Soil selenium in a forested seabird colony: distribution, sources, uptake by plants, and comparison with non-seabird sites.
Selenium (Se) is a trace element with an unusually narrow boundary between deficiency and toxicity (Fordyce 2007). It is often deficient in soils, but can be found at toxic levels in some parts of the world (Hartikainen 2005). The global mean soil Se concentration is 0.4mg[kg.sup.-1] (Fordyce 2007), but concentrations in plants are much lower than the surrounding soil (Watkinson 1962; Wang and Gao 2001). In mammalian herbivores, Se deficiencies are common at herbage levels <0.05 mg[kg.sup.-1] (Allaway and Hodgson 1964) but toxic effects occur with sustained consumption >1 mg[kg.sup.-1] (Hartikainen 2005).
Soil solution Se is divided between Se(IV) (selenite) and the more mobile Se(VI) (selenate), with the likelihood that soil solution Se is maintained by release from soil organic matter (Wang et al. 2012). Inter-conversion between selenite and selenate is controlled by both kinetic and thermodynamic factors (Stolz et al. 2006). Selenate is more thermodynamically stable under typical agricultural conditions of soil pH and redox potential, but selenite is the more stable in poorly drained acid soils (Elrashidi et al. 1987; Seby et al. 2001). Selenate is more readily taken up by roots and translocated to leaves than selenite, but the selenite that reaches leaves is more biologically active (Hopper and Parker 1999). Divalent metal ions (especially calcium) bind selenite strongly (Torres et al. 2010), thus potentially affecting sorptive uptake. Selenate uptake is inhibited by sulfate, whereas selenite uptake is inhibited by phosphate; the phosphate effect on selenite uptake is less severe than the sulfate effect on selenate (Hopper and Parker 1999). Because of the complexity of inter-conversions between different pools of soil Se (Wang et al. 2012) and the interactions occurring in soil solution (Hopper and Parker 1999), defining the boundary between Se deficiency and toxicity in terms of total soil concentrations is problematic (Mincher et al. 2007).
Although volcanism can be locally important (Floor and Roman-Ross 2012), investigations of environmental Se usually emphasise either the soil parent material (Wang and Gao 2001; Fordyce 2007) or human activities. The latter include coal combustion (Ranville et al. 2010), fly ash disposal (Goldberg et al. 2008), and smelting of sulfide ores (Perkins 2011). However, biological agents may also play a role. Supplementing a study in Antarctica (Sun et al. 2000), Xu et al. (2011) showed that Se concentrations on tropical coral islands were dominated by guano input from breeding or roosting seabirds. However, little is known of the Se distribution within seabird-affected soils or its transfer to plants. The proposition that seabirds on land constitute an important Se supply mechanism also deserves wider testing, nutrient-poor coral islands and the cold of Antarctica having quite different hydrological and nutrient cycling regimes from temperate ecosystems.
Seabirds coming ashore to breed or roost are vulnerable to efficient mammalian predators such as rats, cats, mice, and pigs (Towns et al. 2011), so most large landmasses have never been influenced by seabird-derived marine nutrient and trace element subsidies. Exceptions include large islands such as Hawai'i and mainland New Zealand, where the pre-human absence of mammalian predators allowed a widespread distribution of breeding seabirds (Steadman 1995), often many kilometres inland (Worthy and Holdaway 1996). Following mammalian predator introduction alongside successive waves of human arrivals from Polynesia and Europe, the link to the sea (and its nutrients) provided by seabirds largely vanished to leave only a very few remnant seabird breeding sites. One of these sites in mainland New Zealand is the main focus of the research reported here. We tested the hypothesis that soil Se concentrations in seabird colony soil exceed those that can be explained by weathering of parent material, and that the presence of seabirds is a reasonable explanation. Selenium cycling was explored using correlations with other seabird-related elements (nitrogen, N; phosphorus, P), elements related to parent material (iron, Fe; aluminium, Al; zirconium, Zr), and quantities associated with soil processes (organic content as loss on ignition (LOI); [sup.15]N/[sup.14]N, expressed as [[delta].sup.15]N; pH). We also compared Se concentrations with a former seabird site occupied into human times, and with a site from which seabirds have been absent throughout the Holocene. The expectation with these comparisons was that Se concentrations would be highest at the present-day site and lowest at the non-seabird site. Finally, Se concentrations in two plant species from the Westland petrel site were compared with literature values to see if plants growing in seabird colony soil have toxic levels of Se in their foliage.
Material and methods
The three study sites are all in the New Zealand South Island (Fig. 1); the present-day breeding site hosting Westland petrels (Procellaria westlandica) at 42[degrees]08.8'S, 171[degrees]20.5'E, near Punakaiki (West Coast); the former seabird breeding site near Akaroa (43[degrees]49'S, 173[degrees]0.5'E; Banks Peninsula), 200km southeast; and the site with no evidence for seabird breeding throughout the Holocene, on Takaka Hill (41[degrees]01.1'S, 172054.08/E; Nelson), 180 km north-east. Rainfall at all three sites is ~2000 mm (Akaroa 1800 mm, Takaka Hill 2100 mm, Punakaiki, 2500 mm). The soil parent materials are blue-grey muddy sandstone (Punakaiki), basalt and quartzofeldspathic loess (Akaroa), and marble (Takaka Hill). Soil chemical characteristics are listed in Table 1.
