Impact of spruce deadwood logs on physical properties of a loamy sand Podzol in a steep temperate forest

Thinned steep forests are particularly vulnerable to soil physical degradation. Retaining deadwood logs from thinning operations on the forest floor can potentially mitigate soil physical degradation by modifying its physical properties through increased carbon content in steep regions. We aimed to investigate the effect of spruce deadwood logs from thinning operations on the physical properties of a loamy sand Podzol soil in a steep (30°) temperate spruce forest in Bavaria, Germany. The soil organic carbon (SOC) content was 56% higher under deadwood logs compared to the control areas (p‐value = 0.097). Deadwood logs also increased the soil water repellency by 13% (p‐value = 0.269), while decreasing the soil shear strength by 35% (p‐value = 0.001). Shear strength and water repellency strongly correlated with SOC content, with r = −0.87 and r = 0.86, respectively. Although retaining deadwood logs seems a promising carbon sequestration strategy, it can adversely affect soil shear strength and water repellency and potentially lead to soil degradation. Therefore, the choice to keep deadwood logs on the forest floor may align with specific management goals.


| INTRODUCTION
Forests reduce the risk and severity of floods by intercepting water and slowly releasing it into streams during heavy rainfall events (Bradshaw et al., 2007).Forests enhance soil permeability and emit water into the atmosphere through evapotranspiration, which helps minimize rainwater runoff.These processes also protect the soil against water erosion, that is, the detachment or displacement of soil particles by raindrop impact or surface water flow (Ganasri & Ramesh, 2016;Vaezi et al., 2017).Soil water erosion is a major threat to the sustainability of terrestrial ecosystems and one of the most important environmental problems worldwide (Vaezi et al., 2017).For example, it has negative consequences for land resource degradation (e.g., the loss of fertile topsoil), depletion of soil organic matter, and water pollution (Lal, 2004;Pimentel, 2006;Rothwell et al., 2005).The risk of land degradation by soil water erosion is even higher in steep forests (Bradshaw et al., 2007).Approximately 20% of the Earth's forests are situated on steep slopes with inclinations exceeding 15 (Lundbäck et al., 2021).Thus, it is of paramount importance to preserve soils of steep forests for the sustainable delivery of ecosystem services.
Apart from the forest canopy, soil physical properties also play a crucial role in influencing the rate of soil degradation by natural phenomena such as water erosion.The stability of soil is an important factor concerning the risk of water erosion (Blanco-Canqui, 2017;Vaezi et al., 2017).Soil stability can be quantified through shear strength measurements.Soil infiltration rate and saturated hydraulic conductivity are also essential factors in regulating the soil's ability to retain water and prevent runoff.Soils with high infiltration rates and hydraulic conductivities exhibit reduced rates of soil degradation by water erosion (Ran et al., 2012;Vaezi et al., 2017).One significant factor contributing to the enhancement of these hydraulic properties is the presence of soil bio-pores.Bio-pores are large interconnected pores within the soil matrix created by roots and fauna, allowing water to flow more easily.The degree of soil water repellency is another significant determinant influencing soil degradation and erosion by water.Water-repellent soils pose a greater risk of water erosion and runoff compared to hydrophilic soils, primarily due to their reduced infiltration rate (Leighton-Boyce, 2002;Shakesby et al., 2000).
Forest thinning is the strategic removal of a certain proportion of trees to effectively boost resource (e.g., sunlight and soil water) accessibility for the remaining trees (Sohn et al., 2016).In spite of these advantages, thinning reduces rainfall interception and increases the risk of soil degradation by runoff and water erosion (Wang et al., 2018(Wang et al., , 2021)).The risk escalates further under projected climate change patterns, characterized by increased frequency and intensity of rainfall events (Easterling et al., 2000;Ontl et al., 2018;Wang et al., 2021).Reducing the intensity of thinning is recommended as an effective management strategy to mitigate the risk of soil degradation by water erosion during heavy rainfall events.This is attributed to the enhanced rainfall interception and increased soil capacity to retain water (Wang et al., 2021;Zhao et al., 2016).
An innovative and impactful management approach to mitigate soil degradation could involve retaining deadwood from thinning operations on the forest floor.Carbon derived from deadwood finds its way into the soil as dissolved organic matter carried by infiltrating precipitation water, or in the form of fragments by bioturbation (Allison & Martiny, 2008;Ma et al., 2014).Forest management strategies often do not incorporate the retention of deadwood on the forest floor, mainly due to inconclusive findings from past research endeavors (Magnússon et al., 2016;Stutz & Lang, 2017) and a restricted comprehension of its implications, especially in European temperate forest ecosystems (Lombardi et al., 2013).
A recent study investigated the influence of thinning-derived deadwood logs on soil chemical and microbial properties in a managed spruce forest on loamy sand Podzol in Germany (Nazari, Pausch, et al., 2023).After about 15 years, the presence of deadwood logs led to considerable increases in the soil organic carbon (SOC) content at different depths (59% and 56% at 0-4 cm and 8-12 cm depths, respectively).The study suggested that leaving thinning-derived deadwood logs on the forest floor can enhance soil and forest sustainability while promoting carbon sequestration.
Retaining thinning-derived deadwood logs on the forest floor may serve as an innovative management strategy to actively mitigate the risk of soil degradation by modifying soil physical properties, specifically by enhancing water-holding capacity via increased SOC content.There are currently no scientific reports showing the influence of thinning-derived deadwood logs on soil physical characteristics in steep forests.
As a follow-up experiment to Nazari, Pausch, et al. (2023) conducted at the same forest site, this research aimed to assess the impact of deadwood resulting from forest thinning on the soil physical characteristics.We hypothesized that (i) deadwood logs from thinning operations will increase the soil water content, shear strength, water repellency, field-saturated hydraulic conductivity (K fs ), and infiltration rate and (ii) there will be a strong relation between the deadwoodassociated increase in SOC and water contents and the measured soil physical variables.

