Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2025 May 26;27(6):1097–1106. doi: 10.1111/plb.70051

Factors affecting seedling establishment in a warm‐temperate secondary forest affected by oak wilt disease

Celegeer 1, N Watanabe 1, K Otani 1, T Okada 1, M Matsushita 2, M Nakagawa 1,✉
PMCID: PMC12477313  PMID: 40417827

Abstract

  • Diseases are major natural disturbances to forest ecosystems that cause changes to overstory tree structure and influence seedling dynamics by altering environmental conditions via functional traits. Thus, identifying factors affecting seedling dynamics would improve broader understanding of the seedling regeneration process after disease disturbance.

  • We investigated 13,010 current year seedlings from 59 woody species, 44 genera, and 21 families in a Japanese warm‐temperate secondary forest, following Japanese oak wilt disease (JOW) in 2009–2013 and in 2018–2020. We also quantified temporal changes in canopy openness and soil moisture. Structural equation model (SEM) was used to explore the effects of environmental conditions and seedling functional traits on current year seedling density in autumn for 13 dominant species.

  • Canopy openness gradually increased, whereas soil water content decreased throughout the study site. After disruption by JOW, recruitment of current year seedlings increased through improvements in light conditions. The functional traits of current year seedlings were likely affected by soil nutrients in the late JOW period (2011–2013), whereas litterfall production influenced functional traits of current year seedlings in the post‐JOW period (2018–2020). Factors affecting seedling density varied as environmental conditions changed across the JOW periods. Seedling establishment was enhanced by increased canopy openness in the late JOW period and by low litterfall, high soil nutrient content, and high soil moisture in the post‐JOW period.

  • Our study provides evidence that temporal changes in environmental conditions and seedling densities occurring after disruption by JOW differ from the changes in forests impacted by other natural disturbances.

Keywords: Environmental conditions, functional traits, Japanese oak‐wilt disease, seedling establishment, structural equation model

This study revealed that seedling density and functional traits are influenced by changing environmental conditions across different periods of Japanese oak wilt disease disturbance, with key drivers of seedling regeneration varying over time.

graphic file with name PLB-27-1097-g007.jpg

INTRODUCTION

Natural disturbances, such as fire, weather events (hurricane and typhoon), and insect infestations, are important factors that influence tree re‐assembly processes in forests. Ecologists have paid much attention to the tree regeneration process in forests after hurricanes, fires, and typhoons (Shimizu 2005; Comita et al. 2009; Luo et al. 2021). Considering over 12.5 million hectares of forest are disturbed by forest disease globally, including more than 3 million hectares of temperate forest (van Lierop et al. 2015), diseases also act as major natural disturbances that affect forest structure (Brunet et al. 2014; Nakajima & Ishida 2014; Watanabe et al. 2016; Dudney et al. 2020). Changes in tree demography after disease disturbance might re‐shape the species composition and community structure of adult trees and influence seedling dynamics by altering environmental conditions on the forest floor. The seedling stage is a bottleneck stage in the tree life cycle, and the community assembly process of tree seedlings strongly influences future forest structure (Bunker & Carson 2005; Takenaka 2006). However, studies regarding the effects of disease on seedling dynamics remain limited (Shimizu 2005; Nagashima et al. 2019).

Japanese oak wilt disease (JOW) is a major fungal diseases present in warm‐temperate secondary forests in Japan. The disease is caused by Raffaelea quercivora, carried by the ambrosia beetle (Platypus quercivorus); this insect invades the water‐conducting tissues of Fagaceae trees, such as Quercus serrata, Q. variabilis, and Q. acutissima (Kuroda & Yamada 1996; Kuroda 2001; Kinuura & Kobayashi 2006; Kuroda et al. 2012). The stem dynamics of adult trees in JOW‐affected forests have been extensively studied (Saito & Shibata 2012; Nakajima & Ishida 2014; Watanabe et al. 2016; Hata et al. 2017). Nakajima (2019) reported that mortality rates for Q. crispula and Q. serrata after a 10‐year JOW epidemic were 68% and 21%, respectively, in Toyama Prefecture, central Japan. Saito & Shibata (2012) found that Q. crispula had extremely high mortality (90%–100%) in a JOW‐affected forest.