The Westland petrel site has been occupied by seabirds since at least the mid-18th Century (Hawke 2004; Holdaway et al. 2007). Vegetation is dominated by tree ferns (Cyathea spp.) and kamahi (Weinmannia racemosa), with scattered podocarps (especially matai, Prumnopitys taxifolia, and rimu, Dacrydium cupressinum). Nikau (Rhopalostylis sapida, an endemic palm) dominates vegetation towards the valley floor, where petrel burrows are comparatively sparse. Depth distributions of caesium-137 and lead-210 (Hawke 2010) suggest that burrows are longstanding, decades-century-timescale structures. The burrows host a specialised invertebrate fauna for guano processing alongside typical forest soil taxa (Hawke et al. 2012).
The former seabird breeding site at Akaroa was near Purple Peak/Otepatatu-Stony Bay Peak/Taraterehu on the eroded slopes of a Tertiary age volcano. It was identified as a pre-European seabird breeding area from Maori tradition (Andersen 1927; p. 145); seabirds (principally mottled petrels, Pterodroma inexpectata) bred elsewhere on Banks Peninsula well into the 20th Century (Stead 1927). The area is reserved for re-establishment of indigenous vegetation (Stony Bay Peak profile) or used for pastoral agriculture. Fertiliser had never been applied at the time of sampling but application has since begun in the agricultural portion. Previous studies have shown anomalous soil cadmium concentrations attributable to former seabird breeding (Hawke 2003) but a spatial distribution of soil [[delta].sup.15]N dominated by agricultural activity (Hawke 2001).
The control site (no evidence for Holocene seabird breeding) was adjacent to Predator Cave, near Ngama on Takaka Hill. Predator Cave is a rich predator site accumulated by laughing owls (Sceloglaux albifacies). Worthy and Holdaway (1994) found a diverse terrestrial fauna accumulated since the end of the Otiran glaciation into the 19th Century, but no seabirds. In contrast, other laughing owl sites contain a variety of seabirds and have been used to imply seabird breeding nearby (Worthy and Holdaway 1996).
Soil and vegetation sampling
Soil sampling at Punakaiki was in January 2003 and July 2010 at 120-140m altitude on a steep, south-south-east-facing slope. The January 2003 samples involved re-analysis for Se of the five depth profiles collected for a soil P study (Hawke 2005). The depth profiles lay 15 m apart on a transect through the middle of an aggregation of burrows, with each profile excavated to 60 cm depth or to lithic contact. In July 2010, a single, surface soil sample (1-2 cm depth, ~0.5 L) came from each of 14 burrows and five forest floor transect points within a strip 50 by 10m running downslope in the same section of colony as the depth profiles. Burrow samples were collected from 30cm inside the burrow entrance. Six petrel guano deposits were also collected, by scraping from soil litter into a plastic jar. Vegetation sampling was carried out in April 2006, with 10 foliage samples (50-120 g wet weight) collected for each species from a transect running ~75 m downslope from 5 m below the ridgeline (tree fern) or upslope from the valley floor (nikau).
Soil sampling at Akaroa was in September 2000, and May and September 2002. Profiles to 50 cm or lithic contact were collected along a transect on a west-facing side-slope above Akaroa Harbour between Purple Peak and Stony Bay Peak at 490, 530, and 560m, on the ridgeline at 610m, and on the summit of Stony Bay Peak at 806 m. Sampling at the control site on Takaka Hill was in October 2002. Three depth profiles to lithic contact 20 m apart came from a transect across a gentle side-slope leading to a doline, ~100 m from the entrance to the Predator Cave predator site.
Sieved soil (2 mm) was dried (50[degrees]C) and stored dark (20[degrees]C) in polyethylene bags. The pooled guano sample was dried and lightly crushed with a pestle and mortar, and fragments of vegetation and soil were removed by hand. Plant material was dried (50[degrees]C) and ground (1 mm).
Soil, soil parent material, and seabird guano analysis for total recoverable Se used nitric acid/hydrochloric acid digests (US EPA Method 200.2) of finely ground material. Plant material was extracted with tetramethyl ammonium hydroxide (90[degrees]C for 1 h, Fecher et al. 1998; nikau) or microwave-assisted nitric acid/ hydrogen peroxide (CEM MARS 5 Express, Araujo et al. 2002; tree fern). Analysis of soil and plant digests for Se used inductively coupled plasma-mass spectrometry (ICP-MS; Perkin Elmer ELAN DRC Plus, Perkin Elmer ELAN DRC II, or Agilent 7700), the particular instrument used depending on the analysis batch. Analysis of soil for major oxides, Zr, and P used wavelength dispersive X-ray fluorescence (XRF; Siemens 303 AS). LOI was determined at 1000[degrees]C. Total N and [sup.15]N/[sup.14]N were measured on finely ground material using a Europa Geo 20/ 20 continuous flow isotope ratio mass spectrometer. Isotope ratios were calculated as a per mil (%0) deviation from atmospheric N2 and reported as [[delta].sup.15]N. Soil pH was measured on 2:1 water extracts. Well-reputed commercial laboratories were used for all analyses except soil pH: Se, RJ Hill Laboratories Ltd (Hamilton, New Zealand); major oxides, Zr, P, and LO1, Spectrachem Analytical CRL Energy Ltd (Lower Hutt, New Zealand); total N and [[delta].sup15]N, GNS Science Ltd (Lower Hutt, New Zealand).
