted: 14 March 2005  Springer-Verlag 2005

Abstract The stable isotopic composition (d13C) of sediments from lakes are frequently analyzed to reconstruct the proportion of the regional vegetation that used either the C3 or C4 photosynthetic pathways, often without conducting a detailed survey of the current local vegetation. We performed a study on the modern vegetation composition within the Wolfe Creek Meteorite Crater to complement our future paleoecological investigation of the crater. A bull’s-eye pattern exists where C4 grasses dominate an outer ring and salt tolerant species, including shrubs, herbs, chenopods, and halophytic algae, dominate the inner pan of the crater. The ecotone between the inner and outer zones is narrow and occupied by tall (>7 m) Acacia ampliceps, with some C4 grasses in the understory. Along with the highest water table and most saline soils the center of the crater has C3 plants present with the highest d13C and d15N values. The range of d13C and d15N values from the analysis of surface soil organic matter (OM) was much smaller compared with the range of values from plant materials implying that either: (1) the current plant OM has not yet been integrated into the soils, or (2) processes within the soil have acted to homogenize isotopic variability within the crater. The application of a two end member Communicated by Lawrence Flanagan M. J. Wooller (&) Æ M. L. Fogel Geophysical Laboratory, Carnegie Institution of Washington, 5152 Broad Branch Road NW, Washington DC, 20015-1305, USA E-mail: ff[email protected] Tel.: +1-907-4746738 Fax: +1-907-4747979 B. J. Johnson Æ A. Wilkie Department of Geology, Bates College, Lewiston, ME 04240, USA M. J. Wooller Alaska Stable Isotope Facility, Water and Environmental Research Center and School of Fisheries and Ocean Sciences, University of Alaska Fairbanks, Fairbanks, AK 99775, USA

mixing model to calculate %C4 and %C3 biomass from the d13C of surface soil OM was complicated by: (1) the crater containing both a dry habitat with C4 grasses and a central pan with C4 halophytic plants and, (2) the large variation in the d13C of the plants and soil OM. Keywords Carbon Æ Nitrogen Æ C3 Æ C4 Æ Plants

Introduction Changes in environmental conditions, including the concentration of atmospheric CO2, disturbance, temperature, aridity, and salinity, can influence the competitive balance between plants using the C3 and C4 photosynthetic pathways (Ehleringer et al. 1997; Tieszen 1991). C3 and C4 plants possess diagnostic stable carbon isotope ratios (d13C&) (Deines 1980) where C3 plants (most herbs and woody shrubs) are characterized by d13C values averaging 27& (21 to 35&) and C4 plants average 12& (10 to 14&) (Ehleringer 1991). Variations within C3 plants can result from a number of factors including canopy effects and moisture stress either driven by salinity or aridity (O’Leary 1988; Lin and Sternberg 1992). The d13C values of sedimentary organic matter (OM) in tropical and sub-tropical lakes are often used to reconstruct the past regional composition of C3 and C4 plants in lake catchments (e.g., Ficken et al. 2002; Wooller et al. 2003a; Mora et al. 2002). Logistical or financial constraints can sometimes prevent detailed surveys of vegetation being considered for paleoecological research. In these instances regional or global average d13C values for C3 and C4 plants have to be used to interpret sedimentary d13C values. The purpose of this study was to examine the vegetation distribution and d13C variation at a local scale across the Wolfe Creek Meteorite Crater (WCMC), Western Australia in an effort to characterize modern plant inputs into the sedimentary record. The WCMC is one of the sites we are using to

reconstruct the past environments of Australia using multiple lines of evidence including d13C of OM. Considerable variability in d13C of OM from soils has been demonstrated in Northern Australia at the regional and biome level (Bird and Pousai 1997). A detailed local survey of the d13C variation in organic components at a site being considered for paleoecological investigation is valuable so as to critically interpret d13C from sediments. We hypothesized that the stable isotopic composition of the soil OM (SOM) from the WCMC should reflect the modern vegetation composition within the crater, a prerequisite for paleoecological interpretations of d13C analyses of sediments. We hypothesized that the vegetation distribution is related to evapotranspiration, soil moisture, and salinity. This hypothesis originated from our initial observations of the site, which showed that soil in the central part of the WCMC was encrusted with gypsum and was noticeably moist compared with the rest of the soils in the crater. Both soil moisture and salinity are known to influence the d13C of plants (e.g., O’Leary 1988; Lin and Sternberg 1992). Quadrat analyses were performed to quantify the percent C3 and C4 plant cover. The d13C of plants and soils from within the WCMC were determined to examine whether the local C3 and C4 plant distribution was represented in the surface soils. The d15N values of plants and soils were also determined to investigate the magnitude of variation in the vegetation, which can be an effective marker of the nutrient status in a habitat that can subsequently influence vegetation composition, distribution, and structure (e.g., Evans 2001; Robinson 2001; McKee et al. 2002; Wooller et al. 2003a, b; Jones et al. 2004). Soil-moisture and salinity of the groundwater, which are also known to influence the d13C of plants (e.g., Lin and Sternberg 1992; Tieszen 1991; Archibold 1995) were also determined. These findings will subsequently aid interpretation of d13C measurements of total organic carbon (TOC) preserved in a 10 m auger taken from the center of the WCMC (Wooller et al., in preparation).