Because the basal area (BA) of canopy trees is drastically decreased by JOW infection, changes to the forest floor environment, such as increased light levels, might affect seedling recruitment. A previous study in Ogasawara Island showed that the community structure and species composition of seedlings and saplings in Pinus–Schima secondary forest changed after pine wilt disease and typhoon disturbances (Shimizu 2005). Nagashima et al. (2019) studied seedling regeneration after removal of the dead oak trees and clear‐cutting in JOW‐affected forests. These authors reported that clear‐cutting improved seedling recruitment. However, dead oak trees usually remain unmanaged in JOW‐affected forests. Moreover, the environmental conditions at the forest floor will shift after disease disturbance, according to the recovery of canopy trees. Apart from the above studies, the temporal dynamics of seedling communities and their associated impact factors in JOW‐affected, unmanaged forests remain poorly understood. Comita et al. (2009) reported that canopy openness significantly affects seedling survival in the first census interval, whereas soil type significantly affects seedling survival in the final census interval after hurricane disturbances. To clarify the effects of JOW on seedling recruitment over a long period will improve the broader understanding of seedling regeneration after disease disturbance.

The recruitment of tree seedlings is often controlled by environmental conditions on the forest floor, including light, litterfall, soil nutrients, and soil water availability (Nicotra et al. 1999; Padilla & Pugnaire 2007; Gillman 2016; Kremer et al. 2019; Martini et al. 2020; Zhou et al. 2023). In recent years, attempts have been made to predict the directions of changes in tree seedling communities based on environmental conditions and seedling functional traits (Meineri et al. 2020; Umaña et al. 2021; Blonder et al. 2023). There are sets of statistical tools to describe factors that influence seedling regeneration, such as generalized linear models (GLMs), generalized linear mixed models (GLMMs), and structural equation models (SEMs) (Pu et al. 2020; Helluy et al. 2021; Jiang et al. 2022; Li et al. 2022). Among these methods, SEMs constructed by the piecewiseSEM package in R software are powerful tools for estimating both direct and indirect effects; they have recently been used to model complex multivariate relationships to test factors that influenced seedling dynamics (Duan et al. 2022; Jiang et al. 2022; Yin et al. 2023). An understanding of how environmental conditions and seedling functional traits combine to influence seedling dynamics, especially in disturbed forests, is the first goal in efforts to clarify processes of reassembly in tree communities.

Functional traits are morphological, physiological, and phenological characteristics that indirectly influence fitness through their effects on survival, growth, and reproduction (Violle et al. 2007). For example, leaf traits are major functional traits, and reflect resource acquisition strategies, water transport, and adaptation to environmental conditions (John et al. 2017; de la Riva et al. 2018; Li et al. 2018). Specific leaf area (SLA), affected by light conditions and temperature (Meziane & Shipley 1999; Wilson et al. 1999; Kenzo et al. 2015), is strongly associated with plant growth. Leaf dry matter content (LDMC), a proxy for leaf density, is positively correlated with leaf lifespan and negatively correlated with SLA (Wilson et al. 1999; Wright & Cannon 2001). The functional traits of stems and roots are also vital for seedling growth and survival. Plants with low stem dry matter content (SDMC) and root dry matter content (RDMC) tend to have high growth rates and can quickly acquire resources (Shipley & Vu 2002; Birouste et al. 2014).

In this study, we aimed to determine the effects of environmental conditions and seedling functional traits on seedling density in a JOW‐affected, warm‐temperate secondary forest: the Kaisho Forest, Aichi prefecture, Japan. Watanabe et al. (2016) reported mass mortality of Q. serrata and Q. variabilis (ca. 60%) during 2008–2015 due to JOW at this study site. Because of this high mortality, the BA of deciduous trees declined, whereas the BA and stem density of evergreen trees gradually increased (Celegeer et al. 2024). We sought to answer the following questions: (i) How do environmental conditions and seedling densities change after JOW disturbance; (ii) do the effects of environmental factors on functional traits and seedling density change across JOW disturbance periods; and (iii) are environmental conditions more important drivers of seedling density, compared with seedling functional traits? We hypothesized that seedling densities would immediately increase during the peak JOW period due to improved light availability caused by canopy gaps; and that seedling density would be strongly associated with light conditions.