| Study site and experimental design
The research site is situated in Kulmbach, Bavaria, Germany, at coordinates 50 6 0 30 00 N and 11 29 0 12 00 E. It is a managed spruce (Picea abies) forest that has been in existence for 55 years (Figure 1a).The forest has a small scale with an area of about 2 ha.In 1968, 5000 3-year-old spruce seedlings per hectare were planted for timber production.Ten years later in 1978, the first thinning operation was implemented by cutting 500 trees per hectare.The forest is positioned on a north-facing slope with a 30 incline, and its elevation is 360 m above sea level.The area experiences an average annual temperature of 9.1 C and a total yearly precipitation of 957 mm.Mosses and ferns were the prevailing vegetation on the forest floor.The soil is classified as Podzol according to the Federal Institute for Geosciences and Natural Resources of Germany (www.bgr.bund.de),and it features a loamy sand texture based on the USDA taxonomy.The soil is acidic with low biological activity.Some characteristics of the soil (Nazari, Pausch, et al., 2023) are presented in Table 1.In 2006, the forest underwent the second thinning by cutting 20 trees per hectare.The felled spruce trees were cut into smaller logs and randomly scattered across the forest floor (ca.200 logs per ha).We selected three deadwood logs (n = 3) that shared similar characteristics in terms of length (ca.80 cm), diameter (around 60 cm), and degree of decay.The level of decay for the deadwood logs was determined using the "pocket-knife" technique (Lachat et al., 2014).The chosen deadwood logs displayed a moderate degree of decay, meaning that a pocket knife could easily enter through the fibers, but not all the way across (Figure 1b).These logs were in full contact with the forest floor.We opted for logs parallel to the slope to prevent water and debris from gathering excessively on one side.Both field measurements and bulk soil sampling were done in July 2023.