Materials and methods Study site The WCMC (longitude 1910¢E and latitude 12747¢S), located south of Halls creek off of the Great Northern Highway between the Tanami Desert to the east and the Great Sandy Desert to the west, was discovered and named by Reeves and party in 1947 (Reeves and Chalmers 1948). In 1948, the Australian Bureau of Mineral Resources (BMR) carried out the first detailed geological investigation of the feature (Guppy and Matheson 1950) and attributed it to a meteoric impact, which was subsequently confirmed by others (Cassidy 1954; LaPaz 1954; McCall 1965). Ratios of 36Cl/10Be and 41Ca/36Cl from an iron meteorite fragment

from the crater give an age of 300 ka for the crater (Shoemaker et al. 1990). The WCMC is one of the largest and best-preserved impact craters on Earth and is now a National Reserve managed by the Western Australian Department of Conservation and Land Management (CALM). The crater has a mean diameter of approximately 880 m and a rim height of about 30 m above the surrounding plain. At the base of the steep inner walls, there is an abrupt change of slope to a gently sloping sand apron about 280-m wide that is primarily covered in grasses. The apron falls about 6 m to an extremely flat inner floor primarily covered with trees and shrubs. The crater forms a topographically closed depression, but because of its very small catchment area and its position in the arid zone it cannot form a surface-water-fed lake (Bowler 1981, 1986) and the crater hydrology is dominated by groundwater. Subsequently, there is no yearround lake and standing water in the crater consists of only a few isolated pools (Fig. 1). The climate of the area is strongly influenced by the Australian monsoon during the summer months (December through March) and penetration of the Inter Tropical Convergence Zone (ITCZ) from the north, which brings 500 mm of precipitation annually. Evaporative water loss from the site is highest during the months of September through January (e.g., average December evaporation=300–350 mm) with mean annual evaporation 3,400 mm (Commonwealth Bureau of Meteorology 2003), which is evident from a concentration of gypsum forming a crust on the surface of the soil towards the center of the crater. In July 2001, the water table was within 50 cm of the soil surface in the center of the crater, and consequently, freestanding water was present in isolated pools and dissolution pits towards the center of the crater during the driest of months. Vegetation survey Five areas were selected at the WCMC area in July 2001: (1) the area immediately adjacent to and outside of the crater rim, (2) the area along the steep banks leading into the crater, (3) the basal plain in the crater, (4) an ecotone between woodland and grassland in the crater and, (5) the central pan. We selected these areas because they were consistent with obvious topographic or vegetation changes and we wished to examine whether these changes coincided with changes in the stable isotope composition of the plants and soils in the crater. Nine survey stations were set up along a W–E transect (Fig. 1) through the center of the crater in each of the five zones. At each survey station, three 6·6-m plots were set up perpendicular to the transect 10 m apart. Plants were identified and the plant distributions (% cover) were estimated visually as a percentage of the total plant cover within each plant plot. The plants sampled in the crater were Acacia

Fig. 1 a A schematic top view of the crater; b the change in vegetation structure across the crater

ampliceps Maslin (the salt-water wattle), Aristida latifolia Domin, Atriplex sp., Crotalaria sp. Cyperaceae, Dolichandrone heterophylla (R.Br.) F.Muell., Eragrostis setifolia Nees, Eucalyptus brevifolia F.Muell., Flaveria australasica Hook., Hakea sp., Newcastelia sp., Melaleuca lasiandra F.Muelland, Ptilotus exaltatus Nees, Trichodesma zeylanicum R.Br. and Triodia pungens R. Br. One leaf of each of the species (the nearest to the sampling point) present at each site was collected. The %C3 and %C4 cover for each was determined by calculating the total percentage of shrubs, trees and herbs present (C3 vegetation) relative to the total percent of all grasses, Atriplex sp., Flavaria australasica, and sedges present (C4 vegetation). The %C3 and %C4 plant cover for each survey station was determined by averaging the data from the three plots. A soil sample from one of each of these sets of three quadrats was collected for stable isotope analyses of the SOM.

