Calcium, Magnesium and Strontium Concentrations in ... - Science Direct

0 downloads 0 Views 983KB Size Report
The mineralogical analysis of the sclerites of four Mediterranean gorgonians revealed magnesium calcite as the only calcium carbonate polymorph. The. Mg/Ca ...
Estuadne, Coastal and Shelf Science (1995) 40, 87-104

Calcium, M a g n e s i u m and Strontium Concentrations in the Calcite Sclerites of Mediterranean Gorgonians (Coelenterata: Octocorallia)

Markus

G. Weinbauer

~ and Branko

Velimirov

Institut fiir Allgemeine Biologic, Universitiit Wien, Schwarzspanierstr. 17, A-1090 Vienna, Austria Received 3 September 1993 and in revised form 13 January 1994

Keywords: gorgonians; calcium; magnesium; strontium; sclerites; calcite The mineralogical analysis of the sclerites of four Mediterranean gorgonians revealed magnesium calcite as the only calcium carbonate polymorph. The Mg/Ca ratio (0.064-0.098) and the St/Ca ratio (0"0014-0"0025) in the sclerites of the investigated species were the lowest ever reported in octocoral calcite. Calcium concentrations generally did not vary with water depth or colony region, whereas there was a high intraspecific variation of Mg/Ca and Sr/Ca ratios related to water depth and colony region. This variation of the Mg/Ca and St/Ca ratios in the sclerites cannot be explained by the chemical composition of the seawater, as the Mg/Ca and St/Ca ratios of seawater were constant over the depths and seasons investigated (4.0 ± 0-04 and 0.0084 ± 0.00015). The comparison with inorganic precipitated calcite revealed a biological enrichment of Mg and a discrimination against Sr in octocoral calcite. Our data do not support the hypothesis that the Mg concentration is influenced by the growth rate. According to our data on Mg concentrations in gorgonian sclerites and ambient water temperature, the relationship between Mg content in octocorals and ambient water temperature derived by Chave (1954) should probably be revised. A positive correlation between Mg and Sr values in octocoral calcite (including data from literature) intimates that Mg is an important factor influencing the incorporation of Sr into the calcium carbonate matrix. No correlation between Sr concentration and temperature was found, however, growth rate is discussed as a potential factor influencing Sr concentrations.

Introduction T h e mesoskeletons (sclerites, ' spicules ') of octocorals are c o m p o s e d of m a g n e s i u m calcite, whereas calcium carbonate precipitations within the h o r n y axial skeletons m a y be of the calcium carbonate p o l y m o r p h aragonite (Milliman, 1974). According to the classification of mineralization processes presented by L o w e n s t a m a n d W e i n e r (1989), biologically i n d u c e d mineralization is characterized b y processes n o t specifically aPresent address: Institute of Zoology, University of Vienna, Deparmaent of Marine Biology, Althanstr. 14, A-1070 Vienna, Austria. 0272-7714/95/010087+ 18 $08.00/0

