|
The composite sensitivity of any parameter describes the effect that an alteration to that parameter has on all model-generated observations (with observation weights applied) employed in the optimisation process.
This is a very useful variable for assessing the degree to which a parameter is capable of estimation on the basis of the observation dataset available for model calibration. It can also be very useful in deciding
which parameters to temporarily hold fixed in order to improve PEST performance through the user intervention.
|
