|Authors||Brewer, M.J., Sulkava, M., Mäkinen, H., Korpela, M., Nöjd, P. and Hollmén, J.|
|Publication details||In "Intelligent Data Engineering and Automated Learning - IDEAL 2011", 204-211. Eds. H. Yin, W. Wang, V. Rayward-Smith. Springer Verlag, Heidelberg.|
|Publisher details||Springer Verlag, Heidelberg|
|Keywords||growth modelling, logistic function, random effects, dendrometer|
Dendrometers are devices which measure the stem radius of a tree continuously. We studied the use of logistic and generalised logistic models for exploring dendrometer data and for automatically determining the onset and cessation dates of radial increase. We used data measured in two stands in southern Finland to test the performance of the models. In the detection task, the generalised logistic models performed well compared to earlier approaches. In addition, the exploratory analysis revealed distinct differences between growth patterns of trees in different calendar years.