Field collection and preparation of samples from the WCMC Fully expanded, canopy leaves were removed from plants at 50 stations located along the transect within WCMC. The location of each station was recorded using a Global Positioning System (GPS, Garmin 12 XL) and the data are reported relative to a GPS datum point towards the center of the crater (longitude 1910.310¢E and latitude 12747.783¢S). These sites were independent of the sites surveyed using quadrats (described above). A topographic profile across the crater was carried out using a Leica TC 1010 (Certified) total station. The leaves collected were dried in a plant press in the field and were then freeze-dried in the laboratory. Approximately 6 cc of the soil surface (0–2 cm, excluding litter) were collected from each of the 50 sites. These were sealed airtight in screw-top Eppendorf tubes. The soil samples were transported to the laboratory, where the

percentage soil moisture was recorded by subtracting the dry weight (freeze-dried) of the samples of the surface soil from the wet weight. A sub-sample of each soil sample was taken and prepared for stable isotope analyses by reaction with concentrated HCl in an airtight container to remove any carbonates (Harris et al. 2001) A sub-sample of the soil samples was used for salinity and nitrate analyses. Each sub-sample was packed into a 15-ml conical tube to the 3 ml level. A total of 10 ml of distilled water was added to the 13 ml mark and the sample was sealed and shaken vigorously. Each sample was then allowed to settle for 30 min. The liquid from the conical tubes was removed using a pipette and transferred to clean 15 ml conical tubes, which were then sealed and transported to the laboratory. The sediment was discarded and the retained liquid was filtered through a pre-combusted glass fiber filter. The filtrate was tested for salinity using a refractometer (Vista A366ATC). The nitrate concentration of the extracts was measured using a LaMotte colorimetric test kit. Measurements of the nitrate standards and samples were conducted on a Schimadzu UV160U UV-visible recording spectrophotometer. Analyses of nitrate solutions of standard concentration were used to produce a calibration curve with an r2 of 0.99. A water sample was also taken from two of the pools within the central area of the crater (Fig. 1) and tested for salinity. Samples of an algal mat covering the surface of the pools in the WCMC were taken and freeze-dried for stable isotope analyses. The positions of dead Acacia sp. and Eucalyptus sp. trees within the crater were recorded using the GPS. These were recorded so as to mark the recent past extent and distribution of these two C3 tree species. Carbon and nitrogen stable isotope analyses An aliquot (0.700 to 0.800 mg of plants and 12.0– 16.0 mg of soil samples) of each freeze-dried sample collected from the WCMC was weighed into a tin capsule, which was crimped and introduced [via the EA carousel; (Wooller et al. 2001)] into the autosampler (A2100) of a CE Instruments, NA 2500 series, elemental analyzer (EA) at the Geophysical Laboratory (GL). Purified combustion gases (CO2 and N2) were separated in a gas chromatographic column prior to entering a Finnigan Conflo II interface and the stable isotope ratio mass spectrometry (Finnigan MAT, Deltaplus XL). The results are presented in standard delta notation. Acetanalide (C8H9NO) was analyzed (every tenth sample) as a check on the analytical precision (and accuracy) throughout the analyses, which was ±0.2& for d15N (N%= ±0.5) and was ±0.27& for d13C (C%= ±5.4) (n=26). Mixing model and statistical analyses A two-end member mixing model, typical of the type used in paleoecological studies, to calculate the

proportion of C3 and C4 plants composing past vegetation, was used to predict the d13C of SOM as follows: d13 CSOM ¼ ð1  X Þd13 C‰C3 þ X ðd13 C‰C4 Þ where d13C C3=26.5& and d13C C4=11.5&, for the average values of C3 and C4 plants from the crater, respectively, X=fraction of SOM derived from C4 plants, and (1X) = fraction of SOM derived from C3 plants. These predicted values were subsequently compared with the measured d13CSOM. Use of this model relies on the assumption that the modern biomass is incorporated directly into the soil sediment without undergoing isotopic fractionation. Where necessary a Student’s t test was used to determine differences in the stable isotope composition of components (e.g., plants and soils) from different areas of the crater (e.g., pan compared with basal plain). Error bars for measured d13CSOM are derived from the d13C of soil for each vegetation area (Fig. 5a). Error bars for predicted d13CSOM values are derived from the measurement of %C3 and %C4 of the vegetation in three quadrats placed in each vegetation area (Fig. 1a).

