Clim. Past, 12, 1519–1538, 2016 www.clim-past.net/12/1519/2016/ doi:10.5194/cp-12-1519-2016 © Author(s) 2016. CC Attribution 3.0 License. The PRISM4 (mid-Piacenzian) paleoenvironmental reconstruction Harry Dowsett1, Aisling Dolan2, David Rowley3, Robert Moucha4, Alessandro M. Forte5,6, Jerry X. Mitrovica7, Matthew Pound8, Ulrich Salzmann8, Marci Robinson1, Mark Chandler9,10, Kevin Foley1, and Alan Haywood2 1Eastern Geology and Paleoclimate Science Center, United States Geological Survey, Reston, VA 20192, USA 2School of Earth and Environment, University of Leeds, Leeds LS2 9JT, UK 3Department of the Geophysical Sciences, University of Chicago, Chicago, IL 60637, USA 4Department of Earth Sciences, Syracuse University, Syracuse, NY 13244, USA 5Department of Geological Sciences, University of Florida, Gainesville, FL 32611, USA 6GEOTOP, Université du Québec à Montréal, Montréal QC, H3C 3P8, Canada 7Department of Earth and Planetary Sciences, Harvard University, Cambridge, MA 02138, USA 8Department of Geography, Faculty of Engineering and Environment, Northumbria University, Newcastle upon Tyne NE1 8ST, UK 9Center for Climate Systems Research at Columbia University, New York, NY, USA 10NASA Goddard Institute for Space Studies, New York, NY, USA Correspondence to: Harry Dowsett ([email protected]) Received: 16 March 2016 – Published in Clim. Past Discuss.: 21 March 2016 Revised: 22 June 2016 – Accepted: 23 June 2016 – Published: 13 July 2016 Abstract. The mid-Piacenzian is known as a period of rel- 1 Introduction ative warmth when compared to the present day. A compre- hensive understanding of conditions during the Piacenzian The Pliocene, specifically the mid-Piacenzian (3.264 to serves as both a conceptual model and a source for boundary 3.025 Ma), has been a focus of synoptic paleoclimate re- conditions as well as means of verification of global climate search for the past 25 years. The mid-Piacenzian is the most model experiments. In this paper we present the PRISM4 recent time in Earth’s past to have exhibited climates not un- reconstruction, a paleoenvironmental reconstruction of the like those projected for the end of the 21st century (Dowsett, mid-Piacenzian ( ∼ 3 Ma) containing data for paleogeogra- 2007a; IPCC, 2013). With widespread recognition by most phy, land and sea ice, sea-surface temperature, vegetation, experts that anthropogenic drivers have been the dominant soils, and lakes. Our retrodicted paleogeography takes into cause of observed warming since the mid-20th century (Ver- account glacial isostatic adjustments and changes in dynamic heggen et al., 2014), and surface temperatures projected to topography. Soils and lakes, both significant as land sur- rise during the 21st century under all emission scenarios face features, are introduced to the PRISM reconstruction (IPCC, 2013), understanding the Pliocene climate has taken for the first time. Sea-surface temperature and vegetation re- on new importance. While not a temporal analog to future cli- constructions are unchanged but now have confidence assess- mate conditions, there is much to learn about the magnitude ments. The PRISM4 reconstruction is being used as bound- and spatial distribution of processes from this, in essence, ary condition data for the Pliocene Model Intercomparison natural climate laboratory (Crowley, 1990). Project Phase 2 (PlioMIP2) experiments. Borrowing heavily from methodology used by the CLIMAP (Climate/Long-Range Investigation, Mapping and Predictions) for the reconstruction of the last glacial max- imum (LGM), the US Geological Survey (USGS) and col- laborators initiated a large-scale data collection and interpre- tation project: Pliocene Research, Interpretation and Synop- Published by Copernicus Publications on behalf of the European Geosciences Union. 1520 H. Dowsett et al.: The PRISM4 (mid-Piacenzian) paleoenvironmental reconstruction tic Mapping (PRISM) (see Dowsett, 2007a; CLIMAP, 1976). tions for climate model simulations. Confidence in paleoen- The first PRISM reconstructions consisted of data sets for vironmental reconstruction is receiving much interest in the sea-surface temperature (SST), vegetation, land ice distribu- paleoclimate modeling community, and changes necessary to tion and volume, sea-ice cover, land elevation, and sea level address uncertainty are discussed. We also discuss both ter- (Dowsett et al., 1994, 1999). Since its inception in the late restrial and marine high-resolution data being developed as 1980s, PRISM has grown in both size and scope, evolving part of PRISM4 to better understand Piacenzian paleoclimate through four global-scale reconstructions, each one adding a variability. new component of the Piacenzian paleoenvironment or im- proving upon methods to increase confidence in the recon- structions (see Dowsett, 2007; Dowsett et al., 1994, 1996, 2 PRISM