Laboratory quality control checks for Se in soil, parent material, and seabird guano used the certified reference material (CRM) AGAL-10 (riverine sediment, Australian National Measurement Institute; Se content l l.5mg[kg.sup.-1]). Quality control checks for Se in plant foliage used the apple leaf CRM NIST 1515 (Se content 0.050mg[kg.sup.-1]). Soil major component and LOI data all summed to between 98.2% and 99.0%. The standard deviation of repeated analyses of ammonium sulfate (IAEA N-l) for [[delta].sup.15]N was 0.3%0.
Calculation of parent material contribution and enrichment ratio
Estimating the parent material contribution to soil trace element composition usually uses an immobile reference element such as Zr (Hodson 2002). External contributions are then calculated as the difference between the parent material contribution and the observed trace element content (Se in this case). However, soil formation from more than a single parent material makes this approach problematic, especially if the parent materials differ in their susceptibility to weathering. Although soil at the present-day seabird site near Punakaiki formed from a single parent material, soil at the former site near Akaroa formed from both basalt and loess. Fortuitously, basalt and locally sampled loess have identical Se contents (0.8mg[kg.sup.-1]; loess data from Wells 1967), although their Zr contents differ markedly (basalt 292mg[kg.sup.-1], loess 585mg[kg.sup.-1]; D. J. Hawke, unpubl, data). We therefore calculated the parent material contribution (and hence external Se) using LOI to correct for organic content. Because of volume loss during weathering, calculations based on LOI are likely to underestimate parent material contributions. The discrepancy increases with soil age, but conversely estimates based on immobile elements will typically overestimate parent material contribution. For the Punakaiki data, we were able to compare estimates from the Zr content with estimates from LOI; the result of the linear regression analysis ([+ or -]95% Cl) was: external Se (LO1)=0.95 [+ or -] 0.12 external Se (Zr) - 0.06 [+ or -] 0.19 ([r.sup.2]-0.913). While this indicates a close similarity of the two approaches, soil at this site is less than 1000 years old (Hawke 2004).
To put the external Se contributions into a wider context, we calculated enrichment ratios (ER) for the present-day and former breeding-site soil profile data using the equation:
ER = R(sample)/R (crustal mean) (1)
where R is Se concentration/reference element concentration (Zr in this case). Crustal means for Se (0.05 mg[kg.sup.-l]) and Zr (170 mg [kg.sup.-1]) came from Yaroshevsky (2006). The results were classified according to the scale published by Suthedand (2000), with ER values ' <2 representing sites with no or minimal pollution; 2-5 representing sites moderately polluted; 5-20 representing sites with significant pollution, 20-40 representing sites with very strong pollution, and >40 representing those sites that are extremely polluted'. In the current study, 'polluted' is interpreted in a non-pejorative sense, but the results allow comparison with other sites subject to anthropogenic or volcanic contamination.
For surface soil, one-way ANOVA (Kruskal-Wallis test) was used to compare present-day forest floor, petrel burrow, former colony, and control sites; post-hoc pair-wise comparisons used a Tukey test. Pearson correlations and linear regression were used to explore the link between soil Se and potential predictors, involving possible soil binding agents (total Al, total Fe, LOI) or other likely seabird elements and indices (total N, total P, external P, [[delta].sup.15]N, pH). Nikau and tree fern foliage Se concentrations were compared using a Mann-Whitney U-test. SPSS 20 (IBM Corporation, Armonk, NY) was used for all tests, at a significance level of [alpha] = 0.05.
Surface-soil Se concentrations across the three sites were 0.2-4.2mg[kg.sup.-1], encompassing the range of literature values for New Zealand soils (Watkinson 1962; Fig. 2). Highest values came from within burrows at Punakaiki; the control site (no evidence for seabird breeding during the Holocene) recorded the lowest values. Based on a one-way ANOVA, the null hypothesis that the four categories of surface soil came from the same distribution was rejected (Kruskal-Wallis statistic = 16.696, P = 0.001). Pair-wise post-hoc comparisons showed significant differences between petrel burrow soil and the control site (P = 0.005), and petrel burrow soil and the former colony site (P=0.019). The comparison between the control site and the present-day colony site was on the borderline of significance (P = 0.054). All other comparisons were not significant, most notably the present-day and former colony sites (P= 0.232).
The depth profile concentrations were in the same sequence as surface soil, the present-day site being greatest and the control site concentrations being the lowest (Fig. 3a). One of the profiles from the former seabird site (obtained from gently sloping ground on the summit plateau of Stony Bay Peak) overlapped completely with data from the present-day seabird site. Soil profiles from each site category showed a slight mid-depth maximum in Se concentration, but variability between profiles was greater at the present-day seabird site (Fig. 3a).
