Carbon cycling and climate change during the last glacial cycle inferred from the isotope records using an ocean biogeochemical carbon cycle model

Abstract

We construct a simple, vertical one-dimensional ocean model coupled with carbon biogeochemical cycle model in order to study the carbon cycle in association with the glacial–interglacial cycle. The model is time-integrated by using the atmospheric CO 2 concentration and the marine carbon isotope record of the surface and the deep water during the last 130,000 years. Temporal variation of the mean upwelling rate, productivity of organic carbon, productivity of carbonate carbon, and the terrestrial carbon storage are obtained from the mass balances of the total carbon, the 13C of dissolved inorganic carbon (DIC), the total alkalinity, and the dissolved phosphate. Variation of the terrestrial biomass size is similar to the δ 18O curve obtained from deep-sea sediments, suggesting that the terrestrial biomass change has been influenced by climate change. As far as we know, this is the first attempt to reconstruct the temporal variation of the terrestrial carbon storage during the last 130,000 years. The obtained terrestrial carbon storage at the last glacial maximum is within the range estimated in previous studies. Vertical mixing of the ocean is weakened and the bioproductivity decreases globally during the glacial intervals. Although the obtained variations of organic carbon and carbonate productivities are similar to each other, remarkable increases at the deglaciations are found only in carbonate productivity. The variation of nutrient concentration in the intermediate water differs from that in the deep water, while the variations of the total alkalinity in the intermediate and the deep water are in phase. The glacial ocean is characterized by higher alkalinity throughout the water column and lower gradient of DIC and nutrient. This lower gradient, resulting from lower upwelling rate and organic carbon productivity, results in a redistribution of DIC and nutrient between the intermediate and the deep water, but not change the total stock of DIC. On the other hand, higher alkalinity in the glacial intervals results in an increase in the total stock of dissolved inorganic carbon in the ocean because of increased solubility of atmospheric CO 2 from higher alkalinity throughout the water column. Temporal variations in alkalinity are characterized by abrupt decreases corresponding to the deglaciations every 100,000 years, resulting from a remarkable increase in carbonate productivity. This might indicate a relationship with the variation of atmospheric CO 2 level, which shows a gradual decrease through the glacial interval and an abrupt increase at the deglaciations. Higher alkalinity in the glacial intervals would be a result of an integral effect of the lower net carbonate burial rate during those periods. On the other hand, the redistribution of nutrients and dissolved carbon between the intermediate and the deep water shows shorter-term fluctuations. This would not change the atmospheric CO 2 level significantly, probably because the integral effect of net carbonate burial may not be enough to change the total alkalinity over the water column.