Recent research has resulted in the development of models that predict potential yield of 9 common woodland tree species in lowland England and Wales from easily measured site factors, prior to planting. Although such models give good predictions of yield at regional and local levels, errors in predicted yield can be large (e.g. +/- 4 m3 ha-1 yr-1) for individual farm sites. This study will refine the site yield model for ash in order to enable more accurate predictions of yield to be made for individual sites. A consistent method of site classification for yield in Britain will be developed, based on primary and secondary site factors affecting tree growth. This will involve implementation of a comprehensive review of agroclimatic factors affecting growth and yield of tree stands, correlation of site factors in Britain using GIS-based graphical, statistical and multivariate analyses, and identification of key site factors responsible for determining growth. A methodology for construction of site yield models that are applicable to any tree and site in Britain will also be developed. A sampling scheme and a mathematical model for evaluating site and yield variables will be used, while computer methods will be employed for manipulation of site and yield data, as well as for calibrating site yield models based on these data for particular tree species and site types. The site-specific yield model will then be developed by identifying the geographical spread and location of stands of ash trees in England and Wales, highlighting a sample of ash stands in England and Wales for field assessment of site and yield data, collecting and processing of site and yield data from a sample of ash stands, and calibrating the site yield model thus produced. Mathematical methods will be used to combine the refined site yield model for ash with a newly developed silviculture yield model for ash and a computer program will then be developed to predict potential yield and produce out-turn, given information regarding site type and silvicultural treatment. Finally, practical field guides, in the form of look-up tables, will be produced for assessing potential yield and produce out-turn of ash, based on evaluation of site factors prior to planting. The study will lead to the development of a unified methodology for construction of site yield models that are not restricted to either a particular tree species or region of Britain.