| Field and laboratory measurements
Following the removal of the deadwood logs and the upper organic layer, shear strength and K fs were measured beneath the central area of each deadwood log (referred to as "deadwood"), as well as from a location 3 m away from the log (referred to as "control").The choice of the distance between the deadwood location and the control was aimed at eliminating any potential impact from the deadwood and its byproducts.Shear strength was measured at a 1-3 cm depth of the upper mineral soil horizon using a 19-mm hand shear vane (Pilcon, English Drilling Equipment Co. Ltd.England).Five shear tests were performed at each location (pseudoreplicates).We employed the SATURO dual head infiltrometer (Meter Group Inc., Pullman, WA, USA) to measure K fs .A 5-cm insertion ring with a radius of 7.6 cm was utilized at 0-5 cm depth.The measurement configurations were as follows: pressure head one set at 5 cm H 2 O, pressure head two set at 10 cm H 2 O, a soak time of 15 min, a hold time of 15 min, and two pressure cycles.Figure 1c illustrates the dual-head infiltrometer installed at the studied site.Infiltration rate at steady state is recorded at each pressure head.K fs is calculated using the modified and simplified version of the two-ponding head technique developed by Reynolds and Elrick (1990) as follows: where K fs is field saturated hydraulic conductivity (cm s À1 ); D 1 is the Δ is 0.993 Â the infiltrometer ring insertion depth (cm) + 0.578 Â the infiltrometer ring radius (cm); i 1 is the infiltration rate at D 1 (cm s À1 ); and i 2 is the infiltration rate at D 2 (cm s À1 ).
Bulk soil samples were taken at a 0-5 cm depth to measure soil water repellency, organic carbon content, and gravimetric water content.The bulk samples were wet sieved (<2 mm) and stored in polyethylene bags at 4 C before undergoing analysis.Water repellency was determined using the molarity of an ethanol droplet test (Roy & McGill, 2002).Soil samples were oven-dried at 60 C, followed by equilibration at 20 C for a minimum of 48 h.Then, the soil samples were promptly placed in plastic containers, and the surface of the samples was leveled by applying a normal stress of 60.9 Nm À2 (Weber et al., 2021) for 2 min.A sequence of ethanol and deionized water solutions was prepared, ranging from 0.00 to 0.80 m 3 m À3 in increments of 0.01 m 3 m À3 for ethanol concentration.Subsequently, a 60 μL droplet of the ethanol solution was placed with care on the soil sample surface, allowing it to infiltrate for 5 s.The maximum ethanol concentration that did not infiltrate into the soil within this timeframe was then utilized to calculate the soil sample liquid surface tension or water repellency, using the equation outlined by Roy and McGill (2002): where Y is liquid surface tension or water repellency (mN m À1 ); ln is natural logarithm; and X is the molarity of ethanol solution (mol L À1 ).T A B L E 1 Properties of the loamy sand soil (0-4 cm depth) of the studied thinned spruce forest located in Kulmbach, Bavaria, Germany (Nazari, Pausch, et al., 2023).It is important to note that soil water repellency is measured in surface tension units, which has an inverse relationship with the degree of soil water repellency.
Soil total carbon content was measured by dry combustion employing an Elemental Analyzer (Euro EA-CN, Eurovector, Pavia, Italy).Given the lack of carbonates and the soil's low pH, the determined total carbon was considered SOC.Soil water content was determined gravimetrically by oven-drying at 105 C for 48 h.

| Statistical analysis
To assess the normality of data distribution, the Shapiro-Wilk test

| RESULTS
Although not statistically significant, deadwood logs increased the SOC content by 56% (Figure 2a).Deadwood logs did not significantly affect the soil gravimetric water content (Figure 2b).Deadwood logs significantly decreased the soil shear strength by 35%, from 37 to 24 kPa (Figure 3a).Deadwood logs led to a 13% increase in soil water repellency, although the change was not statistically significant (Figure 3b).There were negligible differences between the soil under deadwood logs and the control soil in terms of K fs (control = 14 cm h À1 , deadwood = 17 cm h À1 ) and infiltration rate (control = 31 cm h À1 , deadwood = 34 cm h À1 ) (Figure 4a,b).Shear strength and water repellency of the soil showed strong correlations with SOC content with r = À0.87 and r = 0.86, respectively (Table 2).None of the measured variables were significantly correlated with soil water content (Table 2).

