Carbon and nitrogen status by decay class in fallen dead wood of three pine species in southern Korea

Abstract The importance of a quantitative assessment of C and N contents of dead wood is increasing in forest ecosystems. This study aimed to determine the density and carbon (C) and nitrogen (N) status of dead wood with decay class for three pine species (Pinus densiflora, Pinus rigida, and Pinus koraiensis) in southern Korea. The C concentration in dead wood was significantly different among species (P. densiflora, 50.31%; P. koraiensis, 47.22%; P. rigida, 44.96%), whereas decay class did not affect the C concentration (p > 0.05). The density and C content of dead wood in all species decreased with increasing decay class. The N concentrations of dead wood increased more rapidly in P. rigida and P. koraiensis than in P. densiflora, with an increasing decay class. Thus, the N content of dead wood was unchanged or increased in P. rigida and P. koraiensis, whereas that of P. densiflora decreased because of density reduction with increasing decay class. Our results indicate that the unchanged, increased, or decreased status of C and N in dead wood depends on the species and decay class.


Introduction
Dead wood is a crucial pathway for C and N return in forest ecosystems because of its slow decomposition process, in which the remaining C and N can affect the microbial activity and mineralization of nutrients in forest ecosystems (Ge et al. 2013;Wambsganss et al. 2017;Lagomarsino et al. 2021;Martin et al. 2021). Thus, the quantitative evaluation of C and N content in dead wood is increasing on the stand, regional, national, and global scales (Russell et al. 2015;Noh et al. 2017;Martin et al. 2021).
The density and C and N concentrations of dead wood can be used to quantify C and N release in dead wood. However, those parameters depend on decay class, ecological (species and soil nutrient status), meteorological factors (annual temperature and precipitation), and forest management activities (thinning and fertilization) (Garrett et al. 2012;Harmon et al. 2013;Shorohova and Kapitsa 2014;Herrmann et al. 2015;Kim et al. 2021). Previous studies have reported the influence of decay classes and species on density and C and N concentrations (K€ oster et al. 2015;Cha et al. 2017;Yuan et al. 2017;Herrmann and Bauhus 2018). For example, the dead wood density generally decreases with increasing decay class, with different density reductions among species (K€ oster et al. 2015). These decomposition differences among species are closely related to substrate qualities (C and N concentrations and lignin concentration), and microbial activities (fungal) in forest ecosystems (Laiho and Prescott 2004;Cha et al. 2017;Herrmann and Bauhus, 2018).
Pine species (Pinus densiflora, Pinus rigida, and Pinus koraiensis) are widely distributed in Korean forests. The growing wood stocks (Total: 1,040,000,000 m 3 ) are 254,990,855 m 3 for P. densiflora, 41,214,576 m 3 for P. rigida, and 41,030,174 m 3 for P. koraiensis, respectively (Korea Forest Research Institute 2021). Since most forests in Korea have matured, dead wood is expected to increase through forest management activities, such as thinning and pruning (Korea Forest Service 2018). The stocks of dead wood in Korea were 14.83 m 3 ha À1 (90,401,011 m 3 ), which accounts for approximately 10% of the growing wood stock in Korean forests, showing increasing tendencies of dead wood stocks with an increasing stand age class (Moon et al. 2022).
The substrate qualities, such as wood basic density, lignin, C, and N concentrations, are different among pine species (Garrett et al. 2012;Korea Forest Research Institute 2014;Cha et al. 2017). These differences in substrate qualities among pine species might affect the C and N status of dead wood by decay class. Although several studies in Korea have examined the properties of dead wood or branches for only one pine species or comparisons with other species (not among pine species) based on a stand scale (Cha et al. 2017;Noh et al. 2017;Choi et al. 2021), uncertainties remain about the C and N status of species-specific dead wood in pine forests in Korea. In addition, little information is available to find out how amounts of C and N are released and which factors are responsible for the decomposition of dead wood of pine species. Thus, this study aimed to determine the wood density and C and N status of dead wood by decay class from three pine species (P. densiflora, P. rigida, and P. koraiensis).