Parent-material Se concentrations were all low, being 0.8 mg[kg.sup.-1] (Punakaiki, present-day site), 0.8 mg[kg.sup.-1] (basalt, Akaroa, former site), and <0.5mg[kg.sup.-1] (Takaka Hill, control site). Guano Se content was 3.6 mg[kg.sup.-1]. Enrichment factors at the present-day and former sites were 13.0 [+ or -] 2.4 and 14.4 [+ or -] 4.7, respectively (mean [+ or -] s.d.), corresponding to 'significantly polluted' on the scale proposed by Sutherland (2000). External Se (the concentration of soil Se not derived from parent material) in profiles at the present-day seabird site averaged 1.6 [+ or -] 0.6 mg [kg.sup.-1] (mean [+ or -] s.d.), corresponding to 64 [+ or -] 9% of total soil Se. At the former seabird site, external Se increased from at, or near, zero at the surface to ~0.4 mg [kg.sup.-l] over the depth range 10-35cm, before decreasing again (Fig. 3b). As observed in the total Se profiles, the Stony Bay Peak profile overlapped completely with data from the presentday seabird site. All external Se depth profiles, regardless of site and history, showed a mid-depth maximum similar to that observed for total Se (Fig. 3b).
Correlations between total Se or external Se and other soil quantities (Table 2) were all insignificant, except for pH ([r.sup.2]=0.314 against log(external Se); slope [+ or -] 95% CI, -0.21 [+ or -] 0.15 based on pH as the x-axis) and [[delta].sup.15]N ([r.sup.2] = 0.164). Although [[delta].sup.15]N yielded a significant correlation, its contribution to a linear model predicting external Se concentration was small (pH only, [r.sup.2]=0.240; pH + [[delta].sup.15]N, [r.sup.2]-0.268) because soil pH and [[delta].sup.15]N were significantly correlated (r=-0.524; P=0.012). No significant correlations were found at the former seabird site near Akaroa (data not shown).
Tree fern Se was 0.02-0.13mg[kg.sup.-1] (median 0.05 mg[kg.sup.-1]) compared with 0.03-0.20mg[kg.sup.-1] (median 0.08mg[kg.sup.-1]) for nikau (Fig. 4), an insignificant difference (Mann Whitney U=61.000; P=0.401).
Our study supports seabird breeding as an important contributor to the soil Se status of the Westland petrel site at Punakaiki. Selenium concentrations were much higher than predicted from parent material weathering, and the Se content of petrel guano was well above that of soil parent material but consistent with the values found in soil. Soil Se concentrations were highest at the Westland petrel site, and lowest at the site with no record of seabird breeding. Enrichment factors also supported the idea of a significantly enhanced Se status at both the present-day and former seabird breeding sites.
Intriguingly, concentrations within one profile at the former seabird site near Akaroa were comparable to those at the Westland petrel site. The higher concentrations in this profile could not be explained by differences in A1 and Fe content, Al and Fe being important Se-binding phases (Balistrieri and Chao 1990; Nakamaru et al. 2005; Floor et al. 2011). Although [Al.sub.2][O.sub.3] and [Fe.sub.2][O.sub.3] contents were almost identical (mean 16.7% and 9.9% at Stony Bay Peak v. 16.3% and 10.2% for remaining profiles), the Se concentrations were approximately double those of the other Akaroa profiles. The enhanced Se profile came from an altitude [greater than or equal to] 200 m higher than the other former colony profiles, on the summit plateau of Stony Bay Peak. Although seabird breeding on Banks Peninsula was probably in decline for centuries, isolated colonies on high points persisted well into the 20th Century (Stead 1927); Stony Bay Peak is one of Banks Peninsula's highest locations. Alternatively, the Se content of seabirds varies between even closely related species (Gonzalez-Solis et al. 2002), and seawater Se concentration and speciation varies geographically and seasonally (Sherrard et al. 2004). Consequently, the enhanced Se status on the summit of Stony Bay Peak could also be explained by the presence of a different seabird community, or foraging in a more Se-rich environment.
Soil pH and [[delta].sup.15]N both correlated with soil Se, but they also correlated with each other. At a between-island spatial scale, soil acidity and [[delta].sup.15]N both correlate with seabird breeding density (Mulder 2011). The effect of seabird input on pH arises from nitrification of guano; a multitude of processes can affect soil [[delta].sup.15]N. Seabird soil [[delta].sup.15]N is typically high due to the marine environment where seabirds feed, the high trophic position of seabirds, and an 'open' N cycle where large inputs of guano N are balanced by large losses (Hogberg 1997; Pardo et al. 2006). Nitrogen loss pathways at seabird sites likely to increase [[delta].sub.15]N include ammonia volatilisation (Blackall et al. 2007), nitrate leaching (Harding et al. 2004), and denitrification to nitrous oxide and [N.sub.2]. Although denitrification losses have not been formally measured or inferred at seabird sites, they are probably significant at the Westland petrel site given its high rainfall, comparatively fine soil texture, and high N concentrations.
Although soil redox potential at the Westland petrel site is unknown, the comparatively fine soil texture combined with the humid climate and acidic soil solution pH (range 3.6-4.7; Hawke and Powell 1995) likely favours selenite (Elrashidi et al. 1987; Srby et al. 2001). Selenite is always more strongly adsorbed than selenate (Bar-Yosef and Meek 1987; Balistrieri and Chao 1990), and modelling of selenite uptake by a wide range of soils is best represented (Goldberg et al. 2007) by the equilibrium:
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (2)
rather than formation of protonated or bidentate surface complexes. At the pH of the soils in our study, selenite is present as HSe[O.sub.3] - (rather than [H.sub.2]Se[O.sub.3]), while the surface could be present as either SOH or SO[H.sub.2.sup.+]. Corresponding equilibria pertaining to the Westland petrel system are therefore:
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (3)
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (4)
If reduction of selenate to selenite is important, Eqn 3 becomes the half equation:
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (5)
Equations 4 and 5 both imply relationships with pH, a plot of log(soil Se) (y-axis) v. pH (x-axis) having a slope of zero (Eqn 3), +1 (Eqn 4), or -3 (Eqn 5). A negative correlation between log (external Se) and pH was observed, the slope (-0.066 to -0.357; 95% CI) being intermediate between that predicted by Eqns 3 and 5. We conclude that the reductive sorption typified by Eqn 5 contributes significantly to the distribution of Se within Westland petrel soils, via its relationship with pH.
