3.3.	Computer simulations in pest management
		3.3.1.	Assumptions and oversimplifications
			3.3.1.1.	General goal is to construct as simple a model as possible to meet the objectives established at the outset
			3.3.1.2.	Structure (coarse adjustment)
				Aggregation of like elements that behave similarly (generalization) -- e.g., a highly aggregated pest management model might have a single component to represent "insect" rather than a separate component for each species or each population The level of resolution is selected to include only as much detail as is necessary Simple curve fitting is often adequate for prediction (e.g. prediction of yields), but rational modeling using logical deductions from a set of premises is necessary for mechanistic models
			3.3.1.3.	Parameterization (fine tuning)
				Simply guess at values (so state in model documentation) Obtain values from literature data Conduct carefully controlled experiments to estimate parameters precisely
		3.3.2.	How computer simulations are used in pest management
			3.3.2.1.	Training
				Gives novices experience in management decision-making without risk of disastrous consequences Compresses several seasons into a few minutes or hours -- student can experience a wide range of conditions
			3.3.2.2.	Research
				Already mentioned -- organizing a research program and establishing priorities Conduct experiments that are not possible in the real system (e.g., large factorial experiments can be simulated in order to later zero in on treatments to test in the field) Use simulations to develop pest management strategies or to create decision-making models for growers, extension agents, pest management consultants
			3.3.2.3.	Extension
				Modeling can complement monitoring

				Modeling can help determine when to monitor (e.g. alfalfa pest management program)
				Monitoring can be used to "calibrate" the model

Return to IPM Home Page