The role of snowmelt runoff on the ocean environment and scallop production in Funka Bay, Japan
© Nakada et al.; licensee Springer. 2014
Received: 4 December 2013
Accepted: 2 November 2014
Published: 13 December 2014
This study investigated the role of snowmelt runoff on water circulation, water mass formation, and the production of cultured scallop larvae, as a part of a land-sea linkage, by analyzing hydrological data in conjunction with nutrient data and by conducting sensitivity experiments based on a coupled land-sea model of Funka Bay, Japan, a typical semi-enclosed bay. A comparison between observed data and the simulated runoff showed that, using newly compiled datasets of nutrient concentrations in rivers and groundwater, the model was sufficiently accurate to estimate the terrestrial dissolved inorganic nitrogen (DIN) flux from the river and submarine groundwater discharges (SGDs). The average volume flux from the SGDs accounted for 26% of the riverine runoff flux. The DIN flux from SGDs accounted for a maximum of 40% of the total DIN loading to the bay before the snowmelt period. Sensitivity experiments using an ocean simulation indicated that the freshwater flux supplied by snowmelt runoff not only enhances clockwise circulations along with upwelling along the coast, but also modifies the distributions of wintertime water masses in the bay. However, the snowmelt runoff has little effect on larvae transport since wind forcing, rather than riverine buoyancy, dominates the circulation patterns. The annual density of scallop spat was highly correlated with snowmelt runoffs associated with high DIN concentrations, which suggested that riverine nutrients can increase the biomass of phytoplankton in near-shore seas and improve food availability for scallop spawners, resulting in increased egg production in March to April. Therefore, the nutrient flux from agricultural source areas through the large snowmelt runoff has an important role in larvae production. Land-sea linkages need to be identified to design sustainable and synergetic systems of aquaculture and agriculture for the integrated management of coastal regions.
KeywordsSnowmelt runoff Scallop aquaculture Land-sea linkage OGCM Nutrient flux Submarine groundwater discharge
Fisheries and aquafarms located in coastal seas can be influenced by terrestrial inputs, such as freshwater, nutrients, and momentum through inflow from rivers (Kimmerer ). Aquacultures operate mainly in coastal oceans and currently contribute approximately 48% of the aquatic animal food for human consumption (FAO ; Bondad-Reantaso et al. ). The fish and shellfish populations of coastal seas are also influenced by the input of terrestrial nutrients through submarine groundwater discharge (SGD) (Liu et al. ; Nakada et al. ). Large snowmelt runoffs, along with high nutrient fluxes, can affect fishery production in coastal oceans. However, no study has yet investigated how snowmelt runoff influences the aquaculture production in coastal oceans. Estimation of nutrient loading through rivers and SGDs around the snowmelt season is essential to investigate the role of snowmelt runoff in fishery production.
Recently, a coupled land-sea model was developed to generate an operational ocean prediction system in the bay to inform fishermen of the ocean state in real time (Nakada et al. [2012a]). The model reproduced realistic river runoffs and oceanographic features, such as the clockwise circulation, that agreed quantitatively with observations. New comprehensive analyses can expect to use high-resolution datasets produced by coupled land-sea simulations that are validated by observational datasets. Several studies have employed this approach, using the coupled model to investigate the oceanic dynamical behaviors of sub-mesoscale eddies in coastal oceans (e.g., Zhao et al. ).
In this study, we investigated the role of snowmelt runoff in water circulation, water mass formation, and the production of scallop larvae in the bay to reveal a land-sea linkage using hydrological data in conjunction with nutrient data and by conducting sensitivity experiments based on the coupled land-sea model. The paper is organized as follows: the second section describes the details of numerical methods and nutrient analyses of sampled waters. The third section presents the simulated results and their validations and describes sensitivity experiments for the snowmelt runoff. The fourth section discusses the relationships between the terrestrial runoff and oceanic environment, in terms of nutrient supply and larval transport related to scallop production. The fifth section provides the conclusions.
Overview of the coupled land-sea model
River discharges were predicted using a distributed-tank model, based on radiation, heat, and water mass budgets, by calculating physical components including the solar insolation (S↓), downward longwave radiation flux from the atmosphere (L a ↓), upward longwave radiation flux from the ground surface (L g ↑), sensible heat flux (H), and latent heat flux (IE), which determines rainfall and snowfall (T r ), precipitation (P), evapotranspiration (E), and snowmelt (M) (see Kondo and Watanabe () and Kondo and Yamazaki () for detailed formulae). These budgets can be calculated using the five meteorological properties (rainfall, air temperature [T a ], wind speed, cloud cover, and relative humidity) from the grid point value datasets of the meso-scale model (GPV-MSM) provided by the Japan Meteorological Agency on an hourly basis. The river discharge, Q t , generated by the net water input (P + M − E) is predicted for each grid of the distributed-tank model (Figure 5). The model has three serial tanks, represented as n = 1 to 3, ten model coefficients (A n , B n , and S n ) for the side outlets (Q n ) and bottom outlets (L n ), and specified depth thresholds of water levels (Z n ) in each tank. If snow covers the ground surface (Figure 5) and the snow surface temperature satisfies T s < T a (where T s denotes the ground surface temperature), the snowmelt, M, is calculated by a radiation and heat balance.
