Pelagic marine ecosystem responses to Plio-Pleistocene Climate Change

Photosynthetic nannoplankton play a fundamental role in both ocean food webs and the global carbon cycle. Marine phytoplankton, of which coccolithophore and diatoms are the dominant groups, are also responsible for around half of global primary production (Field et al., 1998). Extent of suitable habitat for these two groups has varied significantly over geological timescales (Lazarus et al., 2014; Gibbs et al., 2016) and anthropogenic climate change is expected to cause large-scalechanges in phytoplankton distributionover the coming centuries (Rost & Riebesell, 2004; Boyd et al., 2016).Future climate states are expected to fall outside of the range of those observed in historic times, but suitable analogies for these conditionsexist in the geological record (Burke et al., 2018). During the mid-Pliocene Warm Period(mPWP, 3.29-2.97Ma), atmospheric CO2 concentrations were comparable to modern values of ~400ppmv, and mean global temperature was around 3C warmer than preindustrial (Haywood & Valdes, 2004). Hence, the mPWP is potentially a good analogy for near-future Earth conditions under Relative Concentration Pathway (RCP) 4.5, an 'intermediate' greenhouse gas emissions scenario (IPCC, 2014; Burke et al., 2018). During the Last Interglacial (LIG, 130-80ka), global temperatures were up to 1C warmer than preindustrial, and global mean sea level around was 4-9m higher(Dutton et al., 2015; Fischer et al., 2018), providing an analogue for a less extreme future climate ina lower RCP scenario
Objectives
1. Quantify the absolute abundance of diatoms, coccolithophores and ichthyoliths over the past 3.5Ma, with a focus on the mPWP and LIG2.
2. Test for covariance between the three groups and climate
3. Model pelagic food webs for both past and future warm periods
4. Predict changes in fish biomass and abundance under future anthropogenic climate change
Using sediment core samples recovered during IODP expeditions 306 and 162; we will investigate the ecological development of two North Atlantic locations (Fig. 2.) over the last 3.5 Ma. These samples are freely available from the IODP core repository at the University of Bremen.We will carry out high-resolution sampling for both the mPWP and LIG at 2ka resolution, reducing to 30ka for background times. Different methods will be employed to calculate the absolute abundance of each phytoplankton group: the SYRACO software will make automated counts of coccolithophores (Beaufort & Dollfus, 2004); and the method of Warnock & Scherer (2014) for diatoms. Fish abundance through the Plio-Pleistocene will be quantified using ichthyoliths, the microscopic remains of fish(e.gSibert et al., 2020). Ichthyoliths and nannoplankton can be obtained from the same sediment samples.We will use stochastic differential equations (SDEs) to integrate abundance data from the three groups withexisting proxies for palaeoclimate (e.g. Lisiecki & Raymo, 2005). This study will assess the possibility of future ecosystem collapse in commercially-important pelagic fisheries using the NEMO-ERSEM (Baretta et al., 1995; Madec, 2008), FABM (Bruggeman & Bolding, 2014) and MIZER (Blanchard et al., 2014) food web modelssimulated for the mPWP, LIG, and late-20thcentury climate scenarios.