Impact of habitat quality on forest plant species colonization

The impact of habitat quality and site history on the recolonization potential of ancient-forest plant species on abandoned farmland was studied in the forest of Ename, Flanders, Belgium. With the exception of a network of fringe relics (linear elements mainly along exploitation roads), our study area was cleared and converted to arable land ca. 1850. From 1869 onward, most fields were gradually abandoned, resulting in a progressive, partly spontaneous reforestation. Each of the 42 actual forest parcels (amounting to 62 ha) was surveyed and a total of 466 plant species were inventoried. Twenty seven of these were identified as ancient-forest plant species and cataloged in a separate subset. Additionally, the spatial distribution of six ancient-forest plant species (Anemone nemorosa, Corylus avellana, Hyacinthoides non-scripta, Paris quadrifolia, Mercurialis perennis, Vinca minor) was systematically surveyed and digitized in a GIS environment. Habitat quality was assessed on the parcel level using intrinsic soil variables on the one side, and historically related variables (length of the agricultural-occupation period, length of woody fringe relies, and total length of fringe relies s.l.) on the other. Soil texture had a major impact on the duration of agricultural land use after deforestation. Soil phosphate content and pH are positively correlated with the duration of the agricultural land use. The number of ancient-forest plant species was negatively affected by the length of the agricultural-occupation period and soil phosphate content, and positively by the total length of the surrounding fringe relies. The same trends are observed studying the systematically surveyed ancient-forest species. We propose that soil phosphate content affected ancient-forest plant species distribution, because it stimulates vigorous vegetation development and as such has a definite effect on evolving competitive plant relationships. Using raster-G.I.S analysis tools, mean and maximum colonization distances and approximate mean and maximum colonization rates per century were calculated for each systematically surveyed species. We conclude that not only seed dispersal capabilities, but also site quality variables play an important role in the colonization process of ancient-forest plant species. In the short term, afforestation of previously heavily fertilized farmland will not result in floristically diverse and, thus, valuable forest habitats. The relatively immobile soil phosphate represents a major barrier. (C) 1999 Elsevier Science B.V. All rights reserved.