A three-dimensional OGCM developed at Kyoto University was employed using the hydrostatic and Boussinesq approximations and a realistic bottom topography. The OGCM implemented a hybrid σ-z vertical coordinate system to improve simulation of the free surface motion of the ocean. To enhance the representation of upper ocean circulation further, this model adopted several sophisticated parameterizations, including a turbulence closure scheme for the mixed layer parameterization (Noh ), UTOPIA (Leonard et al. ) for horizontal advection, and an isopycnal mixing scheme (Gent and McWilliams ; Griffies ). To improve the realistic reproduction of salinity distributions in the coastal ocean or around the aquafarms in Funka Bay, we employed a three-step nesting or downscaling method (Figure 3) that implemented a large-scale, four-dimensional variational data assimilation model (Ishikawa et al. ). The assimilation model (NEST0) covered the western half of the North Pacific at horizontal resolutions of (1/6)° and (1/8)° for longitude and latitude, respectively, and provided accurate boundary conditions for two nested models embedded in the western region. The first nested model (NEST1) without assimilation covered the northwestern North Pacific and centered on the mixed water region at medium resolutions of (1/18)° and (1/24)° for longitude and latitude, respectively. The second nested model (NEST2) without assimilation was resolved with the finest horizontal resolutions of (1/54)° and (1/72)° for longitude and latitude, respectively, to reproduce the small-scale physical features within coastal zones. Note that 78 vertical levels were spaced between 4 m apart near the sea surface and 500 m apart at the bottom. The boundary conditions used in each nested model were provided from upper-level (larger-scale) modeling results using the nesting technique by Oey and Chen (). Assimilated elements in our study were satellite-derived sea surface temperature, sea surface height data, and in situ observation data on temperature and salinity (see Ishikawa et al. () for more details). The river runoff derived from the HaRUM was converted to freshwater flux and input into the meshes closest to the coast of Funka Bay. The freshwater flux was reflected by a decrease in salinity as water was elevated by freshwater inputs into the water column at each surface grid. The high-resolution ocean simulation in NEST2 was spun up using initial conditions following the methods of Ishikawa et al. ().
Reanalysis and predicted meteorological GPV-MSM datasets and the National Centers for Environmental Prediction reanalysis datasets were input into the coupled model using the mean hourly air temperature, precipitation, cloud cover, relative humidity, dew-point temperature, and wind speed. The coupled model was validated using observational data and showed quantitative reproducibility for temporal variations in the bay after several calibrations (Nakada et al. [2012a], [2013a]).
We conducted sensitivity experiments by changing the runoffs in the OGCM to evaluate the influence of snowmelt runoff on circulation and water mass formation in the bay. Two numerical runs were conducted: 1) a case using the terrestrial runoff into the bay as the control run (CR) and 2) a case without the total runoff and SGDs (NF) during spring to summer (March to July).
Water sampling and measurements for nutrient concentrations
Dates of water sampling observations from 2011 to 2013
Groundwater data, including nutrient concentrations measured at observational wells, were downloaded from the website of the National Land Information Division, National and Regional Policy Bureau, Ministry of Land, Infrastructure, Transport and Tourism (http://nrb-www.mlit.go.jp/kokjo/inspect/inspect.html), which provided historical observational datasets of groundwater quality. We removed data that lacked information about the elevation at the well or screen depths, because observation depths at the wells could not be calculated without this information. Data without observed water levels were removed as it was not known whether the screens had been submerged in well water. Furthermore, data with observation depths lower than −70 m or higher than 35 m were removed because groundwater located outside these depths is located far from the coast and contributes little to the SGD. A total of 24 (out of 89) wells were chosen for the period of 1980 to 2006.
Simulated terrestrial volume fluxes to the bay
The tank model without the SGD effect (L3; the bottom outlet in the third tank) displayed a positive interannual runoff trend, leading to a decrease in the model's reproducibility (not shown). Meanwhile, the tank model applied for large watersheds (3,000 km2) did not need the bottom outlet in the lowest tank (e.g., Iwanami et al. ): no SGD condition was required. Watersheds in our study area, which are largely characterized by Neogene or Paleogene volcanic rocks (andesite), retained their high permeability. These facts suggested that the SGD effect is essential for realistic simulations of volume and nutrient fluxes in small watersheds (<500 km2) with high permeability.
Time series of all runoffs were estimated using the coefficients determined by Nakada et al. ([2012a]) and summed to derive the total discharge from all watersheds. The robust time series of the total discharge (Figure 6e) clearly showed seasonal maxima around late April, in the snowmelt period. The interannual variations of snowmelt runoffs were also visible and indicated that the largest snowmelt runoffs occurred in 2010. The maximum freshwater loading to Funka Bay occurred when the discharge was 300 to 600 m3 s−1 around late April.
Figure 6e shows the time series of the simulated volume flux of the SGD, which indicates marked increases in the flux after the seasonal maxima of snowmelt runoff from the rivers. The seasonal maxima of SGDs were 28 to 46 m3 s−1, occurring around June. The time lag between the seasonal maxima of SGDs and the snowmelt runoff was approximately 1 month. During the snowmelt period, the water level in the third tank seasonally increased owing to the generation of snowmelt water, which led to a flux maximum that was mainly caused by leakage from the third tank. The flux of SGD after the seasonal maximum gradually decreased from summer to fall.
Simulated terrestrial nutrient fluxes to the bay
Climatologically averaged DIN riverine fluxes from the eight major rivers
DIN (mol s −1 )
A total of 24 wells were located near the ocean to monitor groundwater quality, distributed fairly evenly across the coastal areas around the bay (Figure 7b). DIN concentrations in the groundwater from the wells were 1.3 to 418.3 μmol L−1, with an average of 50 ± 18 μmol L−1 (Figure 7c), approximately twice the climatologically averaged concentration over all riverine waters (28 ± 8 μmol L−1). A total of 15 sites (63%) showed higher concentrations than the lower limit of the averaged riverine concentrations (20 μmol L−1), and these sites were largely located in or around the watersheds of the Osaru, Yurappu, and Nukibetsu rivers and at the base of Mt. Komagatake. This finding indicated that SGDs from these watersheds can increase the DIN concentration in the coastal bottom water. The average groundwater concentration of NH4 (9 ± 2 μmol L−1) was much lower than the concentration of NO2 + NO3 (41 ± 18 μmol L−1). The average concentration of DIN in stream water (NH4: 23 ± 7 plus NO2 + NO3: 113 ± 36 μmol L−1) was 136 ± 43 μmol L−1 (approximately five times the concentration in riverine water). High-concentration DIN water could flow directly into the coastal water around aquafarms through streams and SGDs from narrow watersheds (<10 km2) in the mountains.
