Deadwood quantity, quality and decomposition processes in Picea abies L. dominated Natural Forest Reserves in Austria

Schlüsselbegriffe: Nekromasse, Zerfallsgeschwindigkeit, Naturschutz, Habitatqualität, Kohlenstoff, Biodiversität, Urwälder, Gemeine Fichte

Available at https://doi.org/10.53203/fs.2602.3

Download as PDF

See below the issue 2/2026 as E-Paper or have a look at our E-Paper archive dating back to 1955.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

Abstract

Forest biodiversity, habitat value and carbon sinks are strongly linked to deadwood quantity (stocks) and quality (decay class). Repeated measurements of unmanaged forests allow to quantify the natural range of deadwood properties for different ecosystems and describe the dynamics over time, which can support recommendations in conservation and carbon management. In our study we repeatedly investigated the growing stock and deadwood conditions in six natural forest reserves (NFR) in Austria using 167 permanent monitoring plots to examine the changes over time. Apart from one, all studied NFR showed constant live tree volume stocks non-significantly changing over 12 years, with a highly variable small tree density, suggesting that these forests are in optimal to climax/terminal phase. Standing deadwood stocks, including stump deadwood, have been increasing in all but one NFR and lying deadwood stocks increased in all but two NFRs. Summarizing measured above ground deadwood and live volume and using deadwood density and carbon fraction data from literature suggest that two NFRs are on average accumulating carbon at about +1 tC/ha per year, and two NFRs have decreasing live and dead carbon stocks at -0.2 to -0.4 tC/ha per year. A consistent relationship between time since death (TSD) and decay class (DC) was found across all sites. Based on the relations between TSD and DC, a model for future deadwood volumes in different decay classes in Norway Spruce-dominated forests is made. Recommendations for future research and monitoring in NFR are given.

Zusammenfassung

Biodiversität, Habitatqualität und Kohlenstoffsenken von Wäldern stehen in einem klaren Zusammenhang mit der Totholzmenge (Bestand) und Qualität (Zersetzungsklasse). Wiederholungsaufnahmen in unbewirtschafteten Wäldern ermöglichen es die natürliche Variabilität von Totholz in unterschiedlichen Waldökosystemen zu quantifizieren und die zeitliche Dynamik zu beschreiben, was bei der Erarbeitung von Empfehlungen im Naturschutz- und Kohlenstoffmanagement helfen kann. In dieser Studie haben wir den Lebend- und Totholzbestand in sechs Naturwaldreservaten (NWR) in Österreich, auf Basis von 167 permanenten Monitoringflächen, hinsichtlich dessen Entwicklung über die Zeit untersucht. Bis auf eines zeigten alle NWR einen konstanten Lebendvorrat ohne signifikante Veränderung über einen Zeitraum von 12 Jahren. Die Stammzahl kleiner Bäume variierte stark, was daraufhin hinweist, dass sich diese Wälder in der Optimalphase bzw. Klimax-/Terminalphase befinden. Stehender Totholzvorrat, einschließlich Stocktotholz, hat bis auf einem in allen NWR zugenommen und der liegende Totholzvorrat hat in allen bis auf zwei NWR zugenommen. Der gesamte gemessene Lebend- und Totholzvorrat weist in Kombination mit den aus der Literatur ermittelten Totholzdichte und Kohlenstoffgehalt darauf hin, dass die mittlere Kohlenstoffmenge in zwei NWRs um etwa +1 tC/ha pro Jahr angestiegen ist, während in zwei NWRs die Menge an Lebend- und Totholz-Kohlenstoff um etwa -0.2 bis -0.4 tC/ha pro Jahr gesunken ist. Ein deutlicher Zusammenhang konnte zwischen der Zeit seit dem Absterben (TSD) und dem Zersetzungsgrad (DC) auf allen Flächen beobachtet werden. Aus der Beziehung zwischen TSD und DC wurde ein einfaches Modell entwickelt, um zukünftige Totholzvorräte in subalpinen Fichtenwäldern abzuschätzen. Empfehlungen für die zukünftige Forschung und Monitoring in NWR werden gegeben.

1 Introduction

Upcoming changes in forest policy, public demand for more protected areas and landowner decisions leading to (partial or temporary) abandonment of active forest management will likely lead to changes in tree species composition and forest structures across various forest types in Europe. The amount of deadwood in forests will likely increase due to no or more extensive forest management, by setting-aside forests for conservation or carbon storage purposes, prolonging rotation periods and/or retaining harvest residues on-site. Increases in deadwood stocks can be also expected from increasing frequency and severity of disturbances (Senf and Seidl, 2021), such as bark beetle outbreaks or storm damage, that will likely become more under climate change will lead to an increasing volume of standing and lying deadwood, depending on disturbance agent and salvage-logging. As deadwood is also one of many indicators for “old-growthness” or naturalness (Borghi et al., 2023; Lassauce et al., 2011; Vandekerkhove et al., 2022), this can also impact biodiversity assessments. However, as habitat value and thus biodiversity is linked by rather complex relations with deadwood quantity and quality (Müller et al., 2015; Vítková et al., 2018), comprehensive information on deadwood and its decomposition is needed to evaluate the consequences of conservation management or extensive forest management.

In Austria, since 1995 almost 200 Natural Forest Reserves (NFR) have been established. An earlier study published in this journal, synthesized data measured in 2009 as part of the project ELENA from five NFRs established between 1997 and 1999 (Ruprecht et al., 2012). In this study, we built on that pilot study to understand changes in deadwood quantity and quality and explore possible limits to deadwood volume stocks using repeated observations of unmanaged natural forests dominated by Norway spruce (Picea abies L.) due its wide distribution in Europe and important contributions to forest carbon stocks and economic value of European forests (Caudullo et al., 2017; FOREST EUROPE, 2020). Norway spruce is still the most widespread tree species in Austria and accounts for almost half of Austrian forest area, despite widespread decline and mortality events (Jandl, 2020). Deadwood volume stocks can be as large or even larger than live tree stocks, depending on development stage, disturbance regimes and decay rates (Harmon et al., 2020; Wijas et al., 2024). While carbon storage in deadwood decreases as decay progresses (and deadwood density decreases), deadwood volume can represent a considerable carbon storage in forests and increasing deadwood carbon stocks contributed to global temperate and boreal forests being a carbon sink between 1990 and 2020 (Pan et al., 2024). For long, live trees have been the focus of carbon assessments, being the primary producers, representing usually the largest carbon component (after organic and mineral soil) and due to easier sampling and biomass estimates, but the importance of dead trees for the forest carbon cycle are becoming more and more acknowledged (Seibold et al., 2021; Wijas et al., 2024). The contributions of deadwood to forest carbon sink depend on changes in deadwood quantity (stocks) and quality (decay stage, decay class). Deadwood decay class is a potent proxy for amount of dry biomass and carbon stored in 1 m³ deadwood and can be assessed in the field (Fravolini et al., 2018; Öder et al., 2021). Increasing deadwood volume stocks will likely be associated with increasing deadwood carbon stocks, unless the deadwood decay classes shift towards more advanced decay stages (Neumann et al., 2023). Thus, assessing deadwood quantity and quality over time allows us to track the contributions of deadwood to carbon stocks and habitat value in NFRs in Austria and contrast them to live tree stocks.