© 1995 Academic Press Limited

88

M. G. IVeinbauer & B. Velimirov

designed for mineralization, but which do, in fact, result in a mineral being formed. The axial deposition of some octocoral species belongs to this type of mineralization, as the mineralization in the hollow central region of the axis is very similar to inorganic precipitation (Lewis et al., 1992) and the aragonite/calcite ratios in the calcium carbonate skeletons are controlled by temperature (Lowenstam, 1964). The axial mineralization of other octocoral species (Ledger & Franc, 1978; Lewis et al., 1992) as well as the sclerite formation of all octocorals, however, corresponds to the type of the biologically controlled mineralization (Lowenstam & Weiner, 1989), since space delineation and specialized cellular and macromolecular mechanisms control the process of calcium carbonate precipitation. In case of the sclerite formation these mechanisms include the control of the Ca metabolism and sclerite calcification by Ca-ATPase activity, the involvement of a carboanhydrase in this process and the formation of an organic matrix and its transport to the site of mineralization (Kingsley & Watabe, 1984a, b, 1985a, b, 1987; Goldberg & Benayahu, 1987). Magnesium and Sr analyses in calcareous skeletons were conducted for the potential use of these elements as ecological and palaeoecological indicators (Dodd, 1967). For example, palaeotemperatures and palaeosalinity were reconstructed by the determination of the Mg/Ca and Sr/Ca ratios in calcite ostracods (Chivas et al., 1985, 1986). However, discordant interpretations have been offered to explain the variations of Mg and Sr concentrations in the calcium carbonate matrix (Chave, 1954; Thompson & Chow, 1955; Dodd, 1967; Milliman, 1974). In octocorals, Ca metabolism, calcification of both sclerites and axis as well as Ca and Mg concentrations of the sclerites vary with colony region (Velimirov & B6hm, 1976; Velimirov & King, 1979; Kingsley & Watabe, 1989; Esford & Lewis, 1990). According to Chave (1954) there is a linear relationship between Mg concentrations in the calcite sclerites of octocorals and the ambient water temperature. However, Velimirov and B6hm (1976) found M g C O 3 concentrations in gorgonians that were somewhat lower than predicted by Chave. Another Mg incorporation hypothesis was established by Moberly (1968) who showed that the Mg content in the magnesium calcite of coralline algae and pelecypods increased with skeletal gro~x_h rate. There are some data on the Sr concentrations of calcite octocorals (Thompson & Chow, 1955; Milliman, 1974; Mat6 et al., 1986), but no information is available on the mechanisms controlling the incorporation of Sr in octocoral skeletons. For the calcite skeletons of echinoderms and bivalves it was assumed that the Sr/Ca ratio is a function of the Sr/Ca ratio in seawater (Thompson & Chow, 1955; Lorens & Bender, 1980). Although there are some attempts to relate Sr concentrations in calcite skeletons of molluscs to water temperature or growth rate, it was not possible to establish simple cause-effect relationships for all species (Milliman, 1974; Rosenberg, 1980). In a study on Mg and Sr in inorganic magnesium calcite, Mucci and Morse (1983) found that the incorporation of Mg promotes the precipitation of St. The present study on the variability of Mg and Sr concentrations in different colony regions of the four most abundant Mediterranean gorgonians collected from various depths, was conducted to answer the following questions. (1) Are Mg and Sr concentrations related to temperature or to the chemical composition of the seawater? (2) Are Mg and Sr in the sclerites fractionated similarly to these elements in inorganic calcite? (3) Can Mg concentrations in the sclerites influence the Sr incorporation as in inorganic calcite?

Chemical composition of octocoral sclerites

89

Material and methods

Study site and general procedure Colonies of Paramuricea clavata, Eunicella cavolinii and E. singularis were collected in the bay of Calvi in the north-west of Corsica, France (42°35'N, 8°45'E), and specimens of Lophogorgia ceratophyta were harvested in Cinque Terre, Italy (44°05'N, 8°45'E). The colonies were cleaned of epiphytes and dried to constant weight at 80 °C. They were then divided into three colony regions: branch tips (5 m m for E. cavolinii, P. clavata and L. ceratophyta and 20 m m in the case of E. singularis, a species with few ramifications), branches (the sum of the end branches without branch tips) and the basal region (rest of the colony). The next step was to separate the cortex from the horny skeletons by tweezers or by grinding. In several colonies the organic material was removed from the sclerites by combusting the cortex at 450 °C for 4 h (Harvell & Suchanek, 1987; Harvell & Fenical, 1989). The cortex mineralization was calculated as the percent proportion of sclerites to the cortex weight. From August 1989 to August 1991 the water temperature in the Bay of Calvi was measured one to three times per month for each metre in depth, from 0 to 40 m. In August 1990 (warm season) and March 1991 (cold season) seawater was collected for chemical analyses with acid-rinsed Niskin bottles at 0, 10, 20, 30 and 40 m depth and filtered immediately through a Whatman GF/F filter. Samples were stored cooled (4 °C) until the chemical analysis was conducted.