Given the positive correlation between Se and soil 815N, one might have expected a correlation between Se and either total or external P; this was not observed, unlike the calcareous system reported by Xu et al. (2011). Whereas all depth increments yielded Se and P concentrations well in excess of predicted parent-material-derived Se, contrasting depth distributions were found. External Se depth profiles showed a distinct mid-depth maximum, whereas external P concentrations (Hawke 2005) are greatest at or near the surface. Combined with the low concentrations in plant foliage, the depth profile pattern is consistent with a minimal role for plants in Se cycling (Jobbagy and Jackson 2001). Binding by soil colloids also involves different mechanisms, even though both inorganic P and Se show strong affinities for Al or Fe oxyhydroxide surfaces (Balistrieri and Chao 1990; Nakamaru et al. 2005). In the case of P, binding is likely to involve Al (or Fe)-humic-P complexes (Hawke and Powell 1995; Gerke 2010). Binding by amorphous Al phases is especially important for Se (Floor et al. 2011). Even though similar depth profiles were observed at all sites, the absence of a correlation with pH at the former site suggests that Se distribution has not been affected by changes in soil processes subsequent to seabird elimination and establishment of agriculture.
Despite the elevated soil Se concentrations, the median values for both plant species (tree fern and nikau) at the Westland petrel site fell close to the minimum concentration of 0.05-0.1 mg[kg.sup.-1] required for mammalian herbivore nutrition (Allaway and Hodgson 1964). None of the individual plant samples came close to the toxic level of ~1 mg[kg.sup.-1] (Hartikainen 2005). In accounting for apparently low rates of Se transfer from soil to plant, Hopper and Parker (1999) found that selenate is more readily taken up by roots than selenite, and that selenite uptake is inhibited by phosphate. Bicarbonate-extractable P at the study site is high (range 25-110mg[kg.sup.-1]), although soil solution P values are not exceptional (range 0.5-1.5 [micro]mol[L.sup.-1]) (Hawke and Powell 1995). Wang et al. (2012) also found that foliage Se concentrations were much lower than soil Se, and correlated with the proportion of soil Se present as selenate. We therefore propose that the comparatively low foliage Se concentrations that we report here derive from a low proportion of selenate (Eqn 5), which in turn results from the combination of soil properties and the humid climate. Phosphate inhibition of selenite uptake as an important driving factor seems unlikely given the reported range of soil solution P concentrations.
Total Se concentrations in the A horizon of New Zealand soils average 0.6 mg[kg.sup.-1] (Wells 1967). This is not exceptionally low in a global context, with (for example) large areas of China having concentrations less than half this value (Wang and Gao 2001). Nevertheless, agricultural animals in many areas of New Zealand respond to Se augmentation (Watkinson 1962). Our results support the hypothesis that seabirds augment soil Se at breeding locations in temperate environments such as New Zealand, with elevated concentrations remaining long after their elimination. Although more study sites, both with and without breeding petrels, would have been advantageous, these are not easy to find. The only other substantive seabird locality remaining in New Zealand (other than offshore islands) is at 1200-1800m altitude near Kaikoura in the north-east of South Island, hosting Hutton's shearwaters (Puffinus huttoni) (Cuthbert and Davis 2002). Documented former colony and seabird-free sites are similarly scarce.
We conclude that soil Se at the Westland petrel site is enhanced by the presence of seabirds, so that anthropogenic extinction of seabirds in New Zealand and elsewhere terminated an important Se source for terrestrial ecosystems. However, availability of seabird Se to terrestrial plants (and their consumers) may have been strongly constrained in wetter areas typified by the Westland petrel site. Although plant foliage Se concentrations were clearly non-toxic and close to minimum levels for herbivore nutrition at our study site, Se availability from seabirds may have been greater in less acidic, drier environments. Investigating the dynamics of Se in this context would be most interesting, especially given that Se deficiency is an important feature of the present day New Zealand landscape.
This study was initially prompted by a suggestion from Rob Fitzpatrick (CSIRO) that analysis of seabird soils for Se might be interesting. Samples were collected with the help of John Clark, Lily Clark, Jenny Hawke, Michael Hawke, Richard Holdaway, and Jenny Vallance; Richard Hold-away also contributed to figure draughting. Chippy Wood (Department of Conservation, DOC) provided guidance through the Westland petrel study site, and the ongoing enthusiasm of the DOC's West Coast Conservancy for petrel research is gratefully acknowledged. Access to the Akaroa study site was kindly granted by Maurice White Native Forest Trust (Hugh Wilson) and by Graeme Curry, and access to the control site on Takaka Hill was facilitated by Mike Endres. Reviewer comments are also acknowledged with thanks.