The objectives of this study are,

(1) to quantify deadwood quantity and quality, separated by standing and lying deadwood for Norway spruce-dominated NFRs in Austria,

(2) to assess changes in live tree and deadwood volume and carbon in Norway spruce-dominated NFRs in Austria over a period of 12 years, in comparison with literature data from other NFRs, and

(3) to make predictions of deadwood dynamics (i.e., changes in deadwood volume over time) by estimating decay rates using decay classes and time since death for Norway spruce-dominated NFRs in Austria.

2 Methods

We used empirical data sampled in six NFRs (Goldeck n = 30, Hutterwald n = 29, Kronawettgrube n = 19, Krimpenbachkessel n = 25, Laaser Berg n = 21 and Schiffwald n = 43, with a total number of 167 sample plots, each measured two times) representing sub-alpine Spruce forests (Piceetum) and montane Spruce-fir forests (Abietetum) with an elevation range from 1000 to 1700 m (Fig. A1). We reanalyzed data used by Ruprecht et al. (2012) to compare the forest conditions in 2008-10 with those in 2021, representing on average 12 years of continuous monitoring using permanent sample plots. Some of the collected data was reported in student thesis (Götze, 2013; Hardalau, 2022; Huber, 2022; Kasseroler, 2011; Starcevic, 2022; Winter, 2009).

 

2.1 Live and deadwood volume

We grouped deadwood into 5 cm diameters classes, consistent with the methodology used by (Ruprecht et al., 2012) for live trees. We removed standing deadwood with diameter at breast height (DBH) smaller than 5 cm and lying deadwood with mid diameter smaller than 10 cm, to be consistent and results comparable between regions and plots. We calculated stem density, basal area and volume per hectare based on a sample plot size of 300 m² (radius 9.77 m). For lying deadwood, we calculated volume assuming the shape of a cylinder, as mid diameter was measured. If mid diameter or deadwood length was not measurable, we assumed it to be the average of all other lying deadwood pieces in that plot, respectively. If both diameter and length were not measurable (e.g. for heavily decomposed pieces or irregular shaped deadwood fragments), we ignored this deadwood piece for volume assessment. For standing deadwood and live trees, we calculated volume using form factors, FF (Eq. 1). Tree height, H, was measured for all standing live and dead trees.

V = FF DBH² π / 4 H        (1)

For standing deadwood, we accounted for changes in stem taper due to crown or stem break by varying FF. For living trees, we calculated FF using the equations of Pollanschütz (1974) also used for the Austrian National Forest Inventory. For standing dead trees with branches still attached to crown, we assumed FF of 0.4, equal to living trees. For standing dead trees without crown, we assumed FF to be 0.7 and for dead stumps and lying deadwood, we assumed FF to be 1, thus having a shape of a cylinder. For lying deadwood, the mid diameter, MD, was used instead of DBH and length, L, instead of H, for calculating volume (Eq. 1). Relations between DBH/MD and H/L are shown in Figs A2-4. We used mean live and deadwood volume stocks reported by Oettel et al. (2020) from low subalpine spruce forests (number plots = 44) and spruce-fir-beech forests (n = 114), to compare our data against literature data on other Austrian NFRs with similar forest types. We note that data on the NFRs reported in Oettel et al. (2020) were not included in Ruprecht et al. (2012) or this study. We used dry wood density and carbon fraction of live trees to estimate their carbon stocks after Neumann et al. (2016). Since there is no published Austrian data available for deadwood density and carbon fraction grouped by the decay classes used here, we used data collected in Switzerland, for similar forest types and management systems (Hararuk et al., 2020). We used density and C fraction of “Picea abies” for all conifers and “Fagus sylvatica” for all broadleaves.

We used histograms (pooling trees into 5-cm classes, consistent with the pilot study) and boxplots to compare changes in stem density and volume stocks of live trees, standing dead and lying dead trees. Kruskal-Wallis tests were used to assess statical differences between measurement times for the six NFRs, as data was non-normally distributed and positively skewed. We calculated differences in two ways, 

(1) between mean values of deadwood on plots grouped by NFR and sampling year and 

(2) by pair-wise comparing plots with the same ID, to apply greater scrutiny to assessed changes over time. 

Option (2) further allowed us to check for significant changes.

 

2.2 Deadwood decay classes and deadwood dynamics

Information on deadwood decay rates, transition time between decay classes or deadwood residence time (Kruys et al., 2002; Russell et al., 2014, 2013; Vrška et al., 2015) is required to understand changes in deadwood quantity (volume) and quality (decay class) over time. For the NFRs used in this study, an assessment of the time since death, henceforth abbreviated with TSD, in years of observed deadwood pieces was made on site. Deadwood decay was assessed using a five-class decay classification system (Ruprecht and Vacik, 2009), expanding the four-class system (Albrecht, 1991; Rauh and Schmitt, 1991) with a fifth class for highly decayed, humus-like deadwood with in total five decay classes, henceforth abbreviated with DC (Table 1). We pooled deadwood volume by decay class, to track changes in deadwood quality over time. We linked DC and TSD using regression functions, to check the time needed for state-transitions through the here used DC system, after replacing letters (A-E) with numbers (1-5) to allow fitting regression functions. We checked for non-linear behavior using second-order polynomial functions, in comparison to linear functions. We considered using mixed effects models, but decided to use separate models to demonstrate the effects of region, deadwood type and species on the relations between TSD and DC. By using ordinary least squares, we assume that independent variable (DC) is error-free. We then used the functions linking TSD and DC for estimating changes in deadwood volume over time. We calculated Nagelkerke pseudo-R² to assess model performance. All analysis and visualization of results was done using R statistical software (R Development Core Team, 2021).

