External management effects on the stand structure of protected forest patches

Abstract

Questions In production forest landscapes, threatened species are increasingly confined to fragments of natural forest. Protecting such fragments is a priority to halt further biological degradation, but its effectiveness depends on external impacts on the fragments. In this study, we ask: how does woody vegetation in protected old‐forest fragments vary depending on management of their adjacent areas, also in combination with natural site conditions? What are the management impacts on habitat quality of the fragments in the long term? Location Pinus‐dominated and Picea mixed‐wood forests in Estonia (hemiboreal Europe). Methods Forest structure and microclimate were measured at the edge and interior of protected old‐forest fragments. Woody vegetation layers, shrub species composition, coarse woody debris, and presence of cut stumps were measured in 107 fragments (adjacent to clearcuts, partial cuts, and unthinned production forests) and in 20 control stands embedded within natural forest. Air temperature and humidity were measured at the edges vs the interior of mixed‐woods. Edge age effects were explored along 80 years of post‐harvest chronosequences. Results All edge types studied had multiple, partly forest‐type‐specific, external effects on stand structure. Some clear‐cutting effects on woody layers reversed in later decades, but there was also persistent impoverishment of deadwood pools of the fragments probably related to unreported tree removal. There was a likely indirect pathway of edge effects through overstorey loss, and shrub cover was reduced in the long term. Exotic species were rare. Edges at clear‐cut stands influenced humidity regimes rather than temperature regimes. Conclusions The moderate external effects on stand structure were tightly linked to low overstorey mortality and a lack of exotic species across the forest landscapes studied. Changes in these key factors constitute risks for the future. Protecting set‐asides from occasional tree cutting, and establishing forest buffers around small fragments and as a precautionary strategy are current priorities in these forest landscapes.

Modeling forest landscape futures: Full scale simulation of realistic socioeconomic scenarios in Estonia

“…We optimized the computation effort by switching between the vector and raster data types using the R packages sf [36] and terra [37]. Forest edge areas were defined as being within 30 m from open land, clear-cuts or early successional stands (up to 20 years old) [38].…”