Sensitivity testing. This refers to the process of examining changes in the 

 output of a model resulting from intentional changes in the input. If large 

 changes are caused by minor changes in the input, the overall results will 

 depend greatly upon the quality of the verification. Unfortunately, for many 

 practical applications, there is some degree of doubt in the verification 

 (Hansen and Kraus 1989). If a model is oversensitive to small changes in 

 input values, the range of predictions will be too broad and will in essence 

 provide no information. 



In summary, numerical models are a valuable complement to prototype 

 data collection and physical (scale) models of coastal processes. However, 

 useful numerical models require empirical input during the calibration and 

 may be based on incomplete data sets. Therefore, the reader is urged to be 

 cautious of the output of any model and to be aware of the results of the 

 verification and sensitivity tests. 



1 23 



Chapter 5 Analysis and Interpretation of Coastal Data 