3 Results

3.1 Size variations of deadwood in Spruce-dominated nature forest reserves using two measurement periods

Stem density distributions of live trees averaged for the six NFRs, separated for the first measurement and remeasurement about 12 years later, reveal slight changes for higher diameter classes and more pronounced changes for live trees smaller than 10 cm DBH, especially for Goldeck and Krimpenbachkessel (Fig. 1). For NFR Krimpenbachkessel, we observed gains in live tree density (predominantly driven by the 0-5 cm class), while all other NFRs lost live tree density (see also Table 2).

We also observed notable shifts in the standing dead trees (Fig. 2) and losses in live trees were commonly associated with gains in standing dead trees (with Goldeck being an exception, Table 2). Again, most change we can observe for smaller diameters. Gains in live trees and losses in deadwood density suggest NFRs are predominantly in “early optimal phase” and “regeneration phase” according to literature (Korpel, 1995; Mayer and Neumann, 1981; Zukrigl et al., 1963), where the forest are accumulating live tree volumes and growth rates are exceeding mortality. Losses in live trees and gains in deadwood would suggest the analyzed NFRs being in “climax phase” and “decay phase”, while losses in both live trees and deadwood would suggest them being in “collapse phase” (see Korpel, 1995; Scherzinger, 1996).

Changes in stem density by diameters class do not reflect changes in height (length for lying deadwood) and volume. Thus, we next looked into changes in volume (proxy for biomass and carbon stocks, assuming that density of live and dead trees are comparable) over time. Decreasing stem density of live trees larger 10 cm (Fig. 1) observed for all NFRs coincide with constant or decreasing volume stocks of live trees (Fig. 4). We show the mean volume stocks by NFR separated in Table 3. The decrease was not significant, except for NFR Schiffwald, which also had the largest loss in live trees > 10 cm DBH (Table 2). Only in NFR Hutterwald mean live tree volume stocks increased notably over 12 years of remeasurements (see also Table 3). Variation of standing and lying deadwood was clearly skewed and more variable than live tree volume. NFRs Schiffwald, Kronawettgrube and Laaser Berg notably increased in deadwood stocks, while the increase was significant only for Schiffwald and Laaser Berg (Fig. 4). The pair-wise comparison of plots yielded in general similar results than using averages by NFR (Table 3), however for NFR Schiffwald the change was not strongly significant (p 0.01). Only for the lying deadwood in Laaser Berg also the pair-wise comparison confirmed the significant differences from Fig. 4. In summary, carbon stocks in live trees and deadwood were increasing in three NFRs (Goldeck, Hutterwald, Kronawettgrube), with the two latter exhibiting increases > 1 tC/ha per year. NFRs Schiffwald, Laaser Berg and Krimpenbachkessel exhibited decreasing live tree and deadwood carbon, with the two latter showing losses with less than -0.2 tC/ha per year. This calculation ignores potential changes in organic and mineral soil layers, as well as ground vegetation.

 

3.2 Comparing stocks of Norway spruce-dominated nature forest reserves with literature data

Comparing with mean volume stocks reported by Oettel et al. (2020) for similar forest types makes it obvious that some NFRs studied here in detail (and not included in the analysis of Oettel et al., 2020) are in different development phases, than the NFRs reported in Oettel et al. (2020) (henceforth called NFR_Oettel2020). NFR Krimpenbachkessel, the only NFR predominantly representing spruce-fir-beech forests, had considerably lower live tree volume stocks than the spruce-fir-beech NFR_Oettel2020. On the other hand, NFR Goldeck with elevation above 1100 meter and dominated by Norway spruce is a typical subalpine spruce forest, but has mean stocks much larger than live tree volume stocks in subalpine spruce NFR_Oettel2020. A value of 30 m³/ha deadwood is often used as a target for deadwood quantities in forest policy (Müller and Bütler, 2010). Here assumed that out of the 30 m³/ha threshold determined for both lying and standing deadwood for Norway spruce forests 24 m³/ha were lying deadwood and 6 m³/ha standing deadwood, based on separate images in the original literature (Ranius and Jonsson, 2007). Our analysis thus underpins that protected and unmanaged forests may have higher or lower stocks compared to literature reference values, depending on site conditions, development stage and disturbance history, and this extends to both live trees and deadwood. The spatially highly variable nature of deadwood makes assessing changes over time particular challenging.

3.3 Deadwood dynamics in Norway spruce-dominated nature forest reserves

Deadwood in forests has not only different diameter and length (volume) and position (standing, lying), but also varying quality, commonly assessed in the field using a decay class system (e.g. Harmon et al., 1987; Müller-Using and Bartsch, 2007; Neumann et al., 2023). Decay classes represent differences in wood density, moisture content, carbon fraction and many other important deadwood traits, including habitat value for many different insects (Jomura et al., 2022; Neumann et al., 2023; Shorohova et al., 2016). Since Oettel et al., (2020) did not report decay classes, we compare here our observations with the volume stocks by decay class, assuming every decay class has the same share (20% since five decay classes) using the earlier used mean volume stocks of NFR_Oettel2020. Conceptually trees after their death start in decay class A (see description in table 1) and move through decay class B-E, until they are fragmented, incorporated in the soil or humus layers and thus became unmeasurable using transects or plots. We note, that there are exceptions, such as internal decay of living standing trees or shedding dead decaying large limbs from live trees or variations according to site and microclimatic conditions. Variations in deadwood quality (using decay class as proxy) help assess habitat value and biodiversity in more detail, than only using total deadwood volume stocks.

The NFRs studied here had more deadwood volume in middle decay classes B-D, while fresh and very heavily decayed deadwood volume (decay classes A and E) was more rarely observed, especially at the first measurements in 2008-2010. 12 years of continuous protection resulted in a more balanced deadwood volume distribution especially for lying deadwood, except for decay class E (Fig. 5). We note again the skewness in deadwood volume, which makes it challenging to track changes over time. The absence of standing deadwood with decay class E is not surprising, since advanced decay reduces wood density and makes fragmentation likely (transferring volume into the lying deadwood pool).

