Persistent Uncertainties in Ocean Net Primary Production Climate Change Projections at Regional Scales Raise Challenges for Assessing Impacts on Ecosystem Services

Abstract

Ocean net primary production (NPP) results from CO2 fixation by marine phytoplankton, catalysing the transfer of organic matter and energy to marine ecosystems, supporting most marine food webs, and fisheries production as well as stimulating ocean carbon sequestration. Thus, alterations to ocean NPP in response to climate change, as quantified by Earth system model experiments conducted as part of the 5th and 6th Coupled Model Intercomparison Project (CMIP5 and CMIP6) efforts, are expected to alter key ecosystem services. Despite reductions in inter-model variability since CMIP5, the ocean components of CMIP6 models disagree roughly 2-fold in the magnitude and spatial distribution of NPP in the contemporary era, due to incomplete understanding and insufficient observational constraints. Projections of NPP change in absolute terms show large uncertainty in CMIP6, most notably in the North Atlantic and the Indo-Pacific regions, with the latter explaining over two-thirds of the total inter-model uncertainty. While the Indo-Pacific has previously been identified as a hotspot for climate impacts on biodiversity and fisheries, the increased inter-model variability of NPP projections further exacerbates the uncertainties of climate risks on ocean-dependent human communities. Drivers of uncertainty in NPP changes at regional scales integrate different physical and biogeochemical factors that require more targeted mechanistic assessment in future studies. Globally, inter-model uncertainty in the projected changes in NPP has increased since CMIP5, which amplifies the challenges associated with the management of associated ecosystem services. Notably, this increased regional uncertainty in the projected NPP change in CMIP6 has occurred despite reduced uncertainty in the regional rates of NPP for historical period. Improved constraints on the magnitude of ocean NPP and the mechanistic drivers of its spatial variability would improve confidence in future changes. It is unlikely that the CMIP6 model ensemble samples the complete uncertainty in NPP, with the inclusion of additional mechanistic realism likely to widen projections further in the future, especially at regional scales. This has important consequences for assessing ecosystem impacts. Ultimately, we need an integrated mechanistic framework that considers how NPP and marine ecosystems respond to impacts of not only climate change, but also the additional non-climate drivers.

CMIP6 Simulations With the CMCC Earth System Model (CMCC‐ESM2)

“…Although this might appear in disagreement with the negative tendency of both CMIP5 an CMIP6 multi‐model means (see Frölicher et al., 2016; Kwiatkowski et al., 2020), it is worth noting that the uncertainty expressed by the inter‐model standard deviation is very large, in some cases up to five times the mean value. It clearly emerges that global mean primary production, its spatial patterns and its temporal evolution under historical conditions and future climate scenarios is still significantly uncertain (Tagliabue et al., 2021). This is due to the complex interplay between physical drivers, ecosystem interactions, and physiological response of organisms in the lower trophic levels and poorly constrained by scarce observational data and diverging remote‐sensing based approaches (Sathyendranath et al., 2020).…”

Anthropogenic climate change drives non-stationary phytoplankton variance

“…Regional changes in phytoplankton biomass are dominated by changes in diatom and small phytoplankton (Table 1). We aggregate biomass across 11 ecological provinces (Vichi et al, 2011;Tagliabue et al, 2021), and present changes in total and PFT biomass over the RCP8.5 scenario in Table 1. The CESM1-LE simulates the largest decline in total phytoplankton carbon concentration in the Atlantic subpolar (ASP) region, where diatom biomass declines by ⇠80 mmol C m 2 , and small phytoplankton biomass increases slightly (⇠8 mmol C m 2 ).…”

“…We classified the marine environment into 11 ecologically cohesive biomes as in Tagliabue et al (2021), which are a consolidation of the 38 ecological regions defined in Longhurst (2007) using multivariate statistical analysis (Vichi et al, 2011).…”

“…Changes in phytoplankton biomass and its variability in the CESM1-LE from 2006 to 2100 for the 11 ecological provinces defined inVichi et al (2011) andTagliabue et al (2021). Units are mmol C m 2 .…”

Marine phytoplankton resilience may moderate oligotrophic ecosystem responses and biogeochemical feedbacks to climate change

“…The more recent CMIP6 includes models with some of these adaptive mechanisms for nutrient use. Here, it is much more uncertain if NPP will decline with warming and stratification (Kwiatkowski et al 2020;Tagliabue et al 2021), but it is yet unclear if a muted response is linked to the mechanisms we have discussed. At this point, it is impossible to say which of these scenarios are correct.…”