Chemical and mineralogical analyses Cortex samples were taken from different colony regions. The cortex of several colonies from the same colony region and water depth was pooled. The weight of the cortex derived per colony was roughly the same. Furthermore, cortex tissue was taken separately from the colony regions of single colonies. In the case orE. cavolinii from 7 m, all the cortex from several colonies was pooled without dividing it into colony regions, because the colonies were small and the number of ramifications was low. The cortex was ground to powder and the organic matter removed from the mesoskeleton by hot (95 °C) commercial domestic bleach. The remaining skeleton was rinsed 10 times in distilled water and reinspected under the stereomicroscope for remaining organic material. Seawater samples were acidified with concentrated H N O 3 to a final solution of 5%. Sclerite samples were dissolved with concentrated H N O 3 and topped up with distilled water to a final solution of 5% H N O 3. Skeletal Ca, Mg and Sr concentrations are given as a percentage of sclerite dry weight and Mg/Ca and Sr/Ca values are given as a mole ratio. The distribution coefficient for Mg and Sr (KMg and Ksr ) is defined as {(Me/Ca) in the skeleton}/{(Me/Ca) in the water} where Me (Mg or Sr) and Ca are the molar concentrations of these elements. Total carbonate content for the calculation of tool% M g C O 3 and mol% SrCO 3 was the sum of %CaCO3, %MgCO3 and %SrCO 3. All samples were taken in triplicate. Calcium, Mg and Sr concentrations were determined by atomic absorption spectroscopy. The precision error of the apparatus (VARIAN SpectrAA 30) was 3.0% for Ca, 1"9% for Mg and 3"3% for Sr as determined by replicate analyses of single samples within 2 days. For the mineralogical analyses, sclerite samples of at least 1 g were milled to a grain size of less than 5 ~m. The mineralogical analyses were carried out with X-ray powder diffraction analysis (Phillips PW1710).

90

M. G. Weinbauer& B. Velimirov

30 ¸

25.

oc 20

15

10

I I I I I I I I I 1 I I ran. Feb. Mar. Apr. May Jun. Jul. Aug. Sep. Oct. Nov. Dec. Month

Figure 1. Seasonal variation of water temperature from different depths in the Bay of Calvi. Values are given as means calculated from all data determined from August 1989 to August 1991. (--A--) 10; (--O--) 20; (--D--) 40m.

Seasonal growth O n 1 September 1990, four colonies ofE. cavolinii at a depth of 10 m were sketched a n d the total b r a n c h length was d e t e r m i n e d in situ with sliding calipers. At the e n d of April 1991 the new growth was m e a s u r e d a n d the new growth from half of the tips clipped off and analysed. Finally, at the end of August 1991 the entire colonies were harvested a n d after m e a s u r e m e n t of all branches the newly grown tips were clipped off. Seasonal growth rates were calculated a n d n a m e d cold season a n d w a r m season growth. Extraction of the sclerites from the cold season a n d w a r m season tips a n d the d e t e r m i n a t i o n of the Ca, M g a n d Sr concentrations were clone as described above.

Statistical analysis T h e variation of parameters related to colony region a n d water depth was analysed with the Kruskal-Wallis one-way analysis of variance using S Y S T A T a n d S T A T V i e w software packages. T h e Wilcoxon signed-rank test was performed in order to analyse the variation of the chemical composition of seawater b e t w e e n seasons.

Results Temperature of the seawater For each depth investigated, the lowest temperature values were f o u n d in F e b r u a r y a n d the highest in August (Figure 1). F r o m April to M a y there was a sharp increase in water temperature, which was more p r o n o u n c e d at 10 m t h a n at 20 m a n d 40 m. T h i s sharper increase in the more shallow waterbody was probably due to faster w a r m i n g above the thermocline located between 20 and 25 m. T e m p e r a t u r e values were still elevated in September and October, b u t decreased rapidly from O c t o b e r to N o v e m b e r . F r o m M a y to October the water temperature was highest at 10 m a n d lowest at 40 m, whereas during the rest of the year the water temperature did n o t vary with depth.

Chemical composition of the seawater T h e m e a n Ca concentration values a n d the M g / C a a n d St/Ca ratios in seawater from all depths and seasons were 373 4- 13, 4.0 4- 0-04 a n d 0-0084 4- 0.00015 p p m , respectively.