We analyzed many functions relating TSD and DC for lying and standing deadwood, by species, elevation and/or size class (Fig. 6, A6-A10). We found no conspicuous differences of TSD by NFRs, but note that NFRs Schiffwald, Laaser Berg and Kronawettgrube had somewhat smaller slopes and thus faster decay (Fig. A5). There were also no significant differences of functions relating TSD and DC between lying and standing deadwood using deadwood type as dummy variable in linear regression functions (p = 0.172), owing to the large variation of TSD in the single DCs (Fig. 6). We also found no significant influence of species on functions relating TSD and DC (see also Figs A8-A9) nor clear non-linear relationships (Fig. 6). Grouping by elevation (pooled into 200 m classes) did not reveal clear patterns (Fig. A10). The variation of TSD was large especially for decay classes B-D (equivalent to 2-4 in Fig. 6) ranging from <5 to >40 years. For standing deadwood a second order polynomial function improved pseudo-R² (coefficient of determination) by 0.003 compared to the linear trend function. Linear functions without intercept had similar pseudo-R² than functions with intercept, but better captured observed TSD for lower DC.

We decided to use a linear trend function without intercept pooling all tree species and deadwood types for estimating the time needed to pass from one decay class to the next higher decay class, which is on average 5 years, rounded to integer (TSD = 4.754 DC or DC = TSD / 4.754, pseudo-R² 0.564). Our reasons include 

(1) the large variation of TSD by DC (also within NFRs, see Fig. A6), 

(2) small differences of functions fitted for separate deadwood types (Figs 6, A6-A9) and 

(3) rare mixed tree species, accounting for 15% of all observed deadwood pieces. 

63% were identified as Norway spruce and for the remaining 22% of observed deadwood pieces, species could not be determined. Assuming the share of Norway spruce among unidentifiable deadwood pieces is similar to that of identifiable pieces (86%), result in about 82% of all deadwood pieces originating from Norway spruce trees.

4 Discussion

4.1 Dynamic rather than static NFRs

To enhance the biodiversity condition of coniferous forest in Austria related to deadwood it is important to gain knowledge on reference values derived from unmanaged forests under natural conditions. A key objective for establishing natural reserves in Austria was to maintain forest biodiversity and provide a reference for discussions on threshold values for sustainable forest management (Frank and Müller, 2003). Few studies have investigated forest dynamics of strictly unmanaged Norway spruce subalpine forests. However, as Norway spruce mortality is expected to increase in the future due to climate change related bark-beetle outbreaks, storms and droughts (Štraus and Bončina, 2025), it is important to understand natural forest dynamics under the past climatic conditions to design close-to-nature management approaches for the future. In this context the findings of our study can support this discussion as it revealed dynamical changes over time of the studied NFRs, under formal protection since 1997-1999 (Ruprecht et al., 2012), based on permanent monitoring plots. This has implications on carbon storage as well as habitat value of these forests. Apart from one, all studied NFR showed constant live tree volume stocks non-significantly changing over 12 years, suggesting that these forests are in optimal to climax/terminal phase (Korpel, 1995; Mayer and Neumann, 1981), based on the measurement of the monitoring plots. Only NFR Schiffwald show indications being in early decay phase, since live stocks are significantly declining and standing deadwood stocks are significantly increasing. Our data on live and dead stocks suggest that NFR Hutterwald and Kronawettgrube are a carbon sink, while Krimpenbachkessel, Laaser Berg and Schiffwald are a carbon source and for Goldeck losses in live carbon are compensated by gains in deadwood. We note that this assessment is based on live trees and deadwood only and based on the available monitoring plots ignoring changes in carbon pools in the soil. More NFRs may be become a C source in the near future, due to higher temperatures and changing disturbance regimes, identified as main drivers for C losses in unmanaged mountain forests (Albrich et al., 2022). Currently the studied NFRs have similar carbon stocks than other Norway spruce-dominated protected forests in Austria (Oettel et al., 2020). Only Goldeck despite representing a typical Spruce-dominated montane forests (see Fig. A1), has considerably higher live stocks (due to higher basal area), more similar to Spruce-fir-beech forests based on literature data. 

Various insects and lichens have a strong preference for sun-exposed stumps and snags (Thorn et al., 2017; Vogel et al., 2021). For maintaining biodiversity related to deadwood it is therefore important to keep a good balance between standing and lying deadwood compartments (Rahman et al., 2008). The size distributions of live trees reported in Ruprecht et al. (2012) were in general similar to that of dead standing and dead lying trees reported here and smaller sized trees were more numerous than larger-sized trees. Small tree density was more variable over time, pointing towards them being more affected by competition and/or browsing compared to larger trees in the upper canopy. In NFR Krimpenbachkessel, a pronounced increase in stem density of live trees smaller than 5 cm DBH, compensated for losses in stem density of larger trees (Kasseroler, 2011) and resulted in a general increase of live tree density over 12 years. Stem density of trees >10 cm DBH was decreasing for all NFRs, albeit at different rates. The number of large trees >75 cm DBH was increasing for all NFRs, with likely positive outcomes for organisms associated with large, old and habitat-bearing trees and old-growth conditions. Larger trees also tend to have more tree-related microhabitats, such as cavities, cracks or dead branches (Larrieu et al., 2018).

Deadwood volume measured in 2021 accounted for on average about 20% of mean live tree volume and thus a somewhat higher share than at the first measurement of 10-20% (Götze, 2013; Kasseroler, 2011; Ruprecht et al., 2012; Winter, 2009). Thus since 2008 in the studied NFRs the habitat value for saproxylic organisms, depending on deadwood (Müller et al., 2020), most likely has increased. Standing and lying deadwood stocks have been increasing for all NFRs (except for lying deadwood in NFR Krimpenbachkessel). Such changes in deadwood volume stocks are the result of the mass balance of deadwood input and deadwood losses (Kušar and Neumann, 2024). Compared to the results of Oettel et al. (2020), which revealed an average deadwood volume of 41.06 m³/ha for low subalpine spruce forests in Austria, the average deadwood amounts in this study are higher. However, as indicated by Nilsson et al. (2002) on many forest sites 30-50 m³/ha of standing deadwood and about 100 m³/ha of lying deadwood were probably very plausible amounts before European forests were subjected to human exploitation. In this context the average deadwood amounts observed in the NFRs are still low, with some exceptions indicating the possible ranges. Also the amount of lying deadwood typically predominates in unmanaged forests (Rimle et al., 2017; Vacek et al., 2015), but the time passed by since no management activities have been observed in the NFRs, might be still too short for a larger accumulation of lying deadwood (Kürsten, 2015; Parigger, 2021; Steininger, 2017). The monitored period of in total 30 years might be still too short to draw conclusions about the general natural dynamics in the forest and there may be other possible trajectories in natural development of these forests (Vacik et al., 2009).

