Introduction

A synthesis of canopy processes can be accomplished at various levels of detail. If historical data are available, then a statistical analysis of that data may provide a kind of synthesis; however, in this case the synthesis is implicit in the statistical tool used, yielding limited insight to us. Alternatively, a mechanistic approach can be used and each relevant process described by appropriate, state-of-the-art, quantitative relations with explicit integration (or synthesis) to achieve an ‘integrated whole’. Clearly, statistical and mechanistic approaches represent extremes of a continuum where all intermediate states are possible. Thus a clear statement of objectives, guiding rules for pursuing these objectives, definition of the system, and evaluation criteria are prerequisites for beginning an orderly synthesis of canopy processes.

This chapter represents an attempt at an orderly synthesis of canopy processes with a reasonably mechanistic approach. The plant-environment model entitled Cupid (Norman & Campbell, 1983; Norman, 1979; Norman, 1982) is used as an example.

Rules for constructing a model

A system of rules for pursuing a synthesis of processes can aid one in resisting the temptation to ‘over sell’ and thus avoid having either to resort to short-term expediency when failure is in sight, or to justify the means deceptively with an end result that was essentially known before the modelling was begun.