An ecosystem modeling study of spatio-temporal variations of phytoplankton distribution in the Okhotsk Sea

Abstract

A three-dimensional ecosystem–physical-coupled model is applied to the Okhotsk Sea. Factors that determine the spatial distribution of phytoplankton in the Okhotsk Sea in autumn and spring are analyzed, and the effects of sea ice on the spring bloom are also discussed. One of the most important factors determining the spatial distribution of phytoplankton in autumn is regional variation in mixed layer depth. The model can explain the spatio-temporal variation of chlorophyll-a concentration in the Okhotsk Sea during the spring blooms in 1997 and 2001. The start of the spring bloom in the Okhotsk Sea depends on the light environment. By controlling the light intensity in sea surface water, sea ice controls the timing of the spring bloom.

Introduction

The Okhotsk Sea is one of the most biologically productive regions in the world, and it supports high fisheries production (Fig. 1). It has been reported that a major fraction of the phytoplankton in this sea are diatoms (e.g., Hanzawa et al., 1981). Previous observations revealed maximum diatom cell numbers in spring and minima in autumn (Ohwada, 1957; Hanzawa et al., 1981). Saitoh et al. (1996) showed that the spring bloom was formed between April and May based on the monthly coastal zone color scanner (CZCS)-chlorophyll (Chl) imagery from 1978 until 1986. Shiomoto et al. (1998) suggested that the Chl-a concentration is lower (0–3 mg/m³) in autumn than during the spring bloom (2–10 mg/m³; Nishihama et al., 1989), and that grazing by macrozooplankton (mainly Copepods) may contribute to variations in Chl-a concentration.

The Okhotsk Sea is well known as one of the southern most seasonal sea ice zones in the Northern Hemisphere. In winter, sea ice formation begins around Shantarsky Bay at the end of November, and sea ice extension reaches its maximum in late February or March. Most of the sea ice disappears by May. Sea ice in the Okhotsk Sea is generally advected southward by the prevailing northerly or northwesterly winds. In the southwestern part of the sea, thick “first-year” ice is primarily advected by the East Sakhalin Current (ESC). Some of the ice is advected toward the offshore warm region and melted, making the surface layer water fresher. Some of this water is then frozen again by cooling. This process leads to formation of “new ice” at ice edges (Ohsima et al., 2001). Therefore, the seasonal change of sea ice volume depends on climatic conditions, and has large interannual variations. The sea ice has been considered to play an important role in the high production at the ice edge in the Okhotsk Sea. A large number of studies paid attention to the spring bloom after the sea ice melting (e.g., at the Bering Sea; Niebauer and Alexnder, 1985). However, there are a few studies (e.g., Matsumoto et al., 2004) that discuss the effect of sea ice on the spring bloom in the Okhotsk Sea.

Recently, numerical models have been applied to examine what controls phytoplankton blooms in the Northern Pacific. Kawamiya et al., 1995, Kawamiya et al., 1997 developed a vertical one-dimensional ecological–physical-coupled model and explained the role of zooplankton grazing pressure in the HNLC at OS Papa. Since then, KKYS (the ecological part of their model) has been applied to several different oceanic regimes, the Bermuda Atlantic Time-Series (BATS) site and Japanese coastal regions. The series of studies showed that with changes in only a few parameters, the model could reproduce nitrate and Chl concentrations in those regimes. KKYS is a simple ecological model, composed of six compartments. Based on KKYS, Kishi et al. (2001) developed an eight-compartment ecological model by dividing phytoplankton and zooplankton each into two size categories. Moreover, building upon their model, North Pacific Ecosystem Model Understanding Regional Oceanography (NEMURO) was developed by the PICES CCCC-MODEL TASK TEAM (Eslinger et al., 2000) during a workshop held in January 2000, in NEMURO, Hokkaido, Japan. NEMURO was developed to simulate the lower trophic ecosystem of the Northern Pacific Ocean. Fujii et al. (2002) showed that NEMURO can reproduce the transition of the seasonal dominant species at station KNOT (44°N, 155°E) in their one-dimensional model.

A few models have been made and applied to algal bloom under sea ice. Fennel et al. (2003) made an ecosystem model on seasonal ice zone of the Southern Ocean. In their model, the effects of the flux of light under sea ice, which is used for photosynthesis, and momentum through sea ice, which is used as physical forcing as a boundary condition, are included. But they do not take into consideration the time-dependent change of the thickness of sea ice, nor do they focus on the effects of sea ice on photosynthesis of planktons.

The purposes of this paper are to examine the factors that control a phytoplankton growth in autumn and spring, and to discuss the effect of sea ice on the spring bloom in the Okhotsk Sea with the aid of the ecosystem model. In 2.1 Modeling the autumn bloom, 3.1 Autumn bloom, we present the application of NEMURO coupled with a three-dimensional physical model to the Okhotsk Sea in autumn. We focused on the factors that determine the spatial distribution of phytoplankton in the Okhotsk Sea in autumn. In 2.2 Modeling the spring bloom, 3.2 Spring bloom, we present a modified model including sea ice (including its temporal changes) and discuss the effects of sea ice on the spring bloom, paying attention to the factors that determine the spatial distribution of phytoplankton in the Okhotsk Sea in spring.