Seasonal dependency of Arctic marine phytoplankton response under greenhouse warming in CMIP5 earth system models

Hyung-Gyu Lim, AOS program in Princeton Univ / NOAA-GFDL, Princeton, NJ, United States, Jong-Seong Kug, Pohang University of Science and Engineering, Pohang, South Korea, Charles A Stock, NOAA/GFDL, Princeton, NJ, United States and John P Dunne, NOAA Geophys Fluid Dynamic, Princeton, United States
Under greenhouse warming, surface warming, sea ice melting, and related positive ice-albedo feedback in the Arctic climate reduce oceanic mixing. The nitrate supply, major limiting factors among nutrients in Arctic Ocean, could be exhausted by enhanced stratification in upper ocean, which processes influence the marine phytoplankton responses against Arctic warming. In this study, we investigate the available dataset of sea ice, nitrate, and chlorophyll concentrations among CMIP5 ESMs of historical runs and RCPs scenarios. As discussed in Cabre et al. (2015), the annual mean Arctic chlorophyll concentration of CMIP5 ESMs simulated to be decreased under Arctic warming. Although sea ice and nitrate concentrations were exhausted among all calendar month, we found that future chlorophyll changes show the increased patterns of spring chlorophyll and decreased patterns of summer chlorophyll. This implies that the chlorophyll concentration responses have a strong seasonal dependency due to different co-limiting condition of light and nutrient. In addition, these patterns of chlorophyll with seasonal dependency were intensified by increased human-influence from RCP 2.6 to RCP 8.5 scenarios. However, chlorophyll responses among CMIP5 ESMs had intermodel diversity as discussed in Cabre et al. (2015), which could be based on different initial condition of nitrate concentration (Vancoppenolle et al. 2013). We investigated that the diversity of initial nitrate can control the timing of oligotrophic onset in May, that enhances the diversity of annual mean chlorophyll responses, (r=0.7 for nitrate initial condition and future chlorophyll responses in RCP8.5). The group of faster transition in May from light limitation to oligotrophic condition (GISS and IPSL) tends to underestimate the surface nitrate concentration compared to World Ocean Atlas climatology. GISS model simulate to be decreased in all calendar month due to poor nitrate basic conditions. Excepting for GISS models, CMIP5 ESMs have been consensus: increased chlorophyll in March and April; decreased chlorophyll in July and August. The phenology of phytoplankton would experience the shift of limiting condition from light to nitrate among May to July, which determine the changes of phytoplankton phenology in future Arctic Ocean.