Figure 9b shows temporal variations in the percentages of volume and DIN fluxes from SGDs in total fluxes. Variations in the percentages of the volume and DIN were similar, but the percentages of the volume flux were generally higher than those of the DIN flux and often exceeded 50% around February. The percentages of the DIN flux from the SGD in the total terrestrial DIN flux increased from 10% to 15% in the snow-cover season (or winter) and reached its seasonal maximum (30% to 40%) around February for each year except 2007 and 2009. Local maxima also occurred between summer and fall, when both the riverine DIN and volume fluxes were low. Seasonal minima were reached during the snowmelt periods, after seasonal maxima, because of the large DIN flux generated by snowmelt runoff from rivers.
Climatologically averaged DIN fluxes from eight major rivers and their errors (Table 2) were calculated using averaged DIN concentrations (with associated errors) for each river (Figure 7) and averaged volume fluxes and errors (RMSEs), with error calculations based on error propagation. Errors in DIN fluxes from each river ranged from 0.02 to 0.28 mol s−1. The average value was 0.26 ± 0.14 mol s−1 or 54% of the averaged value. The error was affected by errors in the averaged DIN concentration (39%), rather than those in the volume flux (27%). The DIN flux from Osaru River was the highest and corresponded to the highest volume flux. In contrast, the Nigori River hardly affected the total DIN flux as it showed the lowest volume flux (1.6 m3 s−1), although the averaged DIN concentration of the Nigori River was the largest. The Yurappu, Nukibetsu, and Osaru rivers contributed 82% of the total DIN flux because their volume fluxes and DIN concentrations were higher than those for other rivers (Figure 7); however, errors were two to ten times higher than those for other rivers. Therefore, more observations of DIN concentrations in these rivers could decrease errors in future studies.
Results simulated using the coupled land-sea model
The coupled model correctly simulated the typical ocean phenomena that occurred during spring and summer, such as the Oyashio inflow during the spring (Ohtani ), the summertime clockwise circulation induced by terrestrial freshwater (Satoh et al. ), and the local wind in the bay (Takahashi et al. ). These phenomena were often captured by conductivity-temperature-depth (CTD) sensor and acoustic Doppler current profiler (ADCP) observations (e.g., Takahashi et al. ), which were used to validate the simulated results. The simulated results were validated using observational datasets (salinity and temperature) at satisfactory model skill scores (Murphy and Epstein ) ranging from 0.5 to 0.8 (see Nakada et al. [2013a]).
The observed velocity field at the surface (−10 m depth) on July 5 (Figure 11e) shows a clockwise circulation that is similar to the simulation, although current speeds for the circulation were generally five times greater than simulated speeds. The underestimation of the surface velocity field can be attributed to the model's vertical resolution; the layer thickness around the surface layer, from 0 to 20 m, is coarse (4 m), which may insufficiently resolve the surface boundary layer. Local inconsistencies between the simulation and observations were found to correspond to detailed flow patterns around the bay mouth and along the northern coast in the outer bay. The inconsistency around the bay mouth can be attributed to tidal components, for example, the 10 cm s−1 velocity observed at the bay mouth (Ohtani ) can be attributed to the M2 tidal component. The local wind pattern can induce the observed inconsistency in the outer bay, because the horizontal resolution of the wind (GPV/MSM) was originally 5 km, insufficient to resolve the 10 km spatial pattern of the wind. The observed speed at a depth of −50 m (Figure 11f) was comparable to the simulated speeds (not shown), which also suggests that the simulation underestimated the surface velocity field.
Sensitivity experiments for snowmelt runoff
Influence of runoff on circulation
The difference between surface salinity maps for the CR case and the NF case, which is without terrestrial runoffs during March to July (middle panels of Figure 12), indicated that the terrestrial discharge associated with snowmelt decreased the salinity field (by approximately −1.1) along the coast. The salinity difference became small during May to June; the areas of marked salinity difference (less than −0.6) in May to June were confined to the areas around major river estuaries.
Differences in surface velocity fields during the period of snowmelt showed that offshore flows from rivers formed along the coast. These seaward flows deflected to the left, leading to a clockwise velocity field around the center of the bay. Maps of the vertical velocity difference (lower panels of Figure 12) at depths from −10 to −30 m showed upwelling in the near-shore area along the coast of the bay and downwelling in the offshore area. The flow pattern characterized by seaward flows and upwelling can be regarded as the estuarine circulation along the coast of the bay. This arrangement suggests that the terrestrial discharge driven by snowmelt could generate the estuarine circulation along the entire coast of the bay. The clockwise velocity field, enhanced by snowmelt runoff, is significant in March to April; however, the estuarine circulation diminished in May to June, and the clockwise flows weakened and moved closer to the coast.
Influence of runoff on water mass formation
Relationship between terrestrial discharges, nutrients, and larval production
After the winter cooling, nutrients in the bay are vertically homogeneous in February (Lee et al. ). A massive spring phytoplankton bloom, characterized by a high chl a concentration (>20 μg L−1), is then induced by high-nutrient waters, increasing solar radiation, and water column stability during March to April (Kudo and Matsunaga ) and is terminated by nitrate (NO3) depletion (Kudo et al. ). The riverine nutrients can contribute to the summertime primary production in the entire bay (Yoshimura and Kudo ). This study found that terrestrial nutrient loading in the coastal ocean is driven predominantly not only by riverine flux, but also by the nutrient flux through the SGD that accounts for a maximum of 40% of the total flux before the snowmelt period. This DIN flux also contributed to the primary production in near-shore seas.