Received 23 May 2012, accepted 10 October 2012, published online 13 November 2012
Allaway WH, Hodgson JF (1964) Symposium on nutrition of forages and pastures: selenium in forages as related to the geographic distribution of muscular dystrophy in livestock. Journal of Animal Science 23, 271-277.
Andersen JC (1927) 'Place names of Banks Peninsula.' (New Zealand Government Printer: Wellington)
Ararjo G, Gonzalez MH, Ferreira AG, Nogueira ARA, Nrbrega JA (2002) Effect of acid concentration on closed-vessel microwave-assisted digestion of plant materials. Spectrochimica Acta B: Atomic Spectroscopy 57, 2121-2132. doi: 10.1016/S0584-8547(02)00164-7
Balistrieri LS, Chao TT (1990) Adsorption of selenium by amorphous iron oxyhydroxide and manganese dioxide. Geochimica et Cosmochimica Acta 54, 739-751. doi:l0.1016/0016-7037(90)90369-V
Bar-Yosef B, Meek D (1987) Selenium sorption by kaolinite and montmorillonite. Soil Science 144, 11-19. doi:10.1097/00010694-198707000-00003
Blackall TD, Wilson L J, Theobald MR, Milford C, Nemitz E, Bull J, Bacon P J, Hamer K, Wanless S, Sutton MA (2007) Ammonia emissions from seabird colonies. Geophysical Research Letters 34, L10801. doi: 10.1029/2006GL028928
Cuthbert R, Davis LS (2002) The breeding biology of Hutton's shearwater. Emu 102, 323-329. doi:10.1071/MU01032
Elrashidi MA, Adriano DC, Workman SM, Lindsay WL (1987) Chemical equilibria of selenium in soils: a theoretical development. Soil Science 144, 141-152. doi:10.1097/00010694-198708000-00008
Fecher PA, Goldman I, Nagengast A (1998) Determination of iodine in food samples by inductively coupled plasma mass spectrometry after alkaline extraction. Journal of Analytical Atomic Spectrometry 13, 977-982. doi: 10.1039/a801671b
Floor GH, Roman-Ross G (2012) Selenium in volcanic environments: a review. Applied Geochemistry 27, 517-531. doi: 10.1016/j.apgeochem. 2011.11.010
Floor GH, Calabrese S, Roman-Ross G, D'Alessandro W, Aiuppa A (2011) Selenium mobilization in soils due to volcanic-derived acid rain: an example from Mt Etna volcano, Sicily. Chemical Geology289, 235-244. doi: 10.1016/j.chemgeo.2011.08.004
Fordyce F (2007) Selenium geochemistry and health. Ambio 36, 94-97. doi: 10.1579/0044-7447(2007)36194:SGAH]2.0.CO;2
Gerke J (2010) Humic (organic matter-Al(Fe)-phosphate complexes: an underestimated phosphate form in soils and source of plant-available phosphate. Soil Science 175, 417-425. doi:10.1097/SS. 0b013e3181flb4dd
Goldberg S, Lesch SM, Suarez DL (2007) Predicting selenite adsorption by soils using soil chemical parameters in the constant capacitance model. Geochimica et Cosmochimica Acta 71, 5750-5762. doi:10.1016/j.gca. 2007.04.036
Goldberg S, Hyun S, Lee LS (2008) Chemical modeling of arsenic(III, V) and selenium(IV, VI) adsorption by soils surrounding ash disposal facilities. Vadose Zone Journal 7, 1231-1238. doi:10.2136/vzj2008.0013
Gonzalez-Solis J, Sanpera C, Ruiz X (2002) Metals and selenium as bioindicators of geographic and trophic segregation in giant petrel Macronectes spp. Marine Ecology Progress Series 244, 257-264. doi: 10.3354/meps244257
Harding JS, Hawke D J, Holdaway RN, Winterbourn MJ (2004) Incorporation of marine-derived nutrients from petrel breeding colonies into stream food webs. Freshwater Biology 49, 576-586. doi:10.1111/j.1365-2427.2004.01210.x
Hartikainen H (2005) Biogeochemistry of selenium and its impact on food chain quality and human health. Journal of Trace Elements in Medicine and Biology 18, 309-318. doi:10.1016/j.jtemb.2005.02.009
Hawke DJ (2001) Variability of [[delta].sup.15]N in soil and plants at a New Zealand hill country site: correlations with soil chemistry and nutrient inputs. Australian Journal of Soil Research 39, 373-383. doi:10.1071/ SR99094
Hawke DJ (2003) Cadmium distribution and inventories at a pre-European seabird breeding site on agricultural land, Banks Peninsula, New Zealand. Australian Journal of Soil Research 41, 19-26. doi: 10.1071/ SR02053
Hawke DJ (2004) Maximum possible age of a petrel breeding colony near Punakaiki (South Island, New Zealand) from radiocarbon and stable isotope analysis of soil. Journal of the Royal Society of New Zealand 34, 1-7. doi:10.1080/03014223.2004.9517760
Hawke DJ (2005) Soil P in a forested seabird colony: inventories, parent material contributions, and N:P stoichiometry. Australian Journal of Soil Research 43, 957-962. doi: 10.1071/SR05075
Hawke DJ (2010) Using [sup.137]Cs and [sup.210]Pb to characterise soil mixing by burrowing petrels: an exploratory study. New Zealand Journal of Zoology 37, 53-57. doi:10.1080/03014221003692829
Hawke DJ, Powell HKJ (1995) Soil solution chemistry at a Westland petrel breeding colony, New Zealand: palaeoecological implications. Australian Journal of Soil Research 33, 915-924. doi:10.1071/ SR9950915
Hawke DJ, Clark JM, Vallance JR (2012) Breeding Westland petrels as providers of detrital carbon and nitrogen for soil arthropods: a stable isotope study. Journal of the Royal Society of New Zealand doi: l0.1080/ 03036758.2011.616211