 

4.2 Estimating deadwood dynamics using functions relating TSD and DC

In this study we were able to derive mean transition or residence time of deadwood in decay classes using the fitted models of time since death (TSD) and decay class (DC). We found no notable deviations between deadwood type, species, elevation and NFR in TSD and DC relations, presumably due to simple field assessments of TSD. The relationships between TSD and DC were linear and non-linear functions did not improve considerably the explained variation and were subject to multi-collinearity (VIF > 30). We note that determining TSD accurately is challenging and may require time-demanding and costly techniques, such as dendrochronology or radiocarbon dating (Petrillo et al., 2016). Keeping the above in mind, the consistent relationship of TSD and DC, across sites, field crews and species is remarkable. It was unexpected to see no clear species-specific effect in the relations of TSD and DC, as literature points towards species and associated resistance of wood to decay (due to heartwood, extractives or other decay-inhibiting factors) can explain variation in deadwood decay rates, unexplained by site conditions (e.g. Edelmann et al., 2023; Harmon et al., 2020). Literature also suggest that deadwood decay slows as air temperature decreases (Harmon et al., 2020; Seibold et al., 2021), which is not supported by the results of this study, using elevation as proxy, which is commonly negatively correlated with temperature in Austria.

Until more detailed empirical data on deadwood decay is available for Austria, for instance using in-situ deadwood experiments (Edelmann et al., 2023; Seibold et al., 2021), we can use the relations of TSD and DC put forth by this study for estimating future deadwood volumes in Norway Spruce forests. This is important for anticipating potential changes in habitat value of saproxylic organisms as well as the carbon storage in deadwood. Table 4 summarizes a small analysis, where we used the deadwood volume stocks measurements 12 years apart, separated by DC and the average transition time of 5 years for a deadwood piece in shift to the next higher DC (see results section). We considered three scenarios, 

(1) no deadwood input, 

(2) deadwood input into DC 1 and 

(3) deadwood input into DC1 and DC2. 

For (1) the simulated deadwood stocks fit quite well the volume stocks in DC 4, but underestimate DC 1-3 (Table 3). For (2) and 5 m³/ha per year – a typical volume growth rate of mountain forests, assuming steady-state volume stocks and volume growth to be equal to tree mortality - the deadwood input after 12 years, would be in total 60 m³/ha. Scenario (2) much better capture DC1-4 stocks, but still underestimate observed stocks in DC2 and DC3. Scenario (3) with 2.5 m³/ha added to DC1 and 2.5 m³/ha added to DC2 each year reproduced the pattern of observed stocks increasing from DC1 to DC3 to DC2.

This estimation neglects deadwood volume loss associated with decay, such as bark shedding or faster decay of sapwood, varying climate conditions as well as the usual irregular deadwood input through tree mortality. Alternative approaches could include empirical determined probability for a deadwood piece to move from one DC to the next (Kruys et al., 2002; Russell et al., 2013) or mechanistic decay models considering site conditions and deadwood properties (Didion et al., 2014; Hararuk et al., 2020; Harmon et al., 2020). The poor performance for strongly decayed deadwood (DC 5) may be associated to 

(1) methodological reasons making it difficult for field crews to detect it separated from soil layers, due to partial cover by soil, litter, other deadwood or ground vegetation, 

(2) its rather ephemeral nature, as it may disintegrate under pressure of a fallen log, footsteps or rockfall and/or 

(3) poorly defined TSD of deadwood with DC 5 due to the current low number of available observations. 

Deadwood with advanced decay may need different sampling such as in-situ density determination of density (Ondrejka et al., 2020), freezing samples before extraction or additional DCs (Holeksa, 2001; Merganičová and Merganič, 2010) to more accurately describe its conditions. Further work can focus on incorporating the relations between TSD and DC in forest ecosystem and tree growth models (Thurnher et al., 2011), decision support tools (Vacik and Lexer, 2014), refine the assumptions using actual deadwood inputs measured using remote sensing (Senf and Seidl, 2021) or estimated tree mortality (Monserud and Sterba, 1999).

In summary, a permanent monitoring system in protected forests such as the Natural Forest Reserve system in Austria offers many insights into forest dynamics that temporary inventories cannot provide, including carbon stocks and habitat value (Frank and Müller, 2003). While time-demanding and labor-intensive to maintain, re-measuring permanent monitoring plots once in 10 years may be a good trade-off of incurred costs and gained information. Remeasurements periods greater than 10 years may come at the risk of losing plot and tree identification material, such as poles and tags, and miss important changes of the forest development stages in temperate forests.

Acknowledgements

We thank the landowners, that made their forest properties available for the Natural Forest Reserve (NFR) program and made this research possible. Without the continuing support and collaboration with Dipl.-Ing. Georg Frank from Austrian Research Centre for Forests (BFW) long-term maintenance of the NFR system would have been impossible. We acknowledge Dipl.-Ing. Dr. Mathias Leiter for his help with data analysis and data harmonization and Dipl.-Ing. Ambros Berger for statistical advice. We are grateful for the help with field work and data entry by Michael Götze, Alexander Schara, Matthias Kasseroler, Johannes Zorzi, Dimitrios Manousidis, Johannes Huber, Darius George Hardalau, Ajdin Starcevic and Herwig Ruprecht. Support was provided by the Erasmus Mundus Joint Master Degree program Master of Science in European Forestry (ID: 101239159). This research was funded in whole or in part by the Austrian Science Fund (FWF) 10.55776/PAT6986524.

