Browsing by Author "Sayani, HR"

Now showing 1 - 1 of 1
  • No Thumbnail Available
    Item
    The CoralHydro2k Database: a global compilation of coral δ18O and Sr/Ca records for reconstructing tropical hydroclimate over the common era
    (American Geophysical Union (AGU), 2021-12-18) Sayani, HR; Walter, R; Felis, T; Cobb, KM; Abram, NJ; Atwood, AR; Arzey, A; Brenner, LD; Dassie, EP; DeLong, KL; Ellis, B; Goodkin, N; Hargreaves, J; Kilbourne, KH; Krawczyk, H; Fischer, MJ; Murty, SA; Moore, AL; Ramos, RDP; Reed, E; Samanta, D; Zinke, J
    Shallow-water corals provide annual to subannual -resolution climate reconstructions from normally data-scarce locations in the tropical to subtropical oceans, enabling us to extend the modern-day observational records back to the preindustrial era, contextualize anthropogenic climate change, and improve the skill of future climate projections. The majority of these coral-based reconstructions utilize oxygen isotope ratios (δ18O), a proxy that tracks the combined change in sea surface temperature (SST) and the oxygen isotopic composition of seawater (δ18Osw) and/or strontium-to-calcium ratios (Sr/Ca), which primarily track SST variability. Paired coral δ18O and Sr/Ca records can be combined to isolate δ18Osw variability, which like salinity reflects changes in the local hydrologic budget. Recently, the PAGES Ocean2k project used published coral records to reconstruct regional SST variability across the tropical oceans (Tierney et al., 2015, Abram et al., 2016). Building on this work, the PAGES CoralHydro2k team has compiled a more comprehensive, machine-readable, and metadata-rich network of paired coral δ18O and Sr/Ca records to help facilitate tropical hydroclimate reconstructions across recent centuries. The CoralHydro2k database currently contains 227 coral proxy records from 120 unique locations that are organized into seven tiers based on the availability of paired proxy data, temporal coverage, and record resolution. The metadata for the new database follows PACTs 1.0 recommendations (Khider et al., 2019), and the database is built using LiPD (McKay and Emile-Geay, 2016) with available R, MATLAB, and Python serializations. Here we describe the structure and spatiotemporal characteristics of this new database and outline a crowdsourced data-submission process to ensure active-curation of records and future updates to the database.