Hodson ME (2002) Experimental evidence for the mobility of Zr and other trace elements in soil. Geochimica et Cosmochimica Aeta 66, 819-828. doi: 10.1016/S0016-7037(01)00803-1
Hogberg P (1997) [sup.15]N natural abundance in soil-plant systems. New Phytologist 137, 179-203. doi: 10.1046/j.1469-8137.1997.00808.x
Holdaway RN, Hawke D J, Hyatt OM, Wood GC (2007) Stable isotopic ([[delta].sup.15]N, [[delta].sup.13]C) analysis of wood in trees growing in past and present colonies of burrow-nesting seabirds in New Zealand. I. [[delta].sup.15]N in two species of conifer (Podocarpaceae) from a mainland colony of Westland petrels (Procellaria westlandica), Punakaiki, South Island. Journal of the Royal Society of New Zealand 37, 75-84. doi:10.1080/03014220709510537
Hopper JL, Parker DR (1999) Plant availability of selenite and selenate as influenced by competing ions phosphate and sulfate. Plant and Soil 210, 199-207. doi: 10.1023/A: 1004639906245
Jobbagy EG, Jackson RB (2001) The distribution of soil nutrients with depth: global patterns and the imprint of plants. Biogeochemistry 53, 51-77. doi:10.1023/A:1010760720215
Mincher B, Mionczynski J, Hnilicka P (2007) Soil redox chemistry limitation of selenium concentration in Carex species sedges. Soil Science 172, 733-739. doi:10.1097/ss.0b013e3180d0a3da
Mulder CPH (2011) Impacts of seabirds on plant and soil properties. In 'Seabird islands: ecology, invasion, and restoration'. (Eds CPH Mulder, WB Anderson, DR Towns, PJ Bellingham) pp. 135-176. (Oxford University Press: New York)
Nakamaru Y, Tagami K, Uchida S (2005) Distribution coefficient of selenium in Japanese agricultural soils. Chemosphere 58, 1347-1354. doi: 10.1016/j.chemosphere.2004.09.086
Pardo LH, Templer PH, Goodale CL, Duke S, Groffman PM, Adams MB, Boeckx P, Boggs J, Campbell J, Colman B, Compton J, Emmett B, Gundersen P, Kjonaas J, Lovett G, Mack M, Magill A, Mbila M, Mitchell M J, McGee G, McNulty S, Nadelhoffer K, Ollinger S, Ross D, Rueth H, Rustad L, Schaberg P, Schiff S, Schleppi P, Spoelstra J, Wessel W (2006) Regional assessment of N saturation using foliar and root [[delta].sup.15]N. Biogeochemistry 80, 143-171. doi:10.1007/s10533-0069015-9
Perkins WT (2011) Extreme selenium and tellurium contamination in soils an eighty year-old industrial legacy surrounding a Ni refinery in the Swansea Valley. The Science of the Total Environment 412-413, 162-169. doi:10.1016/j.scitotenv.2011.09.056
Ranville MA, Cutter GA, Buck CS, Landing WM, Cutter LS, Resing JA, Flegal AR (2010) Aeolian contamination of Se and Ag in the north Pacific from Asian fossil fuel combustion. Environmental Science & Technology 44, 1587-1593. doi:10.1021/es902523m
Seby F, Potin-Gautier M, Giffaut E, Borge E, Donard OFX (2001) A critical review of thermodynamic data for selenium species at 25[degrees]C. Chemical Geology 171, 173-194. doi:10.1016/S0009-2541(00)00246-1
Sherrard JC, Hunter KA, Boyd PW (2004) Selenium speciation in subantarctic and subtropical waters east of New Zealand. Deep-sea Research. Part I, Oceanographic Research Papers 51, 491-506. doi: 10.1016/j.dsr.2003.11.001
Stead EF (1927) The native and introduced birds of Canterbury. In 'Natural history of Canterbury'. (Eds R Speight, A Wall) pp. 204-225. (Philosophical Institute of Canterbury: Christchurch, New Zealand)
Steadman DW (1995) Prehistoric extinctions of Pacific Island birds: biodiversity meets zooarchaeology. Science 267, 1123-1131. doi: 10.1126/science.267.5201.1123
Stolz JF, Basu P, Santini JM, Oremalnd RS (2006) Arsenic and selenium in microbial metabolism. Annual Review of Microbiology 60, 107-130. doi: 10.1146/annurev.micro.60.080805.142053
Sun LG, Xie ZQ, Zhao JL (2000) A 3,000-year record of penguin populations. Nature 407, 858. doi: 10.1038/35038163
Sutherland RA (2000) Bed-sediment associated trace metals in an urban stream, Oahu, Hawaii. Environmental Geology 39, 611-627. doi: 10.1007/s002540050473
Torres J, Pintos V, Dominguez S, Kremer C, Kremer E (2010) Selenite and selenate speciation in natural waters: interaction with divalent metal ions. Journal of Solution Chemistry 39, 1-10. doi:10.1007/s10953-009-9491-3
Towns DR, Byrd GV, Jones HP, Rauzon MJ, Russell JC, Wilcox C (2011) Impacts of introduced predators on seabirds. In 'Seabird islands: ecology, invasion, and restoration'. (Eds CPH Mulder, WB Anderson, DR Towns, PJ Bellingham) pp. 56-90. (Oxford University Press: New York)
Wang Z, Gao Y (2001) Biogeochemical cycling of selenium in Chinese environments. Applied Geochemistry 16, 1345-1351. doi:10.1016/ S0883-2927(01)00046-4
Wang S, Liang D, Wang D, Wei W, Fu D, Lin Z (2012) Selenium fractionation and speciation in agricultural soils and accumulation in corn (Zea mays L.) under field conditions in Shaanxi Province, China. The Science of the Total Environment 427-428, 159-164. doi: 10.1016/j. scitotenv.2012.03.091
Watkinson JH (1962) Soil selenium and animal health. Transactions, Commission IV and V. International Society of Soil Science. pp. 149-154.