MATERIAL AND METHODS

Study site

The study was conducted in the Kaisho Forest (530 ha) in Seto City, Aichi Prefecture, Central Japan (35° 11′ 25″ N, 137° 06′ 55″ E; 120–380 m a.s.l.). This forest is classified as a warm‐temperate secondary forest and was used by locals to collect firewood until the 1960s (Hada et al. 1999). Mean annual temperature was 13.3°C and mean annual precipitation was 1903 mm between 2009 and 2020 (Ecohydrology Research Institute, The University of Tokyo 2020). Mean monthly temperature is highest in August (25.3°C) and coldest in January (1.6°C). In 2008, three permanent plots were established in the granite area of the Kaisho Forest to monitor adult trees with diameter at breast height (DBH) ≥5 cm. Mainly due to differences in length of time since the last logging, tree species composition varied among plots. Plots 1 and 2, which were both 0.25 ha (50 m × 50 m) in size, were dominated by Q. serrata, Clethra barbinervis, and Ilex pedunculosa (Celegeer et al. 2024). Plot 3 (0.16 ha; 40 m × 40 m) was dominated by Q. serrata, Q. variabilis, and Q. glauca. The understory layer consists of evergreen and deciduous broadleaved trees, such as Gamblea innovans, Eurya japonica, Cleyera japonica, and Acer crataegifolium in Plots 1 and 2, with Q. glauca, Ligustrum japonicum, and Camellia japonica in Plot 3 (Watanabe et al. 2016). In 2009, 16 quadrats (each 2 m × 2 m) were established in the upper right corner of a 10 m × 10 m grid in the center area (30 m × 30 m) of each plot to monitor seedling establishment and survival. Watanabe et al. (2016) reported JOW in the Kaisho Forest since 2008 and observed that 55% of Q. serrata and 62% of Q. variabilis trees died in the three permanent plots between 2008 and 2015. To evaluate the natural impact of JOW on seedling establishment, standing dead trees killed by JOW were left undisturbed, that is, not removed.

Measurement of environmental conditions

To determine changes in environmental conditions, we quantified light, soil water content, and litterfall input to each seedling quadrat in the three plots. Hemispherical photographs were taken using a Nikon Cool Pix 950 camera (Nikon, Tokyo, Japan) equipped with a fisheye lens during the leafy season in 2009, 2011–2013, and 2018–2020. Photographs were taken at a height of 50 cm at the center of each quadrat in cloudy conditions in each year. Percentage canopy openness in each quadrat, a proxy for light condition, was estimated from these photographs using ‘Hemispherical 2.0’ in ImageJ software (Beckschäfer 2015).

Soil water content (%) in each quadrat was measured using a soil moisture sensor (HH2 Moisture Meter; Delta‐T, Cambridge, UK) during the summers of 2010–2013 and 2018–2020. In each quadrat, triplicate measurements of soil water content were taken after dry, sunny weather had persisted for at least 1 week; mean values were calculated from the triplicate measures. Sixteen litter traps (each with a surface area of 0.5 m2) were established near seedling quadrats in each plot to collect litterfall. The litterfall was collected approximately once per month, then oven‐dried for 48 h at 80°C (Celegeer et al. 2024). The amount of total litterfall in each litter trap during 2009–2011, 2013, and 2018–2019 was included as an environmental condition (Celegeer et al. 2024).

To avoid disturbance within the quadrats, surface soil samples (100 cm3) were collected outside each quadrat using a 5‐cm diameter soil corer in 2011, then stored in plastic bags. Surface litter was removed prior to soil sample collection. Samples were air‐dried for 2 weeks, sieved with a 2 mm mesh to remove fine roots and stones, then ground. Soil nutrients, including calcium (Ca), potassium (K), magnesium (Mg), sodium (Na), sulfur (S), and silica (Si; mg 100 g−1), were measured using inductively coupled plasma atomic emission spectroscopy (IRIS ICAP; Nippon Jarrell Ash, Japan). Estimates of carbon (C), nitrogen (N), and phosphorus (P) levels were obtained using the soil samples collected with the same method in 2021. The C/N ratio was estimated using a C–N analyser (Macro corder JM1000CN; J‐Science Lab, Kyoto, Japan). Available P was determined using the Truog method (Truog 1930).