Our results revealed that DIN concentrations were higher in the near-shore and aquafarm areas than in offshore observation sites during the snowmelt period. The concentration of chl a in near-shore seas (Figure 4) is strongly associated with terrestrial nutrients. The concentrations of riverine nutrients were much higher than concentrations in the mixed layer of the bay (9 to 12 μmol L−1), even before the spring bloom (Kudo et al. ). This suggests that the high DIN flux from rivers can significantly contribute to the increase in DIN concentration, primary production, and chl a in the shallow coastal seas during the winter.
The late inflow of cold Oyashio water, from late April through to May, can decrease the egg and larval production because the low temperature stunts gonad growth (e.g., Sastry and BlakeSource ). Sensitivity experiments showed that the clockwise velocity modified by terrestrial discharge could diminish the anticlockwise inflow of cold Oyashio water (Figure 14), which can suppress the decrease of the water temperature in the bay. The high correlation between the simulated volume flux and the observed spat density (Figure 15) can secondarily explain how the relationship between larval production and Oyashio water volume is affected by the terrestrial discharge.
The observed chl a in February around the aquafarm is considered to be an index of the food availability for scallop spawners and is used for yearly predictions of spat density (Baba et al. ). The 2013 prediction, using the observed chl a, was inconsistent with the observed spat density because the observed chl a can reflect spatial and temporal variations, or inhomogeneity, of chl a concentration affected by intermittent inputs of both terrestrial and oceanic nutrients through the biochemical mechanism of phytoplankton (Yoshimura and Kudo ). The prediction's representativeness may be limited by the location of the observation site. This study emphasized that predictions of spat density based on chl a concentration could be improved by including information about terrestrial runoff or nutrient flux, instead of examining the space-time inhomogeneity of chl a.
DIN nutrient concentrations from the Nigori and Nukibetsu rivers were twice the climatologically averaged concentration of all riverine waters (28 ± 8 μmol L−1), owing to the supply from agricultural lands in the watersheds. The proportion of agricultural areas in these watersheds determines the baseline concentration of terrestrial nutrients (Ileva et al. ). The nutrient concentration during the snowmelt period was diluted by the large volume of snowmelt water. However, the temporal variation in the nutrient flux was controlled by the volume flux rather than the nutrient concentration. Similar results have been suggested by previous studies (Iwanami et al. ). This tendency for dilution during the snowmelt period may be a general feature of snow-covered coastal areas around the subarctic zone. The spawning density was affected by riverine nutrients in near-shore areas, which suggests that nutrient loading from agricultural lands played an essential role in food availability for spawners. However, additional studies are needed to determine whether the current loading of terrestrial nutrients is appropriate for scallop aquaculture. The increase in the nutrient flux can improve food availability, but it can also induce harmful algal blooms of organisms such as Alexandrium tamarense (Anderson et al. ).
Water circulation and larval migration
The large snowmelt runoff induces upwelling and seaward advection as components of estuarine circulation (Figure 12) and forms nutrient-rich low-salinity water in the surface layer of Funka Bay. The rich nutrients that upwell from the deeper layer may contribute to high chl a concentrations near the aquafarms. Terrestrial water accumulates in the bay because the clockwise circulation, driven by the local wind (Takahashi et al. ) and terrestrial runoff (Satoh et al. ), can restrict the exchange of bay water with external water (Isoda et al. ), resulting in a low-salinity surface water (Hasegawa and Isoda ). After the snowmelt period, larvae can be trapped along the coast and transported to the northern coast by clockwise circulation.
The clockwise circulation can be referred to as a type 2 buoyant plume (Garvine ), as the buoyant water at the river mouth turns left along the coast in the northern hemisphere. This water can be trapped on the coast and self-advected along the left-hand coast (Chapman and Lentz ). Satoh et al. () interpreted the clockwise circulation as a geostrophic flow proceeding along the left-handed coast, induced by an internal Kelvin-wave propagation under the condition of a preexisting oceanic mixed layer, based on the theory proposed by McCreary et al. (). Sensitivity experiments on freshwater inputs in the ocean model indicated that the buoyancy flux supplied by the river runoff contributed secondarily to the velocity field or the clockwise circulation in the bay. The contribution of the riverine buoyancy to the circulation was estimated at approximately 20% (= V r (1 [cm s−1])/V s (5 [cm s−1]) × 100) of the total momentum of the clockwise velocity field, where V r is the typical velocity difference that is revealed by sensitivity experiments or induced by riverine runoffs (Figure 12) and V s is the typical simulated velocity of the clockwise circulation (Figure 11); the remaining 80% was presumably supplied by the wind (Takahashi et al. ). Episodically, strong winds can form sub-mesoscale vortices in the bay and distort the clockwise circulation (Nakada et al. [2012b]). Such vortices have often been observed in other coastal seas and can locally modify ocean currents (Nakada et al. [2013b]).
This study comprehensively investigated the roles of snowmelt runoff in water circulation in terms of riverine buoyancy flux, in water mass formation in terms of freshwater volume flux, and in the production of scallop larvae in terms of riverine nutrient flux, for Funka Bay, based on hydrological and nutrient data. Sensitivity experiments using a high-resolution coupled land-sea simulation were also conducted. The terrestrial freshwater and nutrient (DIN) fluxes from rivers and SGDs into the bay were estimated using a hydrological model and climatological datasets compiled from nutrient concentration data for the rivers, streams, and groundwater. The comparison between observed and simulated runoffs indicated that the model accurately reproduced the terrestrial runoff, allowing for estimates of the terrestrial nutrient flux because variations in terrestrial volume fluxes controlled variations in the terrestrial DIN fluxes. The average volume flux from SGDs accounted for 26% of the flux for riverine runoff. The nutrient flux from SGDs accounted for a maximum of 40% of the total loading to the coastal ocean of the bay in February before the snowmelt period.