References

Albrecht, L., 1991. Die Bedeutung des toten Holzes im Wald. Forstwissenschaftliches Cent. 110, 106–113. https://doi.org/10.1007/BF02741245

Albrich, K., Seidl, R., Rammer, W., Thom, D., 2022. From sink to source: changing climate and disturbance regimes could tip the 21st century carbon balance of an unmanaged mountain forest landscape. For. An Int. J. For. Res. 1–11. https://doi.org/10.1093/forestry/cpac022

Borghi, C., Francini, S., McRoberts, R.E., Parisi, F., Lombardi, F., Nocentini, S., Maltoni, A., Travaglini, D., Chirici, G., 2023. Country-wide assessment of biodiversity, naturalness and old-growth status using national forest inventory data, European Journal of Forest Research. Springer Berlin Heidelberg. https://doi.org/10.1007/s10342-023-01620-6

Caudullo, G., Welk, E., San-Miguel-Ayanz, J., 2017. Chorological maps for the main European woody species. Data Br. 12, 662–666. https://doi.org/10.1016/j.dib.2017.05.007

Didion, M., Frey, B., Rogiers, N., Thürig, E., 2014. Validating tree litter decomposition in the Yasso07 carbon model. Ecol. Modell. 291, 58–68. https://doi.org/10.1016/j.ecolmodel.2014.07.028

Edelmann, P., Weisser, W.W., Ambarlı, D., Bässler, C., Buscot, F., Hofrichter, M., Hoppe, B., Kellner, H., Minnich, C., Moll, J., Persoh, D., Seibold, S., Seilwinder, C., Schulze, E.D., Wöllauer, S., Borken, W., 2023. Regional variation in deadwood decay of 13 tree species: Effects of climate, soil and forest structure. For. Ecol. Manage. 541. https://doi.org/10.1016/j.foreco.2023.121094

FOREST EUROPE, 2020. State of Europe’s Forests 2015. Ministerial Conference on the Protection of Forests in Europe - FOREST EUROPE Liaison Unit Bratislava.

Frank, G., Müller, F., 2003. Voluntary approaches in protection of forests in Austria. Environ. Sci. Policy 6, 261–269. https://doi.org/10.1016/S1462-9011(03)00046-7

Fravolini, G., Tognetti, R., Lombardi, F., Egli, M., Ascher-Jenull, J., Arfaioli, P., Bardelli, T., Cherubini, P., Marchetti, M., 2018. Quantifying decay progression of deadwood in Mediterranean mountain forests. For. Ecol. Manage. 408, 228–237. https://doi.org/10.1016/j.foreco.2017.10.031

Götze, M., 2013. Waldbauliche Analyse des Naturwaldreservates Hutterwald. University of Natural Resources and Life Sciences, Vienna.

Hararuk, O., Kurz, W.A., Didion, M., 2020. Dynamics of dead wood decay in Swiss forests. For. Ecosyst. 7. https://doi.org/10.1186/s40663-020-00248-x

Hardalau, D.G., 2022. Comparative assessment of natural regeneration across natural forest reserves in Austria. University of Natural Resources and Life Sciences, Vienna.

Harmon, M.E., Cromack Jr., K., Smith, B.G., 1987. Coarse woody debris in mixed-conifer forests, Sequoia National Park, California. Can. J. For. Res. 17, 1265–1272. https://doi.org/10.1139/x87-196

Harmon, M.E., Fasth, B.G., Yatskov, M., Kastendick, D., Rock, J., Woodall, C.W., 2020. Release of coarse woody detritus-related carbon: A synthesis across forest biomes. Carbon Balance Manag. 15, 1–21. https://doi.org/10.1186/s13021-019-0136-6

Holeksa, J., 2001. Coarse woody debris in a Carpathian subalpine spruce forest. Forstwissenschaftliches Cent. 120, 256–270. https://doi.org/10.1007/bf02796097

Huber, J., 2022. Charakterisierung der Waldentwicklung im Naturwaldreservat Laaser Berg. University of Natural Resources and Life Sciences, Vienna.

Jandl, R., 2020. Climate-induced challenges of Norway spruce in Northern Austria. Trees, For. People 1, 100008. https://doi.org/10.1016/j.tfp.2020.100008

Jomura, M., Yoshida, R., Michalčíková, L., Tláskal, V., Baldrian, P., 2022. Factors controlling dead wood decomposition in an old-growth temperate forest in central Europe. J. Fungi 8, 673. https://doi.org/10.3390/jof8070673

Kasseroler, M., 2011. Vergleichende Analyse der Naturverjüngung in unbewirtschafteten und bewirtschafteten Gebirgswäldern am Beispiel des NRW Krimpenbachkessel, Salzatal. University of Natural Resources and Life Sciences, Vienna.

Korpel, S., 1995. Die Urwälder der Westkarpaten. Gustav Fischer Verlag, Stuttgart.

Kruys, N., Jonsson, B.G., Ståhl, G., 2002. A stage-based matrix model for decay-class dynamics of woody debris. Ecol. Appl. 12, 773–781. https://doi.org/10.1890/1051-0761(2002)012[0773:ASBMMF]2.0.CO;2

Kürsten, M., 2015. Waldbauliche Analyse des Naturwaldreservats Stoissen in den Leoganger Steinbergen. University of Natural Resources and Life Sciences, Vienna.

Kušar, G., Neumann, M., 2024. Patterns of deadwood volume and dynamics in Slovenian forests. Acta Silvae Ligni 133, 1–12. https://doi.org/10.20315/ASetL.133.1

Larrieu, L., Paillet, Y., Winter, S., Bütler, R., Kraus, D., Krumm, F., Lachat, T., Michel, A.K., Regnery, B., Vandekerkhove, K., 2018. Tree related microhabitats in temperate and Mediterranean European forests: A hierarchical typology for inventory standardization. Ecol. Indic. 84, 194–207. https://doi.org/10.1016/j.ecolind.2017.08.051

Lassauce, A., Paillet, Y., Jactel, H., Bouget, C., 2011. Deadwood as a surrogate for forest biodiversity: Meta-analysis of correlations between deadwood volume and species richness of saproxylic organisms. Ecol. Indic. 11, 1027–1039. https://doi.org/10.1016/j.ecolind.2011.02.004

Mayer, H., Neumann, M., 1981. Stand structure and dynamics in the virgin forests of Rothwald/Lower Austria and in Čorkova Uvala/Croatia. Forstwissenschaftliches Cent. 100. https://doi.org/10.1007/BF02640624

Merganičová, K., Merganič, J., 2010. Coarse woody debris carbon stocks in natural spruce forests of Babia hora. J. For. Sci. 56, 397–405. https://doi.org/10.17221/11/2010-jfs

Monserud, R.A., Sterba, H., 1999. Modeling individual tree mortality for Austrian forest species. For. Ecol. Manage. 113, 109–123. https://doi.org/10.1016/S0378-1127(98)00419-8

Müller-Using, S., Bartsch, N., 2007. Totholz im Elementhaushalt eines Buchenbestandes. Forstarchiv 78, 12–23.