Seedling census and functional traits of seedlings

All tree seedlings (height ≤50 cm) within 16 4‐m2 quadrats in each plot were tagged, identified to species level, monitored, and assessed to determine their condition (dead or alive). This seedling census was conducted three times per year in spring (April–May), summer (June–July), and autumn (October–November) during 2009–2013 and 2018–2020. Seedlings were classified as current year or non‐current year stems based on the presence of cotyledons, number of leaves, height, and stem lignification; seedlings that had emerged from seeds within the census year were classified as current year seedlings, and seedlings that had emerged at least 1 year prior were classified as non‐current year seedlings. Seedlings that lacked green leaves, had dried out, or had broken, snapped, or had irreparably damaged stems were considered dead. We counted current year and non‐current year seedlings in each seedling quadrat, then calculated seedling densities (individuals m−2) per species for both seedling stages in each quadrat for each census year.

We selected 13 species representing 10 genera and 9 families that constituted 74.48 ± 21.06% of the total number of current year seedlings throughout the study period (Table S1); we used these species to investigate 11 below‐ and aboveground functional traits of current year seedlings. Traits included seedling height, leaf area (LA), SLA, LDMC, SDMC, RDMC, leaf nitrogen concentration (LNC), seed mass, root length (RL), specific root length (SRL), and aboveground to belowground biomass ratio (AB ratio), all of which have been associated with seedling survival strategies (Diaz et al. 1998; Meziane & Shipley 1999; Wilson et al. 1999; Shipley & Vu 2002; Birouste et al. 2014; Li et al. 2018; Meineri et al. 2020; Umaña et al. 2021; Blonder et al. 2023). In July and August 2019 and 2020, we excavated 5–24 whole individuals of each of the 13 species from locations adjacent to the plots, then stored them in plastic bags.

The aboveground height (cm), RL (cm) and stem diameter at the base (D0; mm) of each individual was measured; the number of leaves was also counted. All leaves from each individual were scanned (CanoScan LiDE 200; Tokyo, Japan) to obtain image data, and the LA (cm2) for each leaf was estimated using ImageJ v.1.53a software. Fresh mass for leaves (g), stems (g), and roots (g) was measured; tissues were subsequently oven‐dried for 72 h at 60°C. We also determined dry mass of leaves (g), stems (g), and roots (g) to calculate SLA (cm2 g−1), SRL (cm g−1), LDMC (mg g−1), SDMC (mg g−1), RDMC (mg g−1), and AB ratio (g g−1). We measured LNC (mg 100 mg−1) using a C–N analyser (Macro corder JM1000CN; J‐Science Lab, Japan) after leaves had been ground with a bead mill (Tissue Lyser; QA‐85200, Tokyo, Japan, Qiagen). Mean seed mass (g) was calculated by weighing mature seeds collected in the study area (Celegeer et al. 2024).

Data analysis

To determine the effect of JOW on environmental conditions and seedling dynamics, we compared canopy openness, soil water content, and seedling density during three periods: (1) peak JOW period (2009–2010, with Q. serrata and Q. variabilis mortality rates of 22.25% yr−1 and 24.28% yr−1, respectively); (2) late JOW period (2011–2013, with Q. serrata and Q. variabilis mortality rates of 10.36% yr−1 and 13.52% yr−1, respectively); and (3) post‐JOW period (2018–2020; with Q. serrata and Q. variabilis mortality rates of 1.16% yr−1 and 0% yr−1, respectively), based on previous studies in the Kaisho Forest (Watanabe et al. 2016; Celegeer et al. 2024). Canopy openness and soil water content were compared among JOW periods using the Kruskal–Wallis test, and pairwise Wilcoxon rank‐sum test with Holm corrections for post‐hoc tests. These methods were also used to compare the densities for current year and non‐current year seedlings among JOW periods. To summarize soil nutrients in each quadrat, we performed principal components analysis (PCA) after z‐transformation of soil nutrient variables in each quadrat. Predicted scores were obtained from the first (Soil PC1) and second (Soil PC2) principal components for each quadrat. We also calculated the mean values of the 11 functional traits for each species, then performed PCA after z‐transformation of functional trait variables for 13 evaluated species. Next, predicted scores were obtained from the first (Trait PC1) and second (Trait PC2) principal component for each species.

