Controls, variation, and a record of climate change in detailed stable isotope record in a single bryozoan skeleton $ Abigail M. Smith a, * and Marcus M. Key Jr. b a Department of Marine Science, University of Otago, P. O. Box 56, Dunedin, New Zealand b Department of Geology, Dickinson College, Carlisle, PA 17013-2896, USA Received 14 October 2002 Abstract The long-lived (about 20 yr) bryozoan Adeonellopsis sp. from Doubtful Sound, New Zealand, precipitates aragonite in isotopic equilibrium with seawater, exerting no metabolic or kinetic effects. Oxygen isotope ratios (y 18 O) in 61 subsamples (along three branches of a single unaltered colony) range from 0.09 to +0.68x PDB (mean = +0.36x PDB). Carbon isotope ratios (y 13 C) range from +0.84 to +2.18x PDB (mean = +1.69x PDB). Typical of cool-water carbonates, y 18 O-derived water temperatures range from 14.2 to 17.5 jC. Adeonellopsis has a minimum temperature growth threshold of 14 jC, recording only a partial record of environmental variation. By correlating seawater temperatures derived from y 18 O with the Southern Oscillation Index, however, we were able to detect major events such as the 1983 El Nin ˜o. Interannual climatic variation can be recorded in skeletal carbonate isotopes. The range of within-colony isotopic variability found in this study (0.77x in y 18 O and 1.34 in y 13 C) means that among-colony variation must be treated cautiously. Temperate bryozoan isotopes have been tested in less than 2% of described extant species — this highly variable phylum is not yet fully understood. D 2004 University of Washington. All rights reserved. Keywords: Bryozoa; Stable isotopes; y 18 O; y 13 C; Climate; ENSO; New Zealand Introduction Carbonate skeletons of marine invertebrates often con- tain a geochemical record of seawater conditions (e.g., temperature) throughout their lives. Stable isotope paleo- thermometry has been used in both modern and ancient tropical settings (see, e.g., Corfield, 1995), but to a large extent temperate shelf environments have been ignored (one rare exception is Goodwin et al., 2001). Carbonate sediments on Southern Hemisphere temperate shelves are dominated by skeletal remains of bryozoans and molluscs (James, 1997; Nelson et al., 1988). Bryozoans are more significant contributors to cool-water carbonates than their tropical counterparts, so much so that several of the cool-water, shallow marine carbonate facies (e.g., bryomol, bryoderm) are named after this abundant component (Nel- son et al., 1988; Rao, 1996; James, 1997). In some areas of the New Zealand shelf, the ‘bryomol’ province may be up to 80 wt% bryozoans (Nelson et al., 1988). Large erect shelf species may weigh up to 200 g and live some 20 years (Smith et al., 2001). It was Forester et al. (1973) who first suggested that bryozoans precipitate CaCO 3 skeletons in equilibrium with surrounding seawater. They studied mainly cheilostome bryozoans, making no corrections for mineralogy, nor did they correlate the isotopic composition of seawater with skeletal carbonate. Pa ¨tzold et al. (1987) used fine-scale variations in oxygen isotopes in the bryozoan Pentapora foliacea to elucidate seasons and therefore growth rate, as did others later (e.g., Brey et al., 1999, Bader, 2000). Rao and Nelson (1992) concluded that modern temperate bryo- zoan skeletons were reliable indicators of seawater che- mistry, being little affected by kinetic and metabolic considerations. Surveys of Australian living and Recent bryozoans of various mineralogies agreed (Rahimpour- Bonab et al., 1997a; Bone and James, 1997). By the end of the last century it appeared that cool-water bryozoans (as represented by 180 specimens from 49 genera and 65 0033-5894/$ - see front matter D 2004 University of Washington. All rights reserved. doi:10.1016/j.yqres.2003.11.001 $ Supplementary data for this article (Appendix) are available on Science Direct (http://www.sciencedirect.com). * Corresponding author. Fax: +1-64-3-479-8336. E-mail address: abbysmith@otago.ac.nz (A.M. Smith). www.elsevier.com/locate/yqres Quaternary Research 61 (2004) 123 – 133