Wells N (1967) Selenium content of soil-forming rocks. New Zealand Journal of Geology and Geophysics 10, 198-208. doi:10.1080/ 00288306.1967.10428190
Worthy TH, Holdaway RN (1994) Quaternary fossil faunas from caves in Takaka Valley and on Takaka Hill, northwest Nelson, South Island, New Zealand. Journal of the Royal Society of New Zealand 24, 297-391. doi: 10.1080/03014223.1994.9517474
Worthy TH, Holdaway RN (1996) Quaternary fossil faunas overlapping taphonomies, and palaeofaunal reconstruction in North Canterbury, South Island, New Zealand. Journal of the Royal Society of New Zealand 26, 275-361. doi: 10.1080/03014223.1996.9517514
Xu L-Q, Liu X-D, Sun L-G, Yan H, Liu Y, Luo Y-H, Huang J (2011) Geochemical evidence for the development of coral island ecosystem in the Xisha Archipelago of South China Sea from four omithogenic sediment profiles. Chemical Geology 286, 135-145.
Yaroshevsky AA (2006) Abundances of chemical elements in the Earth's crust. Geochemistry International 44, 48-55. doi:10.1134/ S001670290601006X
David J. Hawke (A,B) and Jun-Ru Wu (A)
(A) Department of Applied Sciences and Allied Health, Christchurch Polytechnic Institute of Technology, PO Box 540, Christchurch 8140, New Zealand.
(B) Corresponding author. Email: firstname.lastname@example.org
Table 1. Soil chemical characteristics at former seabird site (Akaroa) and present-day seabird site (Punakaiki) Values are mean [+ or -] s.d. on a whole-soil basis from the surface to 50 cm (or lithic contact). LOI, Loss on ignition at 1000[degrees]C Former [Al.sub.2][O.sub.3] (g [100.sup.-1] g of soil) 6.4 [+ or -] 2.2 [Fe.sub.2][0.sub.3] (g [100.sup.-1] g of soil) 10.1 [+ or -] 3.0 Total P (mg [kg.sup.-1]) 2035 [+ or -] 485 Total Zr (mg [kg.sup.-1]) 344 [+ or -] 59 LOI 17.6 [+ or -] 6.3 pH(water) 4.6 [+ or -] 0.3 Present-day [Al.sub.2][O.sub.3] (g [100.sup.-1] g of soil) 13.6 [+ or -] 1.5 [Fe.sub.2][0.sub.3] (g [100.sup.-1] g of soil) 5.05 [+ or -] 0.70 Total P (mg [kg.sup.-1]) 1015 [+ or -] 470 Total Zr (mg [kg.sup.-1]) 202 [+ or -] 19 LOI 16.8 [+ or -] 7.0 pH(water) 4.1 [+ or -] 0.3 Table 2. Pearson correlation coefficients between soil selenium (Se) concentration and zirconium (Zr) (indicator of parent material), iron (Fe), aluminium (Al) and loss on ignition (LOI) (potential Se binding phases), nitrogen (N) and phosphorus (P) (other potentially seabird derived elements), pH (effect on selenate reduction and sorptive uptake), and [[delta].sup.15]N (indicative of N cycling intensity) in soil profiles from the present-day seabird breeding site at Punakaiki Sample size: n=29, except for pH analysis where n=22 due to accidental loss. * P < 0.05 Zr Fe Al LOI N P Total Se -0.009 0.107 -0.071 0.056 0.061 0.079 External Se -0.156 0.019 -0.156 0.183 0.189 0.213 External P pH [[delta].sup.15]N Total Se -0.009 -0.387 0.368 * External Se -0.155 -0.490 * 0.405 *
|Printer friendly Cite/link Email Feedback|
|Author:||Hawke, David J.; Wu, Jun-Ru|
|Date:||Oct 1, 2012|
|Previous Article:||Spatial heterogeneity of soil physico-chemical properties in contrasting wetland soils in two agro-ecological zones of Lesotho.|
|Next Article:||Nitrous oxide emissions from subtropical horticultural soils: a time series analysis.|