You are here: Home Projects A2) Retaining Forest Structure

A2) Retaining Forest Structure

Retention of structural elements in selectively used forests

Jürgen Bauhus & Patrick Pyttel (until 2020)
Doctoral researchers: Thomas Asbeck (2016 - 2019), Josef Großmann (associated; since 2018) & Andreea Petronela Spinu (since 2019)

University of Freiburg, Faculty of Environment & Natural Resources, Institute of Forest Sciences,
Chair of Silviculture

Background & overview

Tree-related microhabitats (TreMs) are specific structures that occur on standing trees and are an essential habitat component for many forest-dwelling species. Some examples of TreMs are cavities, large dead branches, loose bark, epiphytes, bracket fungi, cracks, or trunk rot. Live or dead trees that support or have the potential to develop TreMs are called habitat trees. The development of both the habitat trees and their TreMs influence the habitat provision for dependent species.

In this project we focus on:

  • the determinants of the occurrence of microhabitats on trees; for example tree species and dimension, environmental variables, and spatial distribution of habitat trees;
  • the influence of forest or tree management intensity and time since cessation of management in reserves on TreM development;
  • the temporal dynamics including longevity of habitat trees and TreMs;
  • the indicator value of TreMs for the richness and diversity of different forest-dwelling taxa.


The subproject comprises at this stage three different PhD projects :

  1. "Tree-related microhabitats as selection criteria for habitat trees in close-to-nature forest management" conducted by Thomas Asbeck (completed in 2019);
  2. "Quality and longevity of habitat trees and their tree-related microhabitats" conducted by Andreea Spinu (commenced in 2019);
  3. "Formation and development duration of microhabitats at single tree and forest stand level" conducted by Josef Grossmann (commenced in 2018).


Tree-related microhabitats as selection criteria for habitat trees in close-to-nature forest management

In the first phase, we have focused on the determinants of microhabitat abundance and diversity on potential habitat trees in 135 one-hectare plots of the ConFoBi design. For that purpose, a full inventory of tree-related microhabitats (TreMs) was carried out according to a detailed catalogue on the largest 15 trees per plot, which had been preselected on the basis of remote sensing data of crown dimensions. An increasing mean diameter at breast height (DBH) of habitat trees led to a higher abundance and diversity of TreMs. The occurrences of eleven different TreMs were related to forest management, forest type, altitude, and mean DBH. Hence, the average abundance and diversity of TreMs may be predicted with readily available forest attributes (Asbeck et al. 2018). We further found that retaining clumped live trees does not provide a greater abundance and richness of most TreMs than dispersed live individuals. These results show that it is more important to select the right retention trees in appropriate quantities rather than focusing on grouping them, which may be done for other reasons. For the Black Forest region, future forest management intensity will have a more pronounced influence on microhabitat provision than climate change, as was found in an interdisciplinary, collaborative analysis with the projects C1 and B6 (Augustynczik et al. 2018). In a collaborative project with A1, it was found that the abundance and diversity of TreMs cannot be sufficiently well predicted with remotely sensed variables including forest structure (e.g. tree height, stand density) and terrain variables (Frey et al., 2020). There was only a weak significant relationship between several predictors and the TreM abundance and diversity. This shows that we still need to identify and quantify TreMs based on terrestrial inventories. However, laser scanning may also be used in the future for a range of TreMs.



Quality and longevity of habitat trees and their tree-related microhabitats

Information on mortality of habitat trees was so far available only for regions such as North-America, Scandinavia, or Australia, where trees have been retained in clearcuts, but not for continuous cover forests in Europe. This information is of critical importance because habitat trees are meant to provide habitat structures for a very long period, ideally until a new generation of mature and over-mature trees have developed. However, if habitat trees already die after a short period of time or if TreMs disappear, habitats for species will be lost or impaired. When selecting habitat trees for retention, there can be trade-offs between the provision of certain highly-valuable TreMs such as bracket fungi, which can host hundreds of other species, and the longevity of habitat trees that are attacked by these fungi.

The prediction of TreM formation was recently investigated with an approach based on survival and reliability analysis (Courbaud et al., 2017). However, the longevity of TreMs has not been studied at all. No time series of the individual development of TreMs exist at this stage. Yet certain TreMs may be only transitory features of habitat trees. For example, loose bark pockets that offer roosting sites for bats may only last a few years. Likewise, dead branches will only stay in the crown until they have sufficiently decayed to brake off. Other TreMs such as tree hollows may be very long-lived. The longevity of different TreMs, which are likely to develop at different stages of development of individual trees may influence the selection and recruitment of habitat trees. For example, one might select trees of different dimension, species, and anticipated longevity to provide a continuity of TreMs over time. Therefore, we focus in this second part of the RTG project on the longevity of habitat trees and their TreMs.

Research questions and hypotheses

  • The overall abundance and diversity of TreMs is significantly higher on dead trees.
  • Longevity of TreMs that are related to wood properties (e.g. crown dead wood, hollows, etc.) is higher in conifers (fir and spruce) than in broadleaves (beech and maple).
  • The mortality of habitat trees is higher than that of average (and smaller) trees, since trees with larger DBH are more likely to be affected by environmental stressors and their vitality is decreased by occurrence of some TreMs (sap runs, occurrence of bracket fungi, mistletoe presence).
  • Habitat tree mortality differs among tree species and is influenced by the landscape context (eg., altitude, local climate).

Approaches, methods, and linkages

The assessment of habitat tree mortality will be based on a combination of repeated ground-based inventories of all ConFobi plots and remote sensing approaches in the wider landscape to quantify the attrition of trees that have been retained, either individually and scattered or in clumps. Based on this, we will develop models that include habitat tree attributes (e.g. species; dimensions; predisposing factors, such as the presence of bracket fungi, etc.), site attributes (e.g. soil type, exposition), and spatial arrangement of habitat trees to predict their mortality.
The assessment of TreM longevity will require the development of a range of different approaches for the different types of TreMs, which are currently developed. Additionally, new techniques for a more objective TreM assessment will be developed, including detailed measurements of the sizes of TreMs (i.e. terrestrial laser scanning, jointly with A1).


Formation and development duration of microhabitats at single tree and forest stand level

In order to effectively protect and conserve biological diversity, knowledge about the origin and persistence of tree microhabitats is of great importance. The overall goal of this project is to increase our knowledge about frequency, distribution and temporal development of tree microhabitats. This will facilitate an assessment of the relative ecological value of tree microhabitats. In the framework of this project, known factors influencing the development of tree microhabitats will be specified and recommendations for forestry practice will be derived. One focus of his work are habitat tree groups and small forest reserves from the Veteran-Tree and Deadwood-Concept (AuT-Konzept) which is applied in the state forest of Baden-Württemberg. Another part of his project deals with structural diversity of urban trees. During a case study in Montréal, Canada he quantified tree microhabitats on urban trees affected by different management approaches. This study has been carried out in collaboration with Christian Messier, who was supported by an Alexander von Humboldt prize.

So far, data collection and analysis of TreM s in urban forests has been completed. This study has been carried out in collaboration with Christian Messier, who was supported by an Alexander von Humboldt prize. Data collection of so called habitat tree groups is also finished, analysis started. The work in this project is closely coordinated with the second project of A2 (Andreea Spinu).




In the future, A2 intends to analyse the influence of different species of habitat trees, dimension, and vitality on their surroundings, in particular the regeneration and regrowth stand. This work will provide a basis for assessing the economic consequences of habitat-tree retention.