Chemical composition of octocoral sderites

91

TABLE 1. Calcium, Mg and Sr concentrations of seawater at different seasons Season

Depth (m)

Ca (ppm)

Mg (ppm)

Warm

0 10 20 30 40 0 10 20 30 40

3734- 13 372+7 371 4- 9 372 -4-23 3744-6 3764- 14 377 4- 4 3694-3 375 4- 3 372 4- 3

905+4 893+19 899 + 13 895 4- 10 882+19 8904-4 904 4- 7 8834-15 911 4- 9 890 4- 1

Cold

mMg/~Ca

Sr (ppm)

4.04-0.01 3.9=1=0.08 4.0 q- 0-06 4-0 -4-0.04 3.9+0-08 3.94-0.02 4.0 4- 0.03 4.04-0.07 4-0 4- 0.04 4.0 4- 0-01

6.94-0.17 6-9=t=0-11 6-8 4- 0-14 6.7 :t: 0.30 6.74-0.16 6-8:t:0.09 6-7 -4-0-03 6.84-0-14 7-0 :t: 0.05 6.8 4- 0.34

mSr/~Ca 0.0085+0.00020 0.0085=t=0.00013 0.0084 + 0-00825 0-0083 + 0.00037 0.00814-0.00020 0-00825:0.00011 0-0082 5:0.00003 0.00854-0-00017 0.0086 + 0.00006 0-0084 4- 0.00042

M a g n e s i u m averaged 895 + 1 0 p p m a n d Sr 6"8 ± 0.11 p p m . I o n c o n c e n t r a t i o n s a n d ratios d i d n o t vary with d e p t h ( T a b l e 1; P > 0 . 1 ) or season ( P > 0 " 2 2 3 ) .

Sclerite dry weight, cortex mineralization and colony region In the zooxanthellate species E. singulaffs the sclerite dry weight (expressed as p e r c e n t of total) was highest in the b r a n c h e s (53.6%), followed b y the basal region (38.0%) a n d lowest in the tips (8.4%; F i g u r e 2). In all o t h e r species, which are azooxanthellate, the sclerite d r y weight was highest in the base, r a n g i n g from 51 to 61%, a n d the lowest values were always f o u n d in the tips ( 1 4 - 1 9 % ) . Lophogorgia ceratophyta was the species with the highest sclerite d r y weight in the basal region (61.0%). I n all species the cortex m i n e r a l i z a t i o n was significantly stronger in the base t h a n in the tips (P 0 " I ) . T h e C a c o n c e n t r a t i o n s o f E. singularis, E. cavolinii, P. clavata a n d L. ceratophyta d i d n o t differ; a n d the C a values o f all samples r a n g e d f r o m 33"3 to 36"6%. T h e C a concentrations s h o w e d no significant variation with d e p t h or colony region for a given species ( P > 0 . 0 5 ) .

92

M. G. Weinbauer & B. Velimirov

E. singularis

E. cavolini Tip SDW: 8.4 ± 2.70 CM: 73.6 ± 3-10

Tip SDW: 18.5 ± 5-07 CM: 78.9 ± 1-17

Branch SDW: 53.6 ± 10-98 CM: 74.1 ± 3.57

Branch SDW: 30-5 ± 7.56 CM: 80.2 ± 3-10

Base SDW: 38-0 ± 9-33 CM: 78.4 ± 2.87

Base SDW: 51.0 ± 4.83 CM: 81-4 ± 2,47

/ l l l l

P. clavata

L. ceratophyta Tip SDW: 14.7 ± 1.99 CM: 66.6 _+5.49

TiP SDW: 13.7 ± 4.27 CM: 66-6 ± 1.91

~

Branch SDW: 25.3 ± 1.97 CM: 67.0 ± 2.92

Branch SDW: 33.0 ± 6.45 CM: 73.3 ± 4.99

Base SDW: 61.0 ± 5.22

Base SDW: 52.4 ± 7.84 CM: 76.9 ± 3-40

CM: 72-4 ± 2.00 75

Figure 2. Sclerite dry weight (SDW) and percent cortex mineralization (CM) in different growth regions of the investigated gorgonians. SDW is given as percentage of total, n=9 for E. singularis; n=10 for E. cavolini; n=10 for P. clavata; n=6 for L. ceratophyta.

M g and Sr variation in the sclerites in relation to water depth and colony region With water depth, E. singularis showed two trends in the M g / C a ratios (Figure 4). T h e Mg/Ca values in the basal region decreased significantly from 20 to 35 m (P