Müller, J., Boch, S., Blaser, S., Fischer, M., Prati, D., 2015. Effects of forest management on bryophyte communities on deadwood. Nov. Hedwigia 100, 423–438. https://doi.org/10.1127/nova_hedwigia/2015/0242

Müller, J., Bütler, R., 2010. A review of habitat thresholds for dead wood: A baseline for management recommendations in European forests. Eur. J. For. Res. 129, 981–992. https://doi.org/10.1007/s10342-010-0400-5

Müller, J., Ulyshen, M., Seibold, S., Cadotte, M., Chao, A., Bässler, C., Vogel, S., Hagge, J., Weiß, I., Baldrian, P., Tláskal, V., Thorn, S., 2020. Primary determinants of communities in deadwood vary among taxa but are regionally consistent. Oikos 129, 1579–1588. https://doi.org/10.1111/oik.07335

Neumann, M., Echeverria, S., Hasenauer, H., 2023. A simple concept for estimating deadwood carbon in forests. Carbon Manag. 14, 1–12. https://doi.org/10.1080/17583004.2023.2197762

Neumann, M., Moreno, A., Mues, V., Härkönen, S., Mura, M., Bouriaud, O., Lang, M., Achten, W.M.J., Thivolle-Cazat, A., Bronisz, K., Merganič, J., Decuyper, M., Alberdi, I., Astrup, R., Mohren, F., Hasenauer, H., 2016. Comparison of carbon estimation methods for European forests. For. Ecol. Manage. 361, 397–420. https://doi.org/10.1016/j.foreco.2015.11.016

Nilsson, S.G., Niklasson, M., Hedin, J., Aronsson, G., Gutowski, J.M., Linder, P., Ljungberg, H., Mikusiński, G., Ranius, T., 2002. Densities of large living and dead trees in old-growth temperate and boreal forests. For. Ecol. Manage. 161, 189–204. https://doi.org/10.1016/S0378-1127(01)00480-7

Öder, V., Petritan, A.M., Schellenberg, J., Bergmeier, E., Walentowski, H., 2021. Patterns and drivers of deadwood quantity and variation in mid-latitude deciduous forests. For. Ecol. Manage. 487, 118977. https://doi.org/10.1016/j.foreco.2021.118977

Oettel, J., Lapin, K., Kindermann, G., Steiner, H., Schweinzer, K.-M., Frank, G., Essl, F., 2020. Patterns and drivers of deadwood volume and composition in different forest types of the Austrian natural forest reserves. For. Ecol. Manage. 463, 118016. https://doi.org/10.1016/j.foreco.2020.118016

Ondrejka, V., Gergeľ, T., Bucha, T., Pástor, M., 2020. Innovative methods of non-destructive evaluation of log quality. Cent. Eur. For. J. 66, 11. https://doi.org/10.2478/forj-2020-0021

Pan, Y., Birdsey, R.A., Phillips, O.L., Houghton, R.A., Fang, J., Kauppi, P.E., Keith, H., Kurz, W.A., Ito, A., Lewis, S.L., 2024. The enduring world forest carbon sink. Nature 631. https://doi.org/10.1038/s41586-024-07602-x

Parigger, P., 2021. Analyse der Walddynamik im Naturwaldreservat Niedergailwald, Kärnten. University of Natural Resources and Life Sciences, Vienna.

Petrillo, M., Cherubini, P., Fravolini, G., Marchetti, M., Ascher-Jenull, J., Schärer, M., Synal, H.A., Bertoldi, D., Camin, F., Larcher, R., Egli, M., 2016. Time since death and decay rate constants of Norway spruce and European larch deadwood in subalpine forests determined using dendrochronology and radiocarbon dating. Biogeosciences 13, 1537–1552. https://doi.org/10.5194/bg-13-1537-2016

Pollanschütz, J., 1974. Formzahlfunktionen der Hauptbaumarten Österreichs. Informationsd. Forstl. Bundesversuchsanstalt Wien 153, 341–343.

R Development Core Team, 2021. R: A language and environment for statistical computing. R Found. Stat. Comput. Vienna, Austria.

Rahman, M.M., Frank, G., Ruprecht, H., Vacik, H., 2008. Structure of coarse woody debris in Lange-Leitn Natural Forest Reserve, Austria. J. For. Sci. 54, 161–169. https://doi.org/10.17221/3102-jfs

Ranius, T., Jonsson, M., 2007. Theoretical expectations for thresholds in the relationship between number of wood-living species and amount of coarse woody debris: A study case in spruce forests. J. Nat. Conserv. 15, 120–130. https://doi.org/10.1016/j.jnc.2007.02.001

Rauh, J., Schmitt, M., 1991. Methodik und Ergebnisse der Totholzforschung in Naturwaldreservaten. Forstwissenschaftliches Cent. 110, 114–127. https://doi.org/10.1007/BF02741246

Rimle, A., Heiri, C., Bugmann, H., 2017. Deadwood in Norway spruce dominated mountain forest reserves is characterized by large dimensions and advanced decomposition stages. For. Ecol. Manage. 404, 174–183. https://doi.org/10.1016/j.foreco.2017.08.036

Ruprecht, H., Vacik, H., 2009. Aufnahmeschlüssel für die Erhebung in Naturwaldreservaten für das Projekt „ ELENA “.

Ruprecht, H., Vacik, H., Steiner, H., Frank, G., 2012. ELENA – a methodological approach for the long term monitoring of natural regeneration in natural forest reserves dominated by Norway Spruce (Vaccinio-Piceetea). Austrian J. For. Sci. 129, 67–105.

Russell, M.B., Woodall, C.W., Fraver, S., D’Amato, A.W., 2013. Estimates of downed woody debris decay class transitions for forests across the eastern United States. Ecol. Modell. 251, 22–31. https://doi.org/10.1016/j.ecolmodel.2012.12.012

Russell, M.B., Woodall, C.W., Fraver, S., D’Amato, A.W., Domke, G.M., Skog, K.E., 2014. Residence Times and Decay Rates of Downed Woody Debris Biomass/Carbon in Eastern US Forests. Ecosystems 17, 765–777. https://doi.org/10.1007/s10021-014-9757-5

Scherzinger, W., 1996. Naturschutz im Wald: Qualitätsziele einer dynamischen Waldentwicklung (Praktischer Naturschutz). Verlag Eugen Ulmer.