Results Vegetation distribution and stable isotope composition of plants The Cyperaceae and herbaceous plants Atriplex sp. and Flaveria australasica were classified as C4 (Table 1). All of the tree species measured from the crater, Acacia ampliceps, Crotalaria sp., Dolichandrone heterophylla, Eucalyptus brevifolia, Hakea sp., Melaleuca lasiandra, Trichodesma zeylanicum, were classified as C3, as were the herbs Newcastelia sp. and Ptilotus exaltatus. The least negative d13Cplant values (9.6&) were determined in the Cyperaceae samples (Table 1 and Fig. 4) found in the central region of the crater (Fig. 2). The Cyperaceae also had the most variable C/N values, which may have been due to the sedge samples being senescent at the time of collection. The algal mat collected from one of the pools had relatively less negative d13C values, 14.6&. The grasses Aristida latifolia, Eragrostis setifolia, and Triodia pungens from the WCMC were all classified as C4 according to the d13C values of their leaves (Table 1). The vegetation in the central (wooded) area of the crater was dominated by the C3 tree Acacia ampliceps, which varied in height from 7 m at the grassland/ woodland ecotone to 2 m at the crater’s center (Fig. 2a). A number of dead Acacia sp. and Eucalyptus sp. trees, ranging from 10 cm to 40 cm in diameter at breast height (dbh), were standing within the wooded area of the crater. The spatial distribution of the dead Eucalyptus sp. trunks seemed to form a circular pattern within the crater; whereas, the dead Acacia sp. trees seemed to be concentrated around the pools in the crater (Fig. 1).

Table 1 Biogeochemical measurements of C3 and C4 plants from the WCMC Species (sample size)

d15N

d15N 1 SD

d13C

d13C 1 SD

%C

%C 1 SD

%N

%N 1 SD

C/N

C/N 1 SD

C3–C4*

Acacia ampliceps (n=38) Aristida latifolia Atriplex sp. (n=7) Crotalaria sp. (n=3) Cyperaceae (n=3) Dolichandrone heterophylla (n=12) Eragrostis setifolia Eucalyptus brevifolia (n=6) Flaveria australasica (n=9) Hakea sp. (n=4) Newcastelia sp. (n=4) Melaleuca lasiandra (n=15) Ptilotus exaltatus (n=5) Trichodesma zeylanicum (n=7) Triodia pungens (n=21) Algae

4.27 0.9 16.25 0.33 9.79 1.55 8.23 0.46 6.83 4.89 0.05 0.69 6.68 1.05 1.28 6.50

3.99

25.86 11.62 11.03 26.03 9.60 26.19 12.08 26.83 12.98 27.10 26.57 27.64 26.09 26.13 11.80 14.59

1.11

33.6 20.8 31.7 42.0 31.7 38.8 37.8 45.1 26.6 36.5 28.4 44.2 29.5 35.4 38.3 7.03

4.1

2.2 1.0 2.5 2.7 1.5 1.1 1.5 1.7 1.6 0.6 0.8 0.9 1.6 1.8 0.8 0.4

0.5

18.4 24.7 22.8 18.4 44.4 44.7 28.7 35.0 20.5 70.0 40.0 61.2 23.1 24.0 66.0 17.9

4.2

C3 C4 C4 C3 C4 C3 C4 C3 C4 C3 C3 C3 C3 C3 C4

4.23 1.14 3.67 1.81 0.85 3.59 1.13 1.01 1.89 3.38 1.19 1.50

0.95 1.25 0.74 1.01 1.55 0.71 1.36 1.53 1.07 2.41 1.04 0.68

3.4 1.1 2.7 5.9 3.9 3.0 9.6 1.4 7.7 11.7 1.1 7.7

1.1 0.4 1.2 0.3 0.7 0.2 0.1 0.1 0.3 0.3 0.4 0.4

2.1 2.1 40.8 10.7 15.6 4.0 20.1 5.5 19.6 11.2 4.4 27.1

*C3 plants (most herbs and woody shrubs) are characterized by d13C values averaging 27& (21 to 35&) and C4 plants average 12& (10 to 14&) (Ehleringer 1991) Fig. 2 a Canopy height of Acacia ampliceps within the WCMC; b the C3and C4 species distribution across the floor of the crater

The highest d13C value for C3 plants was 23& (Acacia ampliceps), which was measured in the central part of the crater (Fig. 3). C3 plants in the region where

Acacia ampliceps is shorter (