phytoplankton and zooplankton carbon concentrations); however, these comparisons 

 result in general agreement between measured and simulated properties (figs. 3- 

 5). This demonstrates the model's ability to simulate the seasonal dynamics of 

 phytoplankton, zooplankton, and major nutrients; however, model output should 

 also be compared to measurements of process rates to see how well the model 

 simulates the internal dynamics of the system. In the discussions that follow, 

 both simulated and measured process rates are used to ensure that internal 

 model dynamics are consistent with observations. I use process rates that were 

 measured during IFYGL or measured in Lake Ontario during other years. Also, 

 where certain processes were not observed in Lake Ontario, information from 

 other field and laboratory studies are used for general comparisons . Simulated 

 and measured rates were also in general agreement, which allows one to use the 

 model to speculate about the relative importance of these processes. This 

 emphasis on process rate comparisons is discussed further below in the context 

 of nonunique parameter estimates. 



ANALYSIS OF LAKE ONTARIO 



This model was exercised within two physical frameworks. First, to explore 

 the controls of phytoplankton production on a lake-scale, seasonal basis, output 

 from the horizontally averaged, two-layer model described above was analyzed 

 with particular emphasis on cycling of phosphorus, the primary limiting nutrient 

 (Scavia 1979). Second, to explore development and maintenance of strong offshore 

 nutrient and plankton gradients, output from a two-dimensional, coupled 

 ecological-physical transport model was analyzed for the spring-summer transition 

 period (Scavia and Bennett 1980). 



Lake-scale, seasonal dynamics - Simulated processes controlling seasonal 

 dynamics of one of the modeled phytoplankton groups are shown in figure 6. 

 Stippled areas on each graph indicate net rate of change of population biomass . 

 In winter and early spring, this algal group, as well as the others, is controllec 

 primarily by the balance between gross primary production and two physical 

 processes, sinking and vertical mixing. During this time, phytoplankton gross 

 production is limited mainly by the availability of light which is controlled 

 both by the amount of incoming solar radiation and the depth to which the 

 phytoplankton are mixed. Mixing between the two model layers becomes quite 

 intense in early spring (fig. 2), indicating that the mixing depth is the depth 

 of the entire water column. Thus, loss to dark, deeper waters prevents sub- 

 stantial increases in algal biomass. 



Riley (1942, 1946, 1963) and Riley et al . (1949) discussed the importance 

 of the relationship between depth of the sunlit zone and mixing depth. They 

 suggested that phytoplankton realize their productivity only after the thermocline 

 begins to develop because it is only after this time that losses to the dark, 

 nonproductive strata are reduced. Significant increases in algal biomass occur 

 only after midspring, when surface waters in Lake Ontario begin to warm and the 

 lake begins to stratify vertically (fig. 6). At this time (early June), 

 phytoplankton populations increase rapidly and concentrations of nutrients they 

 assimilate begin to decrease. The concentrations of nutrients decrease because 

 they also become relatively isolated from the nutrient-rich lower strata. 



62 



