Malindi, Kenya 115 Year Monthly Coral d18O Data ----------------------------------------------------------------------- World Data Service for Paleoclimatology ----------------------------------------------------------------------- NOTE: PLEASE CITE ORIGINAL REFERENCE WHEN USING THIS DATA!!!!! DESCRIPTION: Extended Site, Methods, and Age Model Information All metadata and data for this study can be accessed via: https://doi.org/10.25921/wwrx-z368 ----------------------------------------------------------------------- The coral grew in Malindi Marine Park, Kenya, located on the Kenyan coast, facing the tropical western Indian Ocean. This site is located 15 km south of the mouth of the Sabaki River, which is one of Kenya's major rivers. The site is open to the ocean, and the surface of the colony reaches to a 0.5 m depth from low tide. The SST at the Kenyan coast reaches a maximum of 29.1 degree Celsius in April and a minimum of 25.3 degree Celsius in August, as averaged over the 1951-2002 period in the HADISST data set's monthly SST data (3.5S, 40.5E). Precipitation in this area had two seasonal peaks, with Long Rains beginning from April to May, linked to the southeastern monsoon and Short Rains from October to November linked to the northeastern monsoon. River discharge in the Short Rain period is brought southward to the Malindi coral site by the northeast monsoon winds from December to March, while discharge during the Long Rain period is brought northward by the southeast monsoon from May to November. In October 2002, the coral core was obtained using an air drill from the living colony of Porites lutea. KY-ML-16 was selected for the analysis because it was in the best qualitative condition; the diameter is 55 mm, and the length is 186 cm. The annual bands are clear and continuous, and the bottom of the core is recognized as the 116th annual band from the top, corresponding to 1986. The core was cut into a pair of 5-mm-thick slabs parallel to the growth axis using a high-speed rock saw or a fraise machine. X-radiographs of the slabs were taken to determine the core for the analysis and sub-sampling lines along the maximum growth axis. The X-radiograph of the coral slab revealed clear density bands. A set of high and low- density bands with widths of 10 to 18 mm formed a layered structure. The mean growth rate was 15 mm/year. The boundary from the low to high (white to black in X-radiograph) bands was clear and sharper than that from the high to low (black to white in X-radiograph) bands. The major growth axis was chosen for d18O measurements (sub-sample line). Calcite transformation was examined by X-ray diffraction measurements, which confirmed the absence of any calcite peaks (below 1 percent) in the 3 samples from the top, middle, and bottom parts of this core. We also identified the diagenesis of the primary skeletal aragonite and precipitation of secondary aragonite by petrographic analysis using thin sections from the top, middle, and bottom parts of this core. Dissolution in the thin micritic rim was not identified in this image. Powdered samples for measuring d18O were drilled out with a micro-sampling machine with 1.5 mm interval. Stable isotope analyses were performed using a Finnigan MAT252 isotope ratio mass spectrometer that was equipped with an automated carbonate reaction device (Kiel III) at the Department of Earth and Planetary Science, University of Tokyo. Fifty- to seventy-microgram carbonate samples were reacted with 100 percent phosphoric acid and oxidized to carbon dioxide. All isotope (delta) values are reported with respect to Pee Dee Belemnite (PDB). A laboratory working standard (GS17: d18O = -2.15 per mil, calibrated by an international carbonate standard of NBS-19 (RM8544, d18O = -2.2 per mil PDB published by IAEA)) was used to translate raw measurement results into the PDB scale. The external precision of the powdered carbonate standard is 0.03 per mil (1 sigma) and coral d18O external reproducibility that we measured in this study is about 0.03 per mil (n=28), thus total error contained in coral d18O value is 0.04 per mil. In this study, we apply the following chronology with two anchor points to convert the coral record into a monthly resolution age model. The high- d18O peak corresponds to August, when SST is the lowest and precipitation is low, and the boundary from the low- to high-density band corresponds to November. This density band is followed by a fluorescent band under UV related to the Sabaki river discharge that reaches the coral site in December with the northeast monsoon. Because the coral growth rate changes slightly from year to year, the width of annual bands ranges from 10 to 18 mm (the sub-sampling number ranges from 7 to 16). The annual mean growth rate is 15 mm, and the mean sub-sampling number is 10 2. The typical uncertainty caused by the difference in the sub-sampling number from a set of density bands is 1-2 months. However, the sub-sampling numbers are exact in only a limited number of years. Low resolution caused by a narrow annual band is observed in 1896, 1900, 1950 and 1978. A single anchor point (August) is used for the age model only in 1890, 1915 and 1987 because the low- to high-density boundary (November) is not clear in the annual band of these years.