Standard trees versus underwood: Historical patterns of tree taxon occurrence in coppice forests

Abstract

AbstractQuestionsCoppice woods were once widespread in Europe. It is usually assumed that underwood tree taxon composition was not directly influenced by people, whereas especially Quercus was promoted among standard trees. However, no work has quantitatively tested these assumptions. Our main question was whether there were any patterns in our data to suggest that certain trees occurred more frequently as standards than as underwood, possibly as a result of management decisions.LocationCzech Republic.MethodsWe constructed a database covering 12,000 hectares of forests around AD 1900. We recorded the occurrence of tree taxa for standards and underwood in 1,507 compartments in 21 forest units. We visualised patterns in these data by bar graphs and chord diagrams. We analysed correlation patterns through association analysis and co‐correspondence analysis.ResultsThere was a statistically significant association between the occurrences of standards and underwood in individual compartments for Quercus, Alnus, Carpinus, Fagus, Larix, Populus tremula and Tilia. Only Quercus standards and underwood showed a positive association. The taxon composition of standards and underwood in forest compartments was statistically significantly correlated.ConclusionsNo simple answer can be given to the question whether or not the taxon composition of standards and underwood in coppice woods differed as a result of management. At the level of individual tree taxa, some trees were more common as standards while others, as underwood. However, the prevalence of Quercus among standards was not confirmed because in our study area Quercus was a common underwood tree as well. At the same time, no significant differences were apparent in the taxon composition of standards and underwood in individual forest compartments. It seems difficult to differentiate between the effects of human selection and indirect selection through natural properties and environmental factors. Our results also highlight the need to carefully consider the properties of written historical sources.