Long-term mesoscale variability of modelled sea-ice primary production in the northern Baltic Sea

Abstract

We describe a new ocean-sea ice-biogeochemical model, apply it to the Bothnian Bay in the northern Baltic Sea for the time period 1991-2007 and provide the first long-term mesoscale estimates of modelled sea-ice primary production in the northern Baltic Sea. After comparing the available physical and biogeochemical observations within the study area and the time period investigated with the model results, we show the modelled spatial, intra-and interannual variability in sea-ice physical and biogeochemical properties and consider the main factors limiting ice algal primary production. Sea-ice permeability in the studied area was low compared with the polar oceans, which appeared to be a major reason for the generally low primary production rates. Although the sea ice was less saline in the northernmost parts of the basin, these parts were characterized by sea ice with a larger amount of habitable space, higher levels of photosynthetically active radiation and increased macronutrient availability near the coast, which favoured higher algal growth rates. Other parts of the southern central basin were mostly co-limited by less favourable light conditions (i.e., earlier ice breakups associated with fewer sunlight hours) and lower seawater macronutrient concentrations than in the coastal zones. Although a change towards milder winters (i.e., reduced ice cover, thickness and length of the ice season) was previously detected on a half-century timescale and could partly be seen here, analysis of the temporal evolution of sea-icebiogeochemicalpropertiesshowednosignificanttrendsovertime,thoughthesepropertieswere characterized by large interannual variability.

Drivers of Ice Algal Bloom Variability Between 1980 and 2015 in the Chukchi Sea

“…They showed that shorter ice season corresponded to shorter bloom duration with lower peak bloom concentrations and suggested that even though ice algal communities experienced optimal growth conditions (light, space, and nutrients), they did not have enough time to reach high bloom concentrations under the early melt condition. Two modeling studies that compared ice algal production from early and late ice melt years (Tedesco et al, ; this study) support the findings of the idealized sensitivity analysis in Tedesco and Vichi ().…”

“…In addition to controlling bloom onset and growth rates of the spring ice algal bloom, snow and, to a lesser extent, sea ice thickness, in concert with atmospheric conditions (e.g., air temperatures and wind conditions), impact the timing of ice melt, the loss of the productive bottom layer of ice algae from the sea ice and thus the decline of the ice algal bloom from its peak concentration (Lavoie et al, ; Pogson et al, ; Tedesco & Vichi, ; this study). Our results and others (Haecky & Andersson, ; Duarte et al, ; Tedesco et al, ) suggest that ice algal blooms peak and persist in the presence of some basal melting, depending on the definition of a “bloom period.” Using our threshold value of 5 mg C m −2 to define a bloom, sensitivity analyses showed that productive bottom ice layers disappeared approximately a week to a month earlier under thin snow than under thick snow, regardless of ice thickness. While depth‐integrated biomass sharply decreased during this period of bottom melt, values remained above the bloom threshold until ice retreat.…”

Non‐Redfieldian Dynamics Explain Seasonal pCO2Drawdown in the Gulf of Bothnia

“…However, the primary production data we use have not been filtered and should contain the production of both POC and DOC, making the good agreement between the modeled POC production and observed production intriguing. Tedesco et al (2017) obtained similar results with BFM when modeling sea-ice primary production in the Bothnian Bay. Their modeled production of POC agreed well with the total organic carbon production, measured with the same method as the primary production data used in this study.…”

“…Their modeled production of POC agreed well with the total organic carbon production, measured with the same method as the primary production data used in this study. In the inter-calibration study from the Baltic Sea, where different 14 C methods were compared (Andreasson et al, 2009), it was shown that the method used to measure the TOC production rates, which are used in this study and in Tedesco et al (2017), produced similar, or even lower primary production rates, compared to another method measuring POC production rates. No clear explanation for this inconsistency was given, and it was suggested that more investigations in this matter is needed.…”

“…Here we test, for the first time, a flexible-stoichiometry model coupled to a 3-D circulation model of the Gulf of Bothnia, to examine how stoichiometric flexibility and extracellular DOC production can explain seasonal nutrient and CO 2 dynamics at different contrasting sites. Flexible-stoichiometry models have previously been applied for primary production studies in sea-ice in the Gulf of Bothnia and the Gulf of Finland (Tedesco et al, 2010(Tedesco et al, , 2017Thomas et al, 2017). For pelagic biogeochemistry, flexible-stoichiometry models have been used for 1-D studies at specific locations (Kreus et al, 2015a(Kreus et al, , 2015bVichi et al, 2004), and for modeling cyanobacteria in a dynamic 3-D model (Kuznetsov et al, 2008), but only in the central Baltic Sea and the Gulf of Finland.…”