Tree crown dimension and profile (shape) data are often required as inputs to
competition indices within single-tree growth and yield models. Crown radius and
length were measured for 75 birch (Betula spp.) and 154 Sitka spruce (Picea sitchensis
(Bong.) Carriere) sample trees in permanent sample plots in Wales, and crown shapes of
40 trees were assessed using a "crown window". Crown radius models based on stem
diameter at breast height (dbh) and parameterised for individual stands perform best.
Where a single parameterisation must provide data for all stands, a model incorporating
stem dimensions is best for birch, and a model incorporating stem dimensions, local
stocking and dbh dominance is best for spruce. Crown radius modelling results are poor
for heavily suppressed birch. To ensure realistic model behaviour across multiple time
steps, a crown length model based on tree height alone is best for both species. The
length of the light crown, above the point of maximum crown radius, is roughly two
thirds of birch total crown length and three quarters of spruce crown length. Light
crown profile models give acceptable results for birch and spruce, but shade crown
profile models fail to account for variation in crown shapes, even with spatial variables
as inputs, and an alternative model is suggested. Crown window and terrestrial
photogrammetry crown profile data for three spruce were compared, using the software
PhotoModeler Pro 5 to produce three-dimensional maps of branch tips from
photographs. These methods produce comparable profile data for Sitka spruce up to
four metres in height. Photogrammetric analyses of birch and larger spruce present
considerable difficulties. The crown window can be used more widely, but
photogrammetry can potentially yield more data; this is demonstrated by reconstructing
spruce branch tips for a previous growing season and examining changes in branch
length and crown shape.