| DISCUSSION
Deadwood logs increased the SOC content by 56% (Figure 2a), although the change was not statistically significant.In another study conducted at the same site, deadwood logs enhanced the SOC content by 59% at 0-4 cm depth (Nazari, Pausch, et al., 2023), further confirming our result.The increased SOC content under deadwood logs could be due to the transportation of dissolved and particulate organic matter from slowly decaying spruce deadwood into the soil by rainwater and bioturbation (Stutz et al., 2017;Wambsganss et al., 2017).Another reason could be the incorporation of deadwood-decaying fungal mycelia into the soil that directly contributed to the SOC content or translocated deadwood carbon compounds into the soil (Boberg et al., 2010;Cairney, 2005;Wambsganss et al., 2017).Deadwood logs could also accelerate the microbial decomposition of the litter and the incorporation of litter-derived carbon into the soil (Peršoh & Borken, 2017).Each of these processes could have contributed an unknown share to the observed result.
Overall, the low biological activity of the investigated Podzol soil (Nazari, Pausch, et al., 2023) and the slow decay rate of spruce deadwood (Petrillo et al., 2016) could promote the accumulation of deadwood-associated carbon over time rather than its mineralization to the atmosphere as carbon dioxide.The assessed deadwood exhibited a moderate degree of decay.We anticipate that, as the decay class advances over time, the deadwood will further contribute to the SOC content.We, therefore, recommend keeping spruce deadwood logs on the floor of forests with Podzol soil as a management strategy to support carbon sequestration.
In contrast to our hypothesis and previous studies (e.g., Jensen et al., 2019), the soil shear strength was lower beneath deadwood logs where organic carbon tended to be higher (Figure 3a, Table 2).However, both are within the range of expected values for sandy soils (Schjønning, 1986).A recent meta-analysis showed that an increase in SOC content decreases the bulk density of forest soils (i.e., loosening the soil) and increases their susceptibility to compaction (Nazari, Arthur, et al., 2023).The investigated acidic Podzol has about 80% sand.Even though the deadwood-derived carbon entered the soil, it seems that the low soil biological activity did not allow for organomineral bindings through microbial extracellular polymeric substances.
We also do not expect a well-developed soil structure for the loamy sand, that is, a strong aggregation of mineral and organic particles.
This means that the SOC did not strengthen the soil structure but loosened it, and might have reduced the angle of internal friction of the investigated soil layer.The soil water contents under deadwood logs and in the control areas were similar (Figure 2b), whereas we expected that the soil under deadwood logs would have a higher water content than the control soil, due to less evaporation and higher SOC content.We suggest that future studies investigate which factors and processes influence soil water dynamics under deadwood logs, and if a micro-climate could be identified.Soil water content had no significant correlation with the measured properties (Table 2).This is surprising because soil water content usually governs the physical  and mechanical properties of the soil as well as processes such as the alteration of frictional forces and shear strength.
Deadwood logs increased the soil water repellency (Figure 3b), confirming our hypothesis.Deadwood logs contain hydrophobic organic compounds, such as lignin and waxes (Minnich et al., 2021;Stutz et al., 2019), which could be released into the soil during decomposition.These hydrophobic substances could have coated soil mineral particles and reduced the surface tension, effectively increasing the water repellency of the soil.Organic carbon is also a substantial driver of the severity of soil water repellency; the higher the SOC content, the greater its water repellency (Hermansen et al., 2019;Weber et al., 2021).Thus, the increased SOC content by deadwood logs could have enhanced the water repellency of the soil.The strong correlation between SOC content and water repellency (Table 2) further supports this.Water-repellent soils present an elevated susceptibility to degradation (i.e., water erosion and runoff) as compared to hydrophilic soils, in case of high precipitation after drought events (Leighton-Boyce, 2002;Shakesby et al., 2000).
Deadwood logs only slightly increased the soil K fs and infiltration rate (Figure 4a,b).We assumed that the soil underneath deadwood logs would have high faunal activity, creating pores and channels, and increasing the K fs and infiltration rate.The investigated acidic Podzol soil generally has very low biological activity, even under deadwood logs (Nazari, Pausch, et al., 2023).Accordingly, the contribution of soil fauna (e.g., ants and collembola) to incorporating deadwoodassociated organic carbon and creating macro-pores could be low, although existing.The high consistency of the K fs and infiltration rate measurements, as indicated by the small standard errors, as well as the high values (Figure 4a,b), suggests that these two water transport functions are controlled by the textural macro-porosity (macro-pores between sand particles) rather than by bio-pores.
The choice of sample size (n = 3) in our study was carefully considered in the context of the characteristic features of the investigated forest.Given the small size of the forest (ca. 2 ha), its relatively homogeneous nature, and its steep slope, we aimed to strike a balance between obtaining a representative sample of the ecosystem and the practical constraints associated with fieldwork.In cases where the study area is limited in size and exhibits homogeneity in key environmental variables, a smaller sample size may suffice to capture the variability present.In our case, the homogeneous nature of the forest allowed us to effectively sample the variations in soil properties, but we acknowledge that the small sample size may influence the statistical power of our analysis.A larger sample size could potentially reveal more nuanced differences, particularly in parameters such as water repellency.Future research endeavors in larger or more heterogeneous forest ecosystems may benefit from increased sampling efforts to enhance the robustness and generalizability of the findings.

