From tree-related microhabitats to ecosystem management: A tree-scale investigation in productive forests in Estonia.

A diversity of microhabitats has been suggested to play a key role in mediating the co-occurrence of trees with specific tree-inhabiting biodiversity, which may further influence ecosystem functioning. However, this triple relationship between tree characteristics, tree-related microhabitats (TreMs), and biodiversity has not been described explicitly enough to set quantitative targets of ecosystem management. The two major approaches directly targeting TreMs in ecosystem management are tree-scale field assessment of TreMs and precautionary management, which both require insights into the predictability and magnitude of specific biodiversity-TreM relationships. To obtain such insights, we analysed tree-scale relationships between the diversity of TreM development processes (four classes: peculiarity; pathology; injury; emergent epiphyte cover) and selected biodiversity variables based on 241 live trees (age range 20-188 years) of two species (Picea abies, Populus tremula) in hemiboreal forests in Estonia. We addressed the diversity and abundance of epiphytes, arthropods, and gastropods; their specific response to TreMs was disentangled from tree age and tree size effects. We found that a relatively small improvement in the biodiversity responses studied was attributable solely to TreMs, and that such contribution was more frequently observed in young trees. Unexpectedly, several age- or size-independent effects of TreMs were negative, suggesting trade-offs with other factors of biodiversity relevance (such as tree foliage suppression due to injuries that created TreMs). We conclude that tree-scale microhabitat inventories have only limited potential to resolve the general problem of providing diverse habitats for biodiversity in managed forests. The basic sources of uncertainty are that microhabitat management is mostly indirect (managing TreM-bearing trees and stands rather than TreMs themselves) and that snapshot surveys cannot address various time perspectives. We outline a set of basic principles and constraints for spatially heterogeneous and precautionary forest management that includes TreM diversity considerations. These principles can be further elaborated through multi-scale research on functional biodiversity links of TreMs.

Forest ditch maintenance impoverishes the fauna of aquatic invertebrates: Opportunities for mitigation.

One of the main factors causing biodiversity loss in wetlands is drainage, nevertheless, even drained areas may provide habitat for aquatic fauna in the form of drainage ditches. Assemblages in ditches are regularly disturbed by ditch maintenance, but the extent of these disturbances and mitigation possibilities are poorly documented. We conducted an experimental study in three commercially managed forest plots in eastern Estonia, aiming to find out how ditch network maintenance (DNM) affects the diversity and assemblages of aquatic macroinvertebrates in ditches and remnant pools, and whether this effect can be alleviated by constructing mitigation waterbodies. For comparison we also collected data from natural pools in three undrained forest plots. Before DNM, ditches supported greater number of higher taxa compared to remnant and natural pools and more strictly aquatic taxa, whereas natural pools in undrained plots supported more Trichoptera shredders. After DNM, the diversity in remnant pools decreased. Moreover, majority of the pools dried out, which resulted in further reduction of the richness and abundance of macroinvertebrates. In ditches the diversity dropped immediately after DNM, but recovered in two to three years. Nevertheless, plot-scale richness and abundance did not completely recover. Assemblage shift in ditches took place right after DNM and remained distinct after the four year survey period. Mitigation pools provided habitat for several taxa (especially Odonata) uncommon in other waterbodies in drained and undrained plots. Our results show that DNM in forests substantially impoverishes habitat availability and reduces the abundance and diversity of aquatic macroinvertebrates. We recommend retaining uncleaned sections in ditches and constructing mitigation pools as tools for supporting wetland biodiversity in drained forests.

Restoration dilemmas between future ecosystem and current species values: The concept and a practical approach in Estonian mires.

Ecosystem restoration is gaining political and economic support worldwide, but its exact targets and costs often remain unclear. A key issue, both for predicting restoration success and assessing the costs, is the uncertainty of post-restoration development of the ecosystem. A specific combination of uncertainties emerges when ecosystem restoration would negatively affect pre-restoration species conservation values. Such dilemma appears to be common, but largely ignored in restoration planning; for example, in historically degraded forests, wetlands and grasslands that provide novel habitats for some threatened species. We present a framework of linked options for resolving the dilemma, and exemplify its application in extensive mire restoration in Estonia. The broad options include: redistributing the risks by timing; relocating restoration sites; modifying restoration techniques; and managing for future habitats of the species involved. In Estonia, we assessed these options based on spatially explicit mapping of expected future states of the ecosystem, their uncertainty, and the distribution of species at risk. Such planning documentation, combined with follow-up monitoring and experimentation, can be used for adaptive management, by funding organizations and for academic research.

Shrubby cinquefoil (Dasiphora fruticosa (L.) Rydb.) mapping in Northwestern Estonia based upon site similarities.

BACKGROUND: Different methods have been used to map species and habitat distributions. In this paper, similarity-based reasoning-a methodological approach that has received less attention-was applied to estimate the distribution and coverage of Dasiphora fruticosa for the region in the Baltic states where grows the most abundant population of this species. METHODS: Field observations, after thinning to at least 50 m interval, included 1480 coverage estimations in the species presence locations and 8317 absence locations. Species coverage for the 750 km2 of directly unobserved area was calculated using machine learning in the similarity-based prediction system Constud. Separate predictive sets of site features (e.g. land cover, soil type) and exemplar weights were calibrated for spatial partitions of the study area (probable presence region, unclear region, proved absence region). A modified version of the Gower's distance metric, as used in Constud, is described. RESULTS: The resulting maps depicted the predicted coverage, the certainty of decision when predicting presence or absence, and the mean similarity to the exemplar locations used while predicting. Coverage prediction errors were smaller in the unclear partition-where the species was mostly absent-than in the probable presence partition, where coverage ranged from 0 to 90%. CONCLUSIONS: We call for methodological comparisons using the same data set.

Long-term impacts of forest ditching on non-aquatic biodiversity: conservation perspectives for a novel ecosystem.

Artificial drainage (ditching) is widely used to increase timber yield in northern forests. When the drainage systems are maintained, their environmental impacts are likely to accumulate over time and along accompanying management, notably after logging when new forest develops on decayed peat. Our study provides the first comprehensive documentation of long-term ditching impacts on terrestrial and arboreal biodiversity by comparing natural alder swamps and second-generation drained forests that have evolved from such swamps in Estonia. We explored species composition of four potentially drainage-sensitive taxonomic groups (vascular plants, bryophytes, lichens, and snails), abundance of species of conservation concern, and their relationships with stand structure in two-ha plots representing four management types (ranging from old growth to clearcut). We found that drainage affected plot-scale species richness only weakly but it profoundly changed assemblage composition. Bryophytes and lichens were the taxonomic groups that were most sensitive both to drainage and timber-harvesting; in closed stands they responded to changed microhabitat structure, notably impoverished tree diversity and dead-wood supply. As a result, natural old-growth plots were the most species-rich and hosted several specific species of conservation concern. Because the most influential structural changes are slow, drainage impacts may be long hidden. The results also indicated that even very old drained stands do not provide quality habitats for old-growth species of drier forest types. However, drained forests hosted many threatened species that were less site type specific, including early-successional vascular plants and snails on clearcuts and retention cuts, and bryophytes and lichens of successional and old forests. We conclude that three types of specific science-based management tools are needed to mitigate ditching effects on forest biodiversity: (i) silvicultural techniques to maintain stand structural complexity; (ii) context-dependent spatial analysis and planning of drained landscapes; and (iii) lists of focal species to monitor and guide ditching practices.