chronology 1999, 2010, 2013). These reconstructions serve two pur- poses: to assemble the best information possible to provide a Previous PRISM paleoclimate reconstructions were based conceptual model of the Piacenzian paleoenvironment and to upon a time slab concept developed early in the project to provide the data as quantitative, gridded arrays to the paleo- overcome the inability to make long distance high-resolution climate modeling community for global climate model sim- correlations (Dowsett and Poore, 1991). This was achieved ulations. for SST data and land surface cover reconstruction by av- Early modeling efforts to simulate a warmer Pliocene eraging the warm phase of climate or selecting the vegeta- Earth used atmospheric general circulation models (AGCMs) tion representing the wettest and warmest period within the initiated with higher than preindustrial values of CO2 in the defined temporal slab at each locality (Dowsett et al., 2009, atmosphere (Chandler et al., 1994; Dowsett, 2007b; Hay- 2013; Salzmann et al., 2008, 2013). The PRISM reconstruc- wood et al., 2000; Sloan et al., 1996). These simulations tion therefore approximates the average “interglacial” con- used prescribed boundary conditions in the form of PRISM dition at each site. The initial slab was a ∼ 300 ky interval synoptic reconstructions of SST, land cover, and topography ranging from within the Mammoth to just above the Kaena (Dowsett et al., 1996, 1999, 2007a). reversed-polarity subchrons of the Gauss normal-polarity PRISM3 was the basis for the Pliocene Model Intercom- chron (Fig. 1). This and a number of biochronologic events parison Project (PlioMIP), the first Pliocene multi-model allowed approximate identification of the slab position in comparison using fully coupled atmosphere–ocean gen- both marine and terrestrial settings (see Dowsett, 2007a, and eral circulation models (AOGCMs) (Haywood et al., 2010, references cited therein). 2011). Much effort was put into improving the land sur- The PRISM3 time slab (Dowsett et al., 2010) was refined face cover and ocean components for this penultimate recon- to the interval between 3.264 and 3.025 Ma, easily identified struction (Dowsett et al., 2010). PRISM3 introduced a multi- in marine sequences by its position between oxygen isotope proxy SST reconstruction, a deep-ocean temperature recon- enrichments M2 and G20 (Fig. 1) in the LR04 chronology of struction, and for the first time a biome classification of the Lisiecki and Raymo (2005). The 3.264 to 3.025 Ma range is Pliocene land surface. still appropriate for most components of the PRISM4 recon- The PRISM3 reconstruction also addressed uncertainty in struction (see discussion). both terrestrial and marine paleoclimate estimates (Dowsett New PRISM4 time-series data use a nested chronol- et al., 2012, 2013a; Salzmann et al., 2013). Marine and ter- ogy since various components of the paleoenvironmental restrial data–model comparisons (DMCs) presented in IPCC reconstruction can achieve different maximum resolution. AR5 documented robust large-scale features of the Piacen- PRISM4 marine time series are generated from localities zian climate but at the same time identified areas of disagree- possessing the characteristics needed for precise orbital- ment between data and models, highlighting the need for ad- scale correlation. The challenges of terrestrial age dating and ditional assessments of confidence for both reconstructed and stratigraphic control of many terrestrial archives currently simulated environments (Haywood et al., 2013; Salzmann et limit expansion of the terrestrial reconstruction of PRISM4 al., 2013; Dowsett et al., 2012, 2013a). time series in all but a few locations. PRISM data have also been used to study diversity and The PRISM time slab or PRISM “interval”, as defined ecological niche changes in planktic foraminifers, mollusks, above, occurs within the Piacenzian Age. The Piacenzian is and fish in the face of profound global warming (e.g., Ya- roughly equivalent to the Gauss normal-polarity chron (∼ 3.6 suhara et al., 2012; Saupe et al., 2014; Jacobs, 2015). to 2.6 Ma). Prior to 2010, the Pliocene Epoch included the In this paper we document and summarize our most recent Zanclean, Piacenzian, and Gelasian ages. Thus, it was com- reconstruction, PRISM4. PRISM4 is a conceptual model of mon practice to refer to the PRISM interval as the mid- mid-Piacenzian conditions for which major efforts have been Pliocene. Changes enacted by the International Commis- focused on improving the paleogeographic and cryospheric sion on Stratigraphy revised the placement of the Pliocene– components. New topography/bathymetry, Greenland ice Pleistocene boundary from the base of
Details
-
File Typepdf
-
Upload Time-
-
Content LanguagesEnglish
-
Upload UserAnonymous/Not logged-in
-
File Pages20 Page
-
File Size-