New Caledonia Coral Stable Isotope Data: Readme file --------------------------------------------------------------------- NOAA Paleoclimatology Program and World Data Center A - for Paleoclimatology --------------------------------------------------------------------- NOTE: PLEASE CITE ORIGINAL REFERENCE WHEN USING THIS DATA!!!!! NAME OF DATA SET: New Caledonia Coral Stable Isotope Data LAST UPDATE: 8/1995 (Original Receipt by WDCA Paleo) CONTRIBUTORS: Quinn, T.M., Taylor, F.W., Crowley, T.C., and Link, S. IGBP PAGES/WDCA Data Contribution Series #: 1998-008 SUGGESTED DATA CITATION: Quinn, T.M., Taylor, F.W., Crowley, T.C., and Link, S., 1998, New Caledonia Coral Stable Isotope Data. IGBP PAGES/World Data Center-A for Paleoclimatology, Data Contribution Series # 1998-008. NOAA/NGDC Paleoclimatology Program, Boulder CO, USA. ORIGINAL REFERENCE: Quinn, T.M., Taylor, F.W., Crowley, T.C., and Link, S., 1996, Evaluation of Sampling Resolution in Coral Stable Isotope Records: A Case Study Using Monthly Stable Isotope Records from New Caledonia and Tarawa; Paleoceanography, v. 11, n. 5, 529-542. GEOGRAPHIC REGION: Equatorial Pacific PERIOD OF RECORD: 1956- 1996 AD LIST OF FILES: caledonia.iso.txt, caledonia.iso.xls DESCRIPTION: Coral Site: Amedee Lighthouse, New Caledonia (22°S, 167°E) Coral Species: Porites lutea Sampling Resolution: 1.03 mm/sample Sampling technique: Sequential routing or milling Special Note: Sample numbers 137-140, 293-300, are 326-330 do not exist. Abstract We have generated a 40-year-long, monthly stable isotope record from a Porites lutea coral collected offshore of Amedee, New Caledonia (22°S, 167°E) to investigate the relation between sampling resolution in coral isotope studies and retrieval of sea-surface environmental information. We interpret the high correlation between our oxygen isotope record and a twenty-year long sea-surface temperature record at the monthly timescale (r=0.88) to indicate that our coral isotope record is an accurate monitor of environmental conditions offshore of Amedee. The character of the signal and the percent variance explained in the record at the annual band, at the quasi-biennial oscillation band ((QBO) 2.0-2.4 years), and at the El Nino-Southern Oscillation band ((ENSO) 3-8 years) changes little in response to a reduction in sampling density from monthly to bimonthly to quarterly. Similar results have been obtained in a reanalysis of a coral isotope record from Tarawa, Kiribati. Our results indicate that a significant amount of the information obtained from high-density sampling can also be retrieved from lower-density sampling. In particular, bimonthly sampling yields virtually no drop-off in variance explained, and quarterly sampling is satisfactory for resolving interannual and decadal-scale trends in time series. The proposed sampling approach may enable a more rapid filling in of numerous spatial holes in coral sampling sites needed for reconstruction of long-term decadal-scale variations in climate. Methods Coral Sampling We drilled a ~3.45-m-long, 4.5-cm-diameter core nearly down the vertical axis of maximum growth of the Porites lutea coral head in July 1992 near the lighthouse on Amedee Island, New Caledonia. This sample was collected alive at ~3 m water depth, less than a few hundred meters from the location of the daily SST and SSS measurements. The coral was slabbed to a thickness of 5 mm. X-radiographs of the coral slab were taken under exposure conditions of 55 kV, 3 mA, with an exposure time of 20 s. X-radiographs revealed highly regular and well-developed annual density bands. Physical sampling of the coral skeleton is initially guided by the position of a density-band couplet. Hence sampling resolution is often initially defined as the number of samples per density-band couplet (samples/dbc). Forty density- band couplets were observed to occur in 49.25 cm in our New Caledonia coral, yielding an average annual growth rate for this 40-year period of 1.23 cm/year. Growth rate varied from a low 1.08 cm/year to a high of 1.32 cm/year during this interval. We sampled the coral slab at a resolution of 1.03 mm/sample, which yielded 12 samples/dbc (i.e., we drilled and removed powder every 1.03 mm). Drill bit diameter was 1 mm and sample depth was 0.5 mm. The sampling device used in this study was a computer-aided triaxial sampler (CATS). In this device, the hand- held portion of a dentist's drill is clamped in a small vise which is bolted to the movable base of the sampler; the latter is inclined at ~30° from horizontal and controls the in and out movement of the sample. The coral slab is held in place on a motorized stage that is oriented perpendicular to the base, and this stage controls the up and down movement of the sample. The stage that holds the coral slab is itself attached at 90° to a third motorized stage which controls the right to left movement of the sample. The starting and ending coordinates along with the depth and length of each drilled interval are input to the software program that runs the sampling device. Sampling of corals usually involves a sample length on the order of millimeters; however, the CATS device can be used to sample on the order of tens of microns. Used in the fashion described above, the CATS device generates a series of continuously routed samples along a straight pathway, although sampling along curved pathways is also possible (e.g., mollusc shells). Stable Isotope Analyses We generated a 40-year-long stable isotope record to produce a sufficient number of annual and interannual cycles so that we could perform statistical and spectral analyses with the data. Stable isotopic analyses were performed at the University of Michigan. Prior to isotopic analysis, powdered coralline aragonite samples were vacuum roasted for 1 hour at 200°C. Samples were reacted with anhydrous phosphoric acid at 75°C in individual reaction vessels of a CarboKiel carbonate-extraction system coupled to the inlet of a MAT 251 mass spectrometer. Precision (±1 ?) was monitored by daily analyses of a powdered calcite standard (NBS-20) and was better than 0.04‰ for both oxygen and carbon. The average standard deviation of 46 replicate analyses of coral samples is 0.10‰ for oxygen and 0.07‰ carbon. Values are reported in standard ? notation relative to the Chicago Pee Dee belemnite (PDB) standard after correction for the 17O contribution.