Abstract
Methods are given to construct a unique non-oscillatory sequence through time varying data. This unique sequence gives a basis to test the significance of variations in data. It is also shown to give a basis to predict final steady states from time varying data. Steady state values predicted in this way are found to agree with measured steady-state values. Better prediction is shown than by common models. Methods to test the adequacy of growth models are given and are demonstrated with data. The methods are illustrated with data from the continuous growth of Dunaliella peircci. Time varying cell concentrations are measured after changes of inputs to the growing system. Specific production rates of cells and oxygen as well as the endocellular specific production rates of organic nitrogen, protein, carbohydrate, and chlorophyll a are measured. Specific utilization rates of carbon dioxide and nitrate ion are also measured. The data above are shown to be predicted by calculations from light intensity, residence time, and dimensions of the growth vessel for given initial conditions. Growth rates, washout loci, and photosynthetic quotients are also calculated from the same quantities. The methods to test predictions of mathematical models are demonstrated with the data and for several models, including a Monod-type model.
Moss, Edward Robert (1973). Predictive methods in steady and transient growth of microflora. Doctoral dissertation, Texas A&M University. Texas A&M University. Libraries. Available electronically from
https : / /hdl .handle .net /1969 .1 /DISSERTATIONS -157525.