Finally, we constructed structural equation models (SEMs) to determine the direct and indirect effects of seedling functional traits and environmental conditions on the density of 13 species of current year seedling in autumn of each year. Only data for each species' current year seedlings were included in our model because the current year stage is the key stage for seedling recruitment. To examine temporal differences in factors that influence seedling density in autumn, we modelled each SEM for the late JOW and post‐JOW periods. Due to the absence of data regarding soil water content for 2009 and canopy openness for 2010, we did not construct an SEM for the peak JOW period. The late JOW and post‐JOW models included 2011 and 2013 and 2018–2019 data, respectively. The SEMs were constructed under the assumption that soil nutrient levels do not differ significantly. The models represented direct paths from functional traits and environmental conditions to seedling density, and indirect paths from environmental conditions to seedling density via functional traits (Fig. 1). Before generation of the models, seedling densities in autumn for each species in each quadrat were log‐transformed following the addition of 0.5 to ensure normal distributions. To reduce the dimensionality of soil nutrients and functional traits, predicted scores from PC1 and PC2 for each quadrat were used in the SEMs. Species was included as a random effect in all models. Nonsignificant effects were removed to improve model optimization (P > 0.05). Akaike's information criterion (AICc) and the Chi‐squared test (P > 0.05) were used to evaluate fitness of each model (Shipley 2009).

Fig. 1.

Fig. 1

Open in a new tab

Structural equation models (SEMs) representing connections between environmental conditions (soil water content, canopy openness, litterfall, soil PC1, and soil PC2), seedling functional traits (trait PC1 and trait PC2) and seedling density in autumn of each species in each quadrat. Species was included as a random effect.

All statistical analyses were performed using R software (v. 4.3.0; R Development Core Team 2023). PCA was performed using the R package “factoextra” (Kassambara & Mundt 2017), and SEM analysis was performed using the R packages “nlme” (Pinheiro & Bates 2006) and “piecewiseSEM” (Lefcheck 2016).

RESULTS

Changes in environmental conditions and seedling density

Canopy openness and soil water content significantly differed among JOW periods in all three plots (Fig. 2; P < 0.05). After JOW disturbance, canopy openness increased in all plots, whereas the soil water content decreased (Fig. 2). Soil PC1, which explained 43.3% of the total variation in soil nutrient status, reflected gradients of Si, Ca, K, P, and Mg (Fig. 3). Soil PC2, which explained 18.9% of the total variation, was positively correlated with the C/N ratio and S, whereas it was negatively correlated with Na (Fig. 3). PCA indicated no substantial differences in soil nutrient status among the three study plots (Table S3, Fig. S1).

Fig. 2.

Fig. 2

Open in a new tab

Comparisons of canopy openness (a) and soil water content (b) among JOW periods in each plot in the Kaisho Forest. Colours indicate periods: black, peak JOW; grey, late JOW; white, post‐JOW. The vertical lines at each end of the box represent the range of the maximum and minimum values, and the central line inside the box indicates the median value. Dots placed beyond the vertical lines represent outliers that are >1.5 times further away than the interquartile range indicated by boxes. Letters denote significant differences among JOW periods in each plot (post‐hoc multiple comparisons test; P < 0.05).

Fig. 3.

Fig. 3

Open in a new tab

Biplot for principal components analysis of soil nutrients variables across all seedling quadrants. Leading principal component axes are shown with arrows. Shapes indicate plots: circles, plot 1; regular triangles, plot 2; and squares, plot 3.

During 2009–2013 and 2018–2020, we observed 13,010 current year seedlings (59 woody species, 44 genera, 21 families) and 9810 non‐current year seedlings (68 woody species, 51 genera, 33 families). The densities of current year and non‐current year seedlings for 13 dominant species significantly differed among JOW periods in all three plots (Fig. 4; P < 0.05). The temporal fluctuation patterns of current year seedlings differed among plots. Current year seedling density largely increased throughout JOW periods in Plot 1, whereas the densities of current year seedlings in Plots 2 and 3 were highest in the late JOW period (Fig. 4a). The density of non‐current year seedlings was highest in the post‐JOW period in all three plots (Fig. 4b).

Fig. 4.

Fig. 4

Open in a new tab

Comparisons of density for current year seedlings (a) and non‐current year seedlings (b) of 13 dominant species among JOW periods in each plot. Colours indicate periods: black, peak JOW; grey, late JOW; white, post‐JOW. The vertical lines at each end of the box represent the range of the maximum and minimum values, and the central line inside the box indicates the median value. Dots placed beyond the vertical lines represent outliers that are >1.5 times further away than the interquartile range indicated by boxes. Letters denote significant differences among JOW periods in each plot (post‐hoc multiple comparisons test; P < 0.05).

