Abstract
The Northeast Biogeochemistry and Ecosystem Model (NeBEM) was developed by integrating the Northeast Coastal Ocean Forecast System (NECOFS) with European Regional Seas Ecosystem Model (ERSEM). ERSEM was upgraded to include volume and mass conservation adjustment, total variational diminishing biogeochemical tracer advection scheme, groundwater module, and spatially dependent parameter specifications. NeBEM was first validated through one- and three-dimensional experiments in Massachusetts Bay and then applied to simulate the 2017-2018 physical and biogeochemistry fields in the U.S. region. The model skill assessments demonstrated the NeBEMβs capability of reproducing the seasonal variability of nitrate (ππβ), ammonium (ππ»β), silicate (πππβ), dissolved oxygen (π·π), chlorophyll-a (Chl-a), total alkalinity (ππ΄), dissolved inorganic carbon (π·πΌπΆ), ππ», ππΆπβ, and aragonite saturation state (πΊβ) in a multi-scale region varying from estuaries to continental shelves. Process-oriented studies suggested that the changes in πΊβ was predominantly manipulated by π·πΌπΆ variability in the Middle Atlantic Bight (MAB) and Georges Bank (GB), and π·πΌπΆ plus ππ΄ in the Gulf of Maine (GOM) and Scotian Shelf (SS). Generally, the tidal-mixed areas, such as the western shelf of Nova Scotia, Fundy Bay, Nantucket Shoals, Long Island Sounds, and estuaries connected to the northern GOM, were most susceptible to the OA. From January to April, the inner shelf, especially near rivers, experienced a period of low πΊβ (<1.0), with the largest area occurring in March. During this period, the surface π·πΌπΆ was increased by πΆπβ loading through the air-sea interface via NEC. Over the outer shelf, the total π·πΌπΆ amount was predominantly replenished by the onshore slope-water inflow. The model suggested that warm core rings (WCRs) and eddies (WCEs) played an essential role in enhancing the slope-water transport to the shelf, which accounted for an ~35% increase in the π·πΌπΆ flux. The observed data generally fell within the range of the simulated nππ΄:nπ·πΌπΆ slope. The distribution of simulated nππ΄:nπ·πΌπΆ ratio varied from region to region. The biogeochemical variability of ππ΄ and π·πΌπΆ was primarily controlled by the nitrification/denitrification process in the GOM and MAB, the air-sea πΆπβ exchange in the open sea (OS), and the multiple biogeochemical processes in SS and GB. The influence of climate change on OA was assessed using NeBEM by 1) considering observational data-projected increases in πππ, atmospheric πΆπβ loading, and river discharges, and 2) a downscale climate (NCAR-CESEM-BC)-regional (WRF)-NeBEM coupled model. Both approaches consistently show that regional warming will intensify the anticyclonic residual circulation gyre over GB and the cyclonic gyre in Wilkinson Basin but weaken the cyclonic gyre in Jordan Basin. March will still be the highest probability month with the maximum aea of a yearly minimum πΊβ under changing climate. Increased atmospheric πΆπβ loading against global warming will enlarge yearly minimum πΊβ area during March by 8% in SS, 3% in the GOM, 18% over GB, and increase the probability of having the minimum πΊβ to occur earlier. Warming will increase the probability of having the yearly lowest πΊβ to occur in the bottom layer in the GOM and MAB. Under the climate changes, the primary biogeochemical drivers for ππ΄ and π·πΌπΆ will remain unchanged changed in SS, the GOM, the MAB and OS, even though the contribution of air-sea πΆπβ exchange and NCP will be enhanced. GB is the region with the significant change where the π·πΌπΆ biogeochemical variation will be controlled by air-sea πΆπβ exchange.