The impact of snowmelt runoff on water circulation and water mass formation in the bay was first evaluated by sensitivity experiments using the OGCM. This study revealed, as part of the land-sea linkage, that snowmelt runoff has three important roles in the ocean environment and scallop production: 1) Buoyancy fluxes supplied from the runoff can enhance the clockwise circulation in the bay, but rarely dominate the velocity field. 2) The terrestrial freshwater input can diminish the springtime inflow of the cold OW, leading to an increase in larval production and a decrease in the volume of wintertime TW in the bay, and can modify distributions of the wintertime water masses. 3) Fluxes of high-concentration nutrients from the runoff can form a favorable environment for spawners by improving food availability. Therefore, the snowmelt runoff hardly contributes to larvae transport but can increase egg production through the spring nutrient supply. In addition, upwelling may advect rich nutrients from deeper regions to the surface in the near-shore ocean, and the clockwise circulation may transport larvae northward along the coast.
Agricultural fields were important sources of nutrients for feeding scallop spawners. This study, therefore, demonstrated that the integrated management of agricultural farms and aquafarms (Iwasaki ) in the large bay is essential for controlling the nutrient flux from the land to the ocean. The volume, buoyancy, and nutrient fluxes supplied from terrestrial snowmelt runoff can be key factors in fishery or aquaculture production in subarctic coastal regions, alongside watersheds in the snow zone. Additional studies are needed to elucidate land-ocean linkages and to design sustainable and synergetic systems of aquaculture and agriculture.
SN received a bachelor's degree in Fisheries Science and a master's degree in Fisheries Science (studying in the Faculty of Fisheries Sciences) from Hokkaido University, Hakodate, Japan, in 1999 and 2001, respectively. He received a PhD from Kyushu University (the Department of Earth System Science and Technology, Interdisciplinary Graduate School of Engineering Sciences), Ohnojo, Japan, in 2008. He was a researcher for the multidisciplinary consulting company Knowledge Consulting for Solution, Co., Ltd., Tokyo, Japan, during 2001 to 2003 and JAPAN NUS Co., Ltd., Tokyo, Japan, during 2003 to 2005, and a fellow of Japan Society for the Promotion of Science during 2006 to 2008, at Yonsei University, Seoul, Korea, in 2008, and at the Research Institute of Humanity and Nature (RIHN) during 2009 to 2010. He was mainly engaged in the development of numerical models, analyses of their outputs, and design for visualization systems. He is currently a researcher at the Institute for Liberal Arts and Sciences, Kyoto University, Kyoto, Japan. His current research is predominantly aimed at developing the operational ocean forecast system based on a land-sea-coupled model using data assimilation methods and high-end computers, investigating land-ocean interactions based on the large amount of output datasets produced by high-resolution ocean simulations and high-density observations.
Dr. KB is a researcher in the Hokkaido Research Organization, Fisheries Research Department, Hakodate Fisheries Research Institute, Hakodate, Japan. His research interest is to study the biological mechanisms of scallops, such as spawning, and ecosystems favorable to fishery productions with the aim of predicting the annual spat density.
MS is a researcher in the Hokkaido Research Organization, Fisheries Research Department, Hakodate Fisheries Research Institute, Hakodate, Japan. His research interest is to investigate relationships among fishery productions such as scallops, ocean circulation, water masses, and oxygen-deficient water.
MN received BE and MS degrees (studying in Faculty of Agriculture) from Kyoto University, Kyoto, Japan. He is currently a student in the Graduate School of Fishery, Faculty of Fisheries Sciences, Hokkaido University, Japan. His research interest is to examine the mechanisms of harmful blooming algae, such as Alexandrium tamarense, in terms of nutrient fluxes, pertaining to marine biological systems in the coastal ocean.
YI received a BE degree and a PhD in Science from Kyoto University, Kyoto, Japan, in 1994 and 1999, respectively. He was an Assistant Professor in the Graduate School of Science in Kyoto University in 2001. Currently, he is a Principal Research Scientist in Data Research Center of Marine-Earth Science (DrC), Japan Agency for Marine-Earth Science and Technology (JAMSTEC), Japan, Yokohama (2012). His research interests are in physical oceanography and climate physics, especially numerical modeling and data assimilation systems.
TA received a BS degree (studying in the Faculty of Engineering) from Kyoto University in 1972 and received MS and Doctor of Science degrees at the Graduate School of Science, Kyoto University, Kyoto, Japan. He is currently an executive vice president for education at Kyoto University (from 2010).
KK received BE and ME degrees and a PhD in Electrical and Electronic Engineering (studying at the Faculty of Engineering) from Kyoto University, Kyoto, Japan, in 1983, 1985, and 1994, respectively. He is currently a professor at the Institute for Liberal Arts and Sciences, Kyoto University, Kyoto, Japan. From 1985 to 1998, he has a 14-year career in IBM Japan Ltd., Tokyo, Japan. From 1998 to 2001, he was an associate professor at Iwate Prefectual University. From 2001 to 2003, he was an associate professor at Kyoto University. His research interests include modeling, simulation, and visualization. He is a member of IEEE Computer Society, directors of Visualization Society Japan, and the Institute of Systems, Control and Information Engineers, and a president of Japan Society of Simulation Technology. He received the IEMT/IMC outstanding paper award in 1998, the VSJ contribution award in 2009, and the VSJ outstanding paper award in 2010.