| Study limitations
It is crucial to acknowledge certain limitations inherent in our study, primarily stemming from its nature as a case study conducted in a specific geographic context.The findings presented herein are derived from a small-scale spruce forest in Bavaria, Germany, representing a unique set of environmental conditions.Consequently, the extent to which these results can be generalized to broader ecological contexts or other forest ecosystems is subject to limitations.Therefore, while the insights gained from this study contribute valuable information to the existing body of knowledge, they should be interpreted within the context of this specific case study or similar forest ecosystems.

| CONCLUSION
We observed a considerable increase in the organic carbon content of the soil due to the presence of spruce deadwood logs.Consequently, we suggest that retaining spruce deadwood logs be considered a viable carbon sequestration strategy in forests with Podzol soils.It is however important to note that these deadwood logs may have had adverse effects on the soil, reducing its shear strength and increasing water repellency, with only minor impacts on K fs and the infiltration rate.The decision regarding the presence of deadwood logs on the forest floor may align with our specific management objectives.To further expand our understanding of this topic, we recommend additional research involving deadwood logs from various tree species and different soil types in steep forests around the world.

F
I G U R E 1 The studied thinned spruce forest located in Kulmbach, Bavaria, Germany (a), the spruce deadwood log of moderate decay class (b), and the dual-head infiltrometer installed at a depth of 0-5 cm (c).[Colour figure can be viewed at wileyonlinelibrary.com] was employed.The significance of the impact of deadwood was evaluated through the Paired-Samples T-Test, with a significance level set at α = 0.05 (n = 3).Pearson's correlation test was performed to detect significant associations between the measured soil properties at α = 0.05 (n = 6).The data were analyzed using IBM SPSS Statistics for Windows, version 25 (IBM Corp., Armonk, NY, United States).Bar plots were generated using SigmaPlot 14.0 (Systat, San José, CA, United States).All values presented are arithmetic means.

F
I G U R E 2 Impact of thinning-derived deadwood logs on the soil organic carbon content (SOC) (a) and gravimetric water content (b) at 0-5 cm depth in a temperate spruce forest in Kulmbach, Bavaria, Germany (Paired-Samples T-Test at α = 0.05).Different letters on the bars show a statistically significant difference between the treatments.Error bars indicate the standard error of the mean.n is the number of replicates.[Colour figure can be viewed at wileyonlinelibrary.com]F I G U R E 3 Impact of thinning-derived deadwood logs on the soil shear strength at 1-3 cm depth (a) and water repellency at 0-5 cm depth (b) in a temperate spruce forest in Kulmbach, Bavaria, Germany (Paired-Samples T-Test at α = 0.05).Note that the soil water repellency is measured in surface tension units, which has an inverse relationship with the degree of soil water repellency.Different letters on the bars show a statistically significant difference between the treatments.Error bars indicate the standard error of the mean.n is the number of replicates.[Colour figure can be viewed at wileyonlinelibrary.com]

F
I G U R E 4 Impact of thinning-derived deadwood logs on the soil field saturated hydraulic conductivity K fs (a) and infiltration rate (b) at 0-5 cm depth in a temperate spruce forest in Kulmbach, Bavaria, Germany (Paired-Samples T-Test at α = 0.05).Different letters on the bars show a statistically significant difference between the treatments.Error bars indicate the standard error of the mean.n is the number of replicates.[Colour figure can be viewed at wileyonlinelibrary.com]T A B L E 2 Correlations of the measured soil variables affected by thinning-derived deadwood in the temperate spruce forest (Pearson's correlation test at α = 0.05, n = 6).