Characteristics of seedling functional traits

The mean heights of current year seedlings in the 13 species were 3.99–21.08 cm; LA, RL, and seed mass were 1.68–10.11 cm2, 4.99–21.32 cm, and 0.01–0.63 g, respectively (Table S1). LDMC, SDMC, and RDMC were lowest in Toxicodendron trichocarpum (LDMC: 178.93 mg g−1; SDMC: 140.47 mg g−1; RDMC: 139.33 mg g−1). Quercus serrata had the highest LDMC (361.77 mg g−1), Q. glauca had the highest SDMC (367.99 mg g−1), and Acer palmatum had the highest RDMC (371.23 mg g−1). The ranges of SLA, LNC and SRL were 256.40–773.77 cm2 g−1, 1.25–6.74 mg 100 mg−1 and 60.52–1192.02 cm g−1, respectively. Trait PC1 explained 55.9% of the variation in functional traits and was strongly correlated with height, RL, seed mass, LA, SLA, AB ratio, and SRL; Trait PC2 (18.3%) was strongly correlated with LDMC, SDMC, and RDMC (Fig. 5). Current year seedlings of Q. serrata, Q. variabilis, and Q. glauca, which had high values for height (11.25–21.08 cm), LA (7.68–10.11 cm2), SDMC (310.10–367.99 mg g−1), and seed mass (0.20–0.63 g), and low values for SRL (60.52–157.36 cm g−1) were clustered with positive values in Trait PC1. Acer palmatum, A. crataegifolium, and Cerasus jamasakura, which had high RDMC values (339.46–371.23 mg g−1) were clustered with positive values of Trait PC2.

Fig. 5.

Fig. 5

Open in a new tab

Biplot for principal components analysis of seedling functional trait variables across 13 species. Leading principal component axes are shown with arrows. Letters in black denote species IDs shown in Table S1.

Effects of environmental conditions and functional traits on seedling density

Between 26.7% and 100% of current year seedlings survived during spring to autumn in the late JOW period; 8.1%–100% survived during spring to autumn in the post‐JOW period (Table S1). The final piecewise SEMs for the late JOW and post‐JOW periods explained 27% and 46% of the variation in seedling density in autumn, respectively (Fig. 6; late JOW: χ 2 = 4.706, df = 3, P = 0.195; post‐JOW: χ 2 = 3.14, df = 4, P = 0.534). Predictors of seedling density in the late JOW period were canopy openness and Soil PC2; in the post‐JOW period, the predictors were soil water content, litterfall, and Soil PC1. The direct effects of canopy openness and Soil PC2 on seedling density were significant; both factors positively influenced seedling density in the late JOW period (Fig. 6a). In the post‐JOW period, seedling density was negatively influenced by litterfall and positively influenced by soil water content and Soil PC1 (Fig. 6b). Trait PC1 was negatively correlated by Soil PC2 in the late JOW period and positively correlated with litterfall in the post‐JOW period. Trait PC2 was negatively correlated with Soil PC1 in the late JOW period and negatively correlated with litterfall in the post‐JOW period.

Fig. 6.

Fig. 6

Open in a new tab

Structural equation models (SEMs) for current year seedling density in autumn (a) in late JOW and (b) post‐JOW periods in the Kaisho Forest. Solid arrows with minus values and positive values represent negative (P < 0.05) and positive paths (P < 0.05), respectively. Dashed grey arrows represent no significant paths (P > 0.05). Values adjacent to arrows represent regression coefficients.

DISCUSSION

Effects of JOW on environmental conditions and seedling density

Canopy openness, soil water content, and seedling density all temporally varied in JOW‐affected, warm‐temperate secondary forests. Light availability at the forest floor gradually increased throughout the JOW periods; thus, canopy openness values in 2020 were 1.60–1.81‐fold higher than the initial values observed in 2009. This increase in canopy openness is a result of high mortality in the canopy oak species, Q. serrata and Q. variabilis, which occurred in the Kaisho Forest during the JOW disturbance (Watanabe et al. 2016). A previous study