SS is a professor in the Laboratory of Satellite Oceanography, Division of Marine Bio-resource and Environmental Science, Hokkaido University, Hakodate, Japan. He was a research engineer for the Japan Weather Association, Tokyo, Japan during 1984 to 1993. From 1993 to 2000, he was an associate professor at Hokkaido University. His research interests cover many fields, including fishery oceanography, satellite oceanography, marine ecosystems, and marine GIS.
acoustic Doppler current profilers
- chl a :
the control run case that includes terrestrial runoffs into the bay
conductivity, temperature, depth (CTD) sensors
dissolved inorganic nitrogen
the low-salinity Funka Bay surface water
highly saline Funka Bay water formed in winter
grid point value datasets of the meso-scale model
hydrometeorological and multi-runoff utility model
intertropical convergence zones
the mixed water (MW) that blends FS, the summertime OW (S-OW), and Tsugaru water (TW)
the case without snowmelt runoffs during spring to summer (March to July)
ocean general circulation model
subarctic Oyashio water
submarine groundwater discharge
the summertime OW
We thank Mr. M. Watanobe, Mr. M. Kanamori, and the crew and officers of the R/V Kinsei-Maru of Hakodate Fisheries Research Institute for providing in situ data. We deeply thank Professor I. Kudo (Faculty of Fisheries Sciences, Hokkaido University) and Dr. T. Yoshimura (Central Research Institute of Electric Power Industry) for suggesting improvements to this paper and providing the nutrient datasets (Yoshimura and Kudo ), which greatly helped in the calculation of the terrestrial DIN flux. This work was partially supported by the ‘Hakodate Marine Bio Cluster Project’ in the knowledge Cluster Program from 2009, the Grant-in-Aid for University and Society Collaboration from the Ministry of Education, Culture, Sports, Science and Technology (MEXT), Japan, and the Grant-in-Aid for Research Programs on Climate Change Adaptation (RECCA). We acknowledge the National Aeronautics and Space Administration (NASA) for the MODIS/Aqua data and the Research Institute for Sustainable Humanosphere, Kyoto University, for the archived data of GPV-MSM. We deeply appreciate the constructive comments from two anonymous reviewers and editors.
- Anderson DM, Alpermann TJ, Cembella AD, Collos Y, Masseret E, Montresor M: The globally distributed genus Alexandrium : multifaceted roles in marine ecosystems and impacts on human health. Harmful Algae 2012, 14: 10–35. 10.1016/j.hal.2011.10.012View ArticleGoogle Scholar
- Baba K, Sugawara R, Nitta H, Endou K, Miyazono A: Relationship between spat density, food availability, and growth of spawners in cultured Mizuhopecten yessoensis in Funka Bay: concurrence with El Nino southern oscillation. Can J Fish Aquat Sci 2009, 66: 6–17. doi:0.1139/F08-183View ArticleGoogle Scholar
- Bondad-Reantaso MG, Subasinghe RP, Josupeit H, Cai J, Zhou X: The role of crustacean fisheries and aquaculture in global food security: past, present and future. J Invert Path 2012,110(2):158–165. doi:org/10.1016/j.jip.2012.03.010View ArticleGoogle Scholar
- Chapman DC, Lentz SJ: Trapping of a coastal density front by the bottom boundary layer. J Phys Oceanogr 1994, 24: 1464–1479. doi: http://dx.doi.org/10.1175/1520-0485(1994)024<1464:TOACDF>2.0.CO;2View ArticleGoogle Scholar
- Dai A, Trenberth KE: Estimates of freshwater discharge from continents: latitudinal and seasonal variations. J Hydro 2002, 3: 660–87. doi:10.1175/1525-7541(2002)003<0660:EOFDFC>2.0.CO;2View ArticleGoogle Scholar
- Darecki M, Stramski D: An evaluation of MODIS and SeaWiFS bio-optical algorithms in the Baltic Sea. Remote Sens Environ 2004,89(3):326–350. doi:10.1016/j.rse.2003.10.012View ArticleGoogle Scholar
- FAO yearbook: fishery and aquaculture statistics 2009. Food and Agriculture Organization of the United Nations, Rome, Italy; 2011.
