To produce the sound samples, the marginal statistics, statistical modelling, concatenative synthesis and sinusoidal modelling techniques are all analysis and resynthesis techniques, where a source sample can be considered as a source and the samples are reproduced. There are some parameters, that the authors modified only to limit the number of sonic artifacts present in the audio samples rendered. Only the physically inspired  and additive techniques requires control in an attempt to reproduce the same audio context as the given source sample. For the physically inspired samples, all synthesis models have defaults parameter settings that were used. Only a density or intensity parameter was modified to produced a sound similar to the reference sample. To synthesize using the additive synthesis method, as many default parameters were maintained as possible, and the parameters were tuned by the author to produce a sound as similar to the references sound as possible. The parameters that were tuned manually are: rain intensity; fire intensity; waves size; wind intensity, coldness and variations.