Seibold, S., Rammer, W., Hothorn, T., Seidl, R., Ulyshen, M.D., Lorz, J., Cadotte, M.W., Lindenmayer, D.B., Adhikari, Y.P., Aragón, R., Bae, S., Baldrian, P., Barimani Varandi, H., Barlow, J., Bässler, C., Beauchêne, J., Berenguer, E., Bergamin, R.S., Birkemoe, T., Boros, G., Brandl, R., Brustel, H., Burton, P.J., Cakpo-Tossou, Y.T., Castro, J., Cateau, E., Cobb, T.P., Farwig, N., Fernández, R.D., Firn, J., Gan, K.S., González, G., Gossner, M.M., Habel, J.C., Hébert, C., Heibl, C., Heikkala, O., Hemp, A., Hemp, C., Hjältén, J., Hotes, S., Kouki, J., Lachat, T., Liu, J., Liu, Y., Luo, Y.H., Macandog, D.M., Martina, P.E., Mukul, S.A., Nachin, B., Nisbet, K., O’Halloran, J., Oxbrough, A., Pandey, J.N., Pavlíček, T., Pawson, S.M., Rakotondranary, J.S., Ramanamanjato, J.B., Rossi, L., Schmidl, J., Schulze, M., Seaton, S., Stone, M.J., Stork, N.E., Suran, B., Sverdrup-Thygeson, A., Thorn, S., Thyagarajan, G., Wardlaw, T.J., Weisser, W.W., Yoon, S., Zhang, N., Müller, J., 2021. The contribution of insects to global forest deadwood decomposition. Nature 597, 77–81. https://doi.org/10.1038/s41586-021-03740-8

Senf, C., Seidl, R., 2021. Mapping the forest disturbance regimes of Europe. Nat. Sustain. 4, 63–70. https://doi.org/10.1038/s41893-020-00609-y

Shorohova, E., Kapitsa, E., Kazartsev, I., Romashkin, I., Polevoi, A., Kushnevskaya, H., 2016. Tree species traits are the predominant control on the decomposition rate of tree log bark in a mesic old-growth boreal forest. For. Ecol. Manage. 377, 36–45. https://doi.org/10.1016/j.foreco.2016.06.036

Starcevic, A., 2022. Forest dynamics of the natural forest reserve Schiffwald. University of Natural Resources and Life Sciences, Vienna.

Steininger, B., 2017. Beschreibung der Waldentwicklung im Naturwaldreservat Luxensteinwand. University of Natural Resources and Life Sciences, Vienna.

Štraus, H., Bončina, A., 2025. The vulnerability of four main tree species in European forests to seven natural disturbance agents: lessons from Slovenia. Eur. J. For. Res. 267–282. https://doi.org/10.1007/s10342-024-01754-1

Thorn, S., Bässler, C., Svoboda, M., Müller, J., 2017. Effects of natural disturbances and salvage logging on biodiversity – Lessons from the Bohemian Forest. For. Ecol. Manage. 388, 113–119. https://doi.org/10.1016/j.foreco.2016.06.006

Thurnher, C., Klopf, M., Hasenauer, H., 2011. Forests in transition: a harvesting model for uneven-aged mixed species forests in Austria. Forestry 84, 517–526. https://doi.org/10.1093/forestry/cpr021

Vacek, S., Vacek, Z., Bílek, L., Hejcmanová, P., Štícha, V., Remeš, J., 2015. The dynamics and structure of dead wood in natural spruce-beech forest stand – a 40 year case study in the Krkonoše national park. Dendrobiology 73, 21–32. https://doi.org/10.12657/denbio.073.003

Vacik, H., Lexer, M.J., 2014. Past, current and future drivers for the development of decision support systems in forest management. Scand. J. For. Res. 29, 2–19. https://doi.org/10.1080/02827581.2013.830768

Vacik, H., Rahman, M.M., Ruprecht, H., Frank, G., 2009. Dynamics and structural changes of an oak dominated Natural forest reserve in Austria. Bot. Helv. 119, 23–29. https://doi.org/10.1007/s00035-009-0053-5

Vandekerkhove, K., Meyer, P., Kirchmeir, H., Piovesan, G., Hirschmugl, M., Larrieu, L., Kozàk, D., Mikolas, M., Nagel, T., Schmitt, C., Blumröder, J., 2022. Old-growth criteria and indicators for beech forests (Fageta).

Vítková, L., Bače, R., Kjučukov, P., Svoboda, M., 2018. Deadwood management in Central European forests: Key considerations for practical implementation. For. Ecol. Manage. 429, 394–405. https://doi.org/10.1016/j.foreco.2018.07.034

Vogel, S., Bussler, H., Finnberg, S., Müller, J., Stengel, E., Thorn, S., 2021. Diversity and conservation of saproxylic beetles in 42 European tree species: an experimental approach using early successional stages of branches. Insect Conserv. Divers. 14, 132–143. https://doi.org/10.1111/icad.12442

Vrška, T., Přívětivý, T., Janík, D., Unar, P., Šamonil, P., Král, K., 2015. Deadwood residence time in alluvial hardwood temperate forests - A key aspect of biodiversity conservation. For. Ecol. Manage. 357, 33–41. https://doi.org/10.1016/j.foreco.2015.08.006

Wijas, B.J., Allison, S.D., Austin, A.T., Cornwell, W.K., Cornelissen, J.H.C., Eggleton, P., Fraver, S., Ooi, M.K.J., Powell, J.R., Woodall, C.W., Zanne, A.E., 2024. The Role of Deadwood in the Carbon Cycle: Implications for Models, Forest Management, and Future Climates. Annu. Rev. Ecol. Evol. Syst. 133–155. https://doi.org/10.1146/annurev-ecolsys-110421-102327

Winter, M.B., 2009. Protection efficiency and regeneration preconditions of Norway Spruce (Picea Abies (L.) Karst.) in the upper montane mountain forest oft he natural forest reserve Goldeck in Carinthia. University of Natural Resources and Life Sciences, Vienna.

Zukrigl, K., Eckhart, G., Nather, J., 1963. Standortskundliche und waldbauliche Untersuchungen im Urwaldresten der niederösterreichischen Kalkalpen. Mitteilungen der Forstl. Bundesversuchsanstalt Mariabrunn 62, 278.

Appendix