- Garvine RW: The impact of model configuration in studies of buoyant coastal discharge. J Mar Res 2001, 59: 193–225. doi:10.1357/002224001762882637View ArticleGoogle Scholar
- Gent PR, McWilliams JC: Isopycnal mixing in the ocean circulation model. J Phys Oceanogr 1990, 20: 150–155. doi:10.1175/1520-0485(1990)020<0150:IMIOCM>2.0.CO;2View ArticleGoogle Scholar
- Griffies SM: The Gent-McWilliams skew flux. J Phys Oceanogr 1998, 28: 831–841. doi:10.1175/1520-0485(1998)028<0831:TGMSF>2.0.CO;2View ArticleGoogle Scholar
- Hanawa K, Mitsudera H: Variations of water system distribution in the Sanriku coastal area. J Oceanogr Soc Japan 1986, 42: 435–446. 10.1007/BF02110194View ArticleGoogle Scholar
- Hasegawa N, Isoda Y: Fresh water budget of Funka Bay. Umi to Sora 1997, 73: 113–121. (in Japanese with English abstract)Google Scholar
- Ileva NY, Shibata H, Satoh F, Sasa K, Ueda H: Relationship between the riverine nitrate-nitrogen concentration and the land use in the Teshio River watershed, North Japan. Sustain Sci 2009, 4: 189–198. doi:10.1007/s11625-009-0081-5View ArticleGoogle Scholar
- Ishikawa Y, Awaji T, Toyoda T, In T, Nishina K, Nakayama T, Shima S, Masuda S: High-resolution synthetic monitoring by a 4-dimensional variational data assimilation system in the northwestern North Pacific. J Mar Syst 2009, 78: 237–248. doi:10.1016/j.jmarsys.2009.02.016View ArticleGoogle Scholar
- Isoda Y, Takeyama Y, Azumaya T: Ocean structural change around the diatom bloom in Funka Bay. Bull Coast Oceanogr 2000, 38: 3–14. (in Japanese with English abstract) (in Japanese with English abstract)Google Scholar
- Iwanami H, Wada T, Sakamoto K, Kudo I, Chikita KA: Mechanism of nutrients supply from the Tokachi River catchment considering characteristics of discharge and land use. J Jpn Assoc Hydro Sci 2013,43(1):3–24.Google Scholar
- Iwasaki S: Fishers-based watershed management in Lake Saroma, Japan. Ocean & Coastal Manag 2013, 81: 58–65. 10.1016/j.ocecoaman.2012.07.019View ArticleGoogle Scholar
- Kimmerer WJ: Effects of freshwater flow on abundance of estuarine organisms: physical effects or trophic linkages? Mar Eco Prog Series 2002, 243: 39–55. 10.3354/meps243039View ArticleGoogle Scholar
- Kondo J, Watanabe T: Studies on the bulk transfer coefficients over a vegetated surface with a multilayer energy budget model. J Atmo Sci 1992,49(23):2183–2199. doi:10.1175/1520-0469(1992)049<2183:SOTBTC>2.0.CO;2View ArticleGoogle Scholar
- Kondo J, Yamazaki T: A prediction model for snowmelt, snow surface temperature and freezing depth using a heat balance method. J Appl Meteorol 1990, 29: 375–384. doi:10.1175/1520-0450(1990)029<0375:APMFSS>2.0.CO;2View ArticleGoogle Scholar
- Kosaka Y, Ito H: Japan. In Scallops: biology, ecology and aquaculture. Edited by: Shumway SE, Parsons GJ. Elsevier, Amsterdam, the Netherlands; 2006:1093–1141. 10.1016/S0167-9309(06)80049-9View ArticleGoogle Scholar
- Kudo I, Matsunaga K: Environmental factors affecting the occurrence and production of the spring phytoplankton bloom in Funka Bay, Japan. J Oceanogr 1999, 55: 505–513. 10.1023/A:1007887307322View ArticleGoogle Scholar
- Kudo I, Yoshimura T, Yanada M, Matsunaga K: Exhaustion of nitrate terminates a phytoplankton bloom in Funka Bay, Japan: change in SiO 4 :NO 3 consumption rate during the bloom. Mar Ecol Prog Ser 2000, 193: 45–51. 10.3354/meps193045View ArticleGoogle Scholar
- Lee CW, Kudo I, Yokokawa T, Yanada M, Maita Y: Dynamics of bacterial respiration and related growth efficiency, dissolved nutrients and dissolved oxygen concentration in a subarctic coastal embayment. Mar Freshw Res 2002, 53: 1–7. 10.1071/MF01003View ArticleGoogle Scholar
- Leonard BP, MacVean MK, Lock AP: Positivity-preserving numerical schemes for multidimensional advection. NASA Tech Memo 106055: 62 ICOMP-93–05, Washington, DC; 1993.Google Scholar
- Liu SM, Li RH, Zhang GL, Wang DR, Du JZ, Herbeck LS: The impact of anthropogenic activities on nutrient dynamics in the tropical Wenchanghe and Wenjiaohe estuary and lagoon system in East Hainan, China. Mar Chemist 2011, 125: 49–68. 10.1016/j.marchem.2011.02.003View ArticleGoogle Scholar
- McCreary JP, Zhang S, Shetye SR: Coastal circulation driven by river outflow in a variable-density 1 1/2-layer model. J Geophys Res 1997, 102: 15535–15554. 10.1029/97JC00985View ArticleGoogle Scholar
- Minoda T, Fuji A: Biological environments in Funka Bay. Tokai University Press p 1106, (In Japanese), Coastal Oceanography of Japanese Islands; 1985.Google Scholar
- Miyake H, Tanka I, Murakami T: Outflow of water from Funka Bay, Hokkaido, during early spring. J Oceanogr Soc Japan 1988, 44: 163–170. doi:10.1007/BF02302640View ArticleGoogle Scholar
- Motovilov Y, Gottschalk GL, England K, Rodhe A: Validation of distributed hydrological model against spatial observations. Agric Forest Meteo 1999, 98–99: 257–277. doi:10.1016/S0168- 1923(99)00102-1View ArticleGoogle Scholar
- Murphy AH, Epstein ES: Skill scores and correlation coefficients in model verification. Mon Weather Rev 1989, 117: 572–581. doi:10.1175/1520-0493(1989)117<0572:SSACCI>2.0.CO;2View ArticleGoogle Scholar
- Nakada S, Yasumoto J, Taniguchi M, Ishitobi T: Submarine groundwater discharge and seawater circulation in a subterranean estuarine beneath a tidal flat. Hydrol Process 2011, 25: 2755–2763. doi:10.1002/hyp.8016View ArticleGoogle Scholar
- Nakada S, Ishikawa Y, Awaji T, In T, Shima S, Nakayama T, Isada T, Saitoh S: Modeling runoff into a region of freshwater influence for improved ocean prediction: an application in Funka Bay. Hydro Res Lett 2012, 6: 47–52. 10.3178/hrl.6.47View ArticleGoogle Scholar
- Nakada S, Uenaka T, Ishikawa Y, Matsui K, Sakamoto N, Koyamada K, Awaji T, Saitoh S: A visualization study of vortexes extracted from the tremendous ocean simulation data: an application in Funka Bay. Jpn Soc Sim Tech 2012,4(4):145–152. doi:org/10.1016/j.pocean.2013.10.008Google Scholar
- Nakada S, Ishikawa Y, Awaji T, In T, Koyamada K, Watanobe M, Okumura H, Nishida Y, Saitoh S: An integrated approach to the heat and water mass dynamics of a large bay: high-resolution simulations of Funka Bay, Japan. J Geophys Res Ocean 2013, 118: 1–18. doi:10.1002/jgrc.20262View ArticleGoogle Scholar
- Nakada S, Hirose N, Senjyu T, Fukudome K, Tsuji T, Okei N: Operational ocean prediction experiments for smart coastal fishing. Prog Oceanogr 2013, 121: 125–140. doi.org/10.1016/j.pocean.2013.10.008 doi.org/10.1016/j.pocean.2013.10.008 10.1016/j.pocean.2013.10.008View ArticleGoogle Scholar
- Nash JE, Sutcliffe JV: River flow forecasting through conceptual models part I - a discussion of principles. J Hydro 1970,10(3):282–290. doi:10.1016/0022-1694(70)90255-6View ArticleGoogle Scholar
- Noh Y: Sensitivity to wave breaking and the Prandtl number in the ocean mixed layer model and its dependence on latitude. Geophys Res Lett 2004, 31: L23305. doi:10.1029/2004GL021289Google Scholar
- Oey LY, Chen P: A nested-grid ocean model - with application to the simulation of meanders and eddies in the Norwegian coastal current. J Geophys Res 1992, 97: 20063–20086. 10.1029/92JC01991View ArticleGoogle Scholar
- Ohtani K: Funka Bay (I) physical environments. Bull Coastal Oceanogr 1981, 19: 68–80. (in Japanese) (in Japanese)Google Scholar
- Ohtani K, Kido K: Oceanographic structure in Funka Bay. Bull Fac Fish Hokkaido University, Hakodate, Japan; 1980.Google Scholar
- Rosa AL, Isoda Y, Uehara K, Aiki T: Seasonal variations of water system distribution and flow patterns in the southern sea area of Hokkaido, Japan. J Oceanogr 2007, 63: 573–588. doi:10.1007/s10872-007-0051-4View ArticleGoogle Scholar
- Sastry AM, BlakeSource NJ: Regulation of gonad development in the bay scallop, Aequipecten irradians . Lamarck Biol Bull 1971,140(2):274–283. 10.2307/1540074View ArticleGoogle Scholar
- Satoh C, Isoda Y, Shimizu M: Clockwise circulation in the upper layer of Funka Bay in summer. Bull Coast Oceanogr 2003, 40: 181–188. (In Japanese with English Abstract)Google Scholar
- Shimada H, Nishida Y, Ito Y, Mizushima T: Relationship among growth and survival of cultured scallops ( Patinopecten yessoensis JAY), and environmental condition in the coastal area off Yakumo, Funka Bay, Hokkaido, Japan. Sci Rep Hokkaido Fish Exp Stn 2000, 27: 49–62. (In Japanese with English Abstract)Google Scholar
- Simpson JH: Physical processes in the ROFI regime. J Mar Syst 1997, 12: 3–15. doi:10.1016/S0924-7963(96)00085-1View ArticleGoogle Scholar
- Takahashi D, Kido K, Nishida Y, Kobayashi N, Higaki N, Miyake H: Dynamical structure and wind-driven upwelling in a summertime anticyclonic eddy within Funka Bay, Hokkaido, Japan. Conti Shelf Res 2007, 27: 1928–1946. 10.1016/j.csr.2007.04.001View ArticleGoogle Scholar
- Takahashi D, Miyake H, Nakayama T, Kobayashi N, Kido K, Nishida Y: Response of a summertime anticyclonic eddy to wind forcing in Funka Bay, Hokkaido, Japan. Conti Shelf Res 2010, 30: 1435–1449. doi:10.1016/j.csr.2010.05.003View ArticleGoogle Scholar
- Taniguchi M, Burnett WC, Cable JE, Turner JV: Investigations of submarine groundwater discharge. Hydrol Process 2002, 16: 2115–2129. doi:10.1002/hyp.1145View ArticleGoogle Scholar
- Uiboupin R, Arino O (2010) Study of snowmelt impact on SST and TSM fields in the coastal zone of Barents Sea. IEEE International Geoscience and Remote Sensing Symposium, pp 4212–4215, doi:10.1109/IGARSS.2010.5654375Google Scholar
- Willmott CJ, Matsuura K (2001) Terrestrial water budget data archive: monthly time series (1950-1999). , [http://climate.geog.udel.edu/~climate/html_pages/archive.html] Willmott CJ, Matsuura K (2001) Terrestrial water budget data archive: monthly time series (1950-1999).
- Yoshimura T, Kudo I: Riverine nutrient loadings and their impact on primary production in Funka Bay, Japan. Oceanogr Jpn 2003, 12: 185–193. (in Japanese with English abstract) 10.5928/kaiyou.12.185View ArticleGoogle Scholar
- Yoshimura T, Kudo I: Seasonal phosphorus depletion and microbial responses to the change in phosphorus, availability in a subarctic coastal environment. Mar Chemi 2011, 126: 182–192. 10.1016/j.marchem.2011.06.003View ArticleGoogle Scholar
- Yoshimura K, Sakimura T, Oki T, Kanae S, Seto S: Toward flood risk prediction: a statistical approach using a 29-year river discharge simulation over Japan. Hydro Res Lett 2008, 2: 22–26. doi:10.3178/hrl.2.22View ArticleGoogle Scholar
- Zhao K, Nakada S, Sakamoto N, Koyamada K, Bajaj C, Ishikawa Y, Awaji T, In T, Saitoh S: A visualization for the dynamic behaviors of the mixture of water mass for Northwestern Pacific near Japan. Int J Model Simul Sci Comput 2013,4(1341002):18. doi:10.1142/S179396231341002XGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited.