Stage 1:  Review the requirement for predictions

Whenever environmental noise predictions are proposed, it is necessary to have a clear understanding of the reasons for the predictions, the significance of any decisions to be made on the basis of them, any variable characteristics of the sound field in question that may represent sources of uncertainty (including longer term sources of variability or change, as may be related to seasonal effects or future imminent development that may passively or directly change the noise environment), and the likely feasibility of conducting predictions.

It is at this point that consideration must be given to the type of noise data that the assessment calls for, under the following categories:

• Absolute values, where the specific numeric value of the calculation is important (e.g. comparison against known benchmark).
• Relative difference values (which may between sources or locations), where developing attenuation strategies or complementing the design or findings of a measurement study.

The relevance and reliability of any existing assessment data that is available should be evaluated to determine whether this may negate the need for further assessments. The limitations of any available existing data will need to be considered against the difficulties that may face any attempt to develop new prediction data.

Finally, it is important at this point to identify any well-defined threshold values that trigger significantly different assessment outcomes. The knowledge of such thresholds, in conjunction with estimates of what predicted values that might be expected from the study (see following screening exercise) is critical to determining the requirement for, and scope of, subsequent detailed modelling exercises.