Material and methods
2.1. Study areas, wood density, and C and N analysis The National Forest Inventory (NFI) in Korea has been conducted to provide basic data for forest management (Korea Forest Research Institute 2011). We collected dead wood samples from three pine species (P. densiflora, P. rigida, and P. koraiensis) for three years from a range of forest locations for the National Forest Inventory (NFI) in the 20 regions of southern Korea (Figure 1). The mean annual temperature and precipitation in the regions over the last 30 years  were 10.7-14.9 C and 987-1,675 mm, respectively (Korea Meteorological Administration 2021). The decay class was classified into four levels based on visual observations (Korea Forest Research Institute 2011). We sampled 10 pieces of dead wood for each decay class in each species from at least three other regional replicates [3 species Â 10 pieces Â 4 decay classes] (Tables 1 and  2). The samples were collected from pure forests because we could not identify species at advanced decay classes. Based on the sixth Korea NFI guidelines, dead wood was defined as detrital components with a diameter of at least 6 cm and a length of at least 1 m. Our measurements were restricted to fallen dead wood because of the scarcity of stand dead wood samples. Dead wood samples were collected as follows: From each selected log, sampling disks (5-10 cm) representing the decay class of the selected dead wood were (1) cut with a chainsaw, (2) sealed with a plastic zipper bag, (3) transported to the laboratory, and (4) oven-dried at 85 C for 48 h. Each sample was weighed (5), and some were (6) ground to analyze C and N concentrations. The dead wood volume was calculated based on Huber's formula (Eq. 1), which is widely used to estimate the volume of dead wood (Lagomarsino et al. 2021). Dead wood density was calculated by dividing the oven-dry weight by the sample volume. The dead wood density was used to convert the volume measured in the field into dry weight. The C and N concentrations in dead wood were analyzed using an elemental analyzer (Vario Macro cube, Langenselbold, Germany). The C and N contents of dead wood in each decay class were calculated using Eq. (2). Meanwhile, net sinks or sources mean that dead wood reserved or released C and N during decomposition, resulting in increased or decreased C and N contents of dead wood, respectively. (2)

Statistical analysis
All statistical analyses were conducted using R software v4.2.0. The data were analyzed by a general-liner-model Figure 1. Location of the sampling regions in South Korea; site (a) and dead wood at decay class IV (b) of P. koraiensis.
using the "glm" function in the R package based on main fixed effects [species type (S) and decay class (C)] and interactive effects (S Â C). In addition, principal component analysis (PCA) was conducted to determine the information from the multivariate data set by plotting all dead wood data with the first two principal components. Principal component analysis was performed using the "prcomp" function in the R package to summarize the variation in dead wood (physical and chemical properties). The correlation between the properties and the decay class of dead wood in this study could be explained by the wood density, which can represent the decay class as a quantitative variable with 95% confidence ellipses of samples by the decay class.

Wood density and moisture content
The wood density and moisture content of dead wood were significantly affected by species and decay class (p < 0.05), with no interaction effect (p > 0.05) ( Figure 2). For example, the wood density of pine species was higher in P. densiflora (0.47 g cm À3 ) than in P. rigida (0.41 g cm À3 ) and P. koraiensis (0.37 g cm À3 ) at decay class I (little decomposed) and significantly decreased (I: 0.42 g cm À3 , II: 0.34 g cm À3 , III: 0.27 g cm À3 , and IV: 0.21 g cm À3 ) with increasing decay class. However, dead wood densities became more similar among species as the decay class increased. The moisture content of dead wood increased with increasing decay class.

C and N status of dead wood
The species and decay class-specific C and N concentrations are shown in Table 2. The C concentrations in dead wood were affected by species (P. densiflora: 50.31%, P. koraiensis: 47.22%, and P. rigida: 44.96%), but the decay class did not affect the C concentration in dead wood (p > 0.05) ( Figure 3). The C content of   dead wood was similar to the decreasing pattern of the density. The C content was higher in P. densiflora (174 kg C m À3 ) across the decay class than in other pine species (P. koraiensis: 136 kg C m À3 and P. rigida: 130 kg C m À3 ), with a significantly decreasing pattern with increasing decay class across species (I: 202 kg C m À3 , II: 162 kg C m À3 , III: 124 kg C m À3 , and IV: 101 kg C m À3 ) (Figures 4 and 5). P. densiflora (54.7%) showed a higher C reduction than other pine species (P. koraiensis: 45.4% and P. rigida: 46.2%) from decay classes I to IV (Figure 4).
The N concentration and content were significantly affected by species and class with interaction effects (Figures 3 and 4). The N concentration was significantly higher in P. densiflora (0.53%) than in P. koraiensis (0.28%) and P. rigida (0.19%) in decay class I. However, rapidly increased N concentrations were observed in P. koraiensis (0.28-0.63%) and P. rigida (0.19-0.38%) compared with P. densiflora (0.53-0.61%) from decay classes I to IV (Figure 3). N contents of dead wood decreased in P. densiflora from decay class I (2.43 kg N m À3 ) to IV (1.36 kg N m À3 ), whereas N contents increased in P. koraiensis (0.99-1.30 kg N m À3 ) or unchanged in P. rigida (0.79-0.81 kg N m À3 ) from decay class I to IV (Figure 4).

Relationship between C or N contents and properties of dead wood
The first and second principal components (PCA) accounted for 84.7% of the total variation in the properties of dead wood. PCA 1 accounted for 49.9% of the variation, where the first axis was primarily correlated with the density and C content of dead wood. PCA 2 contributed to 34.8%, in which the second axes were  mainly correlated with N content and N concentration of dead wood. There were positive correlations between density and C content and between N concentration and N content, whereas density and C content were negatively correlated with moisture content. Meanwhile, the N content of dead wood was not related to density ( Figure 5).

Wood density
Although the densities of dead wood in decay class I were similar to the basic density of living trees of pine species (P. densiflora: 0.47 g cm À3 , P. rigida: 0.50 g cm À3 , and P. koraiensis: 0.41 g cm À3 )(Korea Forest Research Institute 2014), the results have highlighted the distinct density reduction in class IV in three pine species. These changes in density are crucial because it is a good indicator of the decay process of dead wood (Harmon et al. 2013;K€ oster et al. 2015).
In this study, density reduction was higher in P. densiflora than in other species. The significantly higher N concentration of P. densiflora in the initial decay class may explain its higher density loss than that of the other pine dead wood. This is because a substrate with a high N concentration leads to rapid mass loss (Yang et al. 2010;Ge et al. 2013;Cha et al. 2017;Choi et al. 2021). In addition, abiotic factors such as air temperature, precipitation, elevation, and soil nutrient status might affect the dead wood density reduction. In this study, the mean air temperature and elevation of sampled regions were 13.41 C and 355 m for P. densiflora, 12.94 C and 437 m for P. koraiensis, and 12.88 C and 251 m for P. rigida, respectively. Considering that the air temperature decrease as the elevation increase, the mean temperature of P. koraiensis forests might be lower than 12.94 C. Thus, sampled distribution characteristics of P. densiflora growing at a relatively high air temperature and the low altitude might affect the activities of microorganisms involved in decomposition, resulting in a higher decrease in the wood density in P. densiflora. A similar result has been reported that the decomposition rate of leaf and fine root litters decreased regardless of the type of litter as the altitude increased, which was due to a decrease in soil microbial communities as temperature decreased (Zhou et al. 2015).

C and N status of dead wood
The values of dead wood in this study fall within the range of other C and N concentrations of branches (C: 45.4-48.2%, N: 0.2-0.35%) and dead wood (C: 48.5-51.3%, N: 0.15-0.82%) of Korean pine species (Garrett et al. 2012;Cha et al. 2017;Noh et al. 2017;Yuan et al. 2017;Choi et al. 2021;Martin et al. 2021).
The higher C concentration in P. densiflora than in other pine dead wood could be attributed to the   (density, moisture, N concentration, and C, N contents). Different colors of the ellipse circle represent 95% confidence of dead wood samples by decay class. difference in lignin concentration among the tree species because the C concentration of dead wood is positively correlated with lignin concentration (Thomas and Martin 2012). Chong and Park (2008) reported that the dead wood of P. densiflora (29.32%) presented a higher lignin concentration than that of P. rigida (28.64%) and P. koraiensis (27.85%). However, the decay class in this study did not affect the C concentration in dead wood. The unchanged C concentrations by decay class are in agreement with previous results (Yang et al. 2010;Romashkin et al. 2021), which could be due to confounding factors, such as species-specific properties, stand characteristics, forest management, and site conditions (Thomas and Martin 2012;Herrmann et al. 2015;K€ oster et al. 2015).
The C content of dead wood was higher in P. densiflora (242 kg C m À3 ) than in other pine species (P. koraiensis: 180 kg C m À3 and P. rigida: 183 kg C m À3 ) in decay class I, but similar between species (P. densiflora: 109 kg C m À3 ; P. koraiensis and P. rigida: 98 kg C m À3 ) in decay class IV. The C contents in this study are comparable to that of Pinus sylvestris (198 kg C m À3 ) and Picea abies (185 kg C m À3 ) in decay class I; Pinus sylvestris (107 kg C m À3 ) and Picea abies (99 kg C m À3 ) in decay class IV in hemiboreal forests (Stak_ enas et al. 2020).
The C content of dead wood significantly decreased in all pine species, with a higher loss rate in P. densiflora than in the other pine species. Considering that the C concentration was not significantly different among species and classes, the higher losses of C contents are explained by a higher density reduction in P. densiflora than in other pine dead wood. The loss rates of C contents in this study from decay class I to IV are comparable with the values of 52% of P. sylvestris dead wood after 36 years of decay (Herrmann and Bauhus, 2018).
In this study, the significantly different N concentrations of dead wood among species could be due to differences in the inherent characteristics of N uptake (Berg and Laskowski 2006). These differences in N concentrations in dead wood among species have been reported in previous studies (Laiho and Prescott 2004;Garrett et al. 2012;Herrmann and Bauhus 2018). The N concentrations of dead wood rapidly increased in P. koraiensis (0.28-0.63%) and P. rigida (0.19-0.38%) from decay class I to IV, but slightly increased in P. densiflora (0.53-0.61%). These increased N concentrations of dead wood with increasing decay class are consistent with other results (K€ oster et al. 2015). In addition, Noh et al. (2017) have reported that the N concentration of dead wood increased from decay class I (0.15%) to IV (0.42%) in 75-year-old P. densiflora stand.
The increased N concentration of dead wood in our study could be due to density reduction and N accumulation by microorganisms, immobilization, and inputs by through fall (Garrett et al. 2012;K€ oster et al. 2015;Noh et al. 2017). However, rapidly increased N concentrations were observed in P. koraiensis and P. rigida, which had lower initial N concentrations compared with a high initial N concentration of P. densiflora. This is consistent with the results that substrates with a low initial N could have more N gain with increasing decay class (Laiho and Prescott 2004;Yang et al. 2010). Meanwhile, the poorer the initial substrate, the more translocation occurs when decaying microbial organisms such as fungi translocate nutrients from the soils to the logs because of colonizing to compete with surrounding trees when soil nutrients are limited (Wells and Boddy 1990;Laiho and Prescott 2004).
The remaining N content of dead wood was not related to density loss because N concentrations varied considerably with decay class and species. Although our N concentrations in dead wood did not change from classes I to III, dead wood density reduction implied that the net weight of N decreased between classes I and III. However, the remaining N content increased or remained unchanged in decay class IV compared with class I because of the rapidly increased N concentration of dead wood in P. koraiensis and P. rigida. Thus, the dead wood of those species became net N sinks in decay class IV, whereas P. densiflora was a source of N. A similar result has been reported that the net sink or source of dead wood is largely determined by species, which is closely related to substrate qualities, such as C, N, P, and lignin concentration because of microbial activities (Laiho and Prescott 2004;Herrmann and Bauhus 2018).
We acknowledge a few limitations of our studies on quantifying the C and N contents of dead wood. First, we sampled cross-sectional disks (5-10 cm) representing the decay class from the fallen dead wood in forests without consideration of the effect of the location of wood (top, middle, low, heartwood, and sapwood) and bark. However, the properties (wood density, C and N concentrations) of fallen long-dead wood might be varied with the location of the wood as well as the bark cover (Harmon et al. 2013). These heterogeneous properties could affect the decomposition process differently. Second, the C and N contents of dead wood in our study might be slightly overestimated than actual forests because we did not consider the existence of internal cavities in decay logs by small soil animals which can decrease total mass. Therefore, our results on the wood density and C and N concentrations or contents by decay classes in three pine species should be interpreted cautiously. And, further studies are needed for understanding the role of the factors mentioned above for the expansion and extrapolation into forest ecosystems. Nevertheless, it is meaningful because our study is the first attempt to track the C and N status of species-specific dead wood by decay class in three pine species in southern Korea.

Conclusions
This study was conducted to determine the wood density and C and N status of dead wood by decay class from three pine species (P. densiflora, P. rigida, and P. koraiensis) investigated across southern Korea.
Our results indicate that the species and decay class affect the variations in wood density and C and N concentration of dead wood, resulting in differences in C and N contents of dead wood among three pine species. The C content of dead wood was related to the decreasing density pattern of each species, showing higher C loss rates in P. densiflora than in the other pine dead wood. In contrast, the N content of dead wood was not related to the density pattern due to the variation of N concentration depending on species. The N concentrations of dead wood rapidly increased in P. koraiensis (0.28-0.63%) and P. rigida (0.19-0.38%) compared to P. densiflora (0.53-0.61%) from decay class I to IV. Thus, the N content of P. koraiensis and P. rigida dead wood were net sinks of N in decay class IV due to the rapidly increased N concentration, whereas that of P. densiflora was a net source of N. These results indicated that the species and decay class-specific estimation should be considered when the N contents of deadwood are calculated particularly.

Disclosure statement
No potential conflict of interest was reported by the author(s).