Prediction of conifer natural regeneration in a 'data-poor' environment

greater interest in natural regeneration. This paper describes a project that designed and tested a model to predict the likelihood of natural regeneration in an environment where long-term datasets were not available. A spreadsheet based model known as REGGIE (REGeneration GuIdancE) was designed based on first principles and silvicultural experience. It was tested on 129 sites of four conifer species on a wide range of sites throughout Britain; at each site an expert judged the likelihood of regeneration in the next 5 years in one of five classes: 0-20%, 21-40%, 41-60%, 61-80% and 81-100%. The REGGIE model agreed with the expert prediction on 63 of the 129 sites (48.8%). The validation data were then analyzed using an ordinal logistic regression. The minimal adequate model included fewer terms compared with REGGIE and, not surprisingly, was more accurate with respect to the expert prediction on 113 of the 129 sites (87.6%). An advantage of the ordinal logistic model is that we have devised a simple score based method of application which is easy to apply in the field. Informal validation of this model has suggested that it has potential to be used by forest managers as part of a strategy to raise understanding of how to use natural regeneration when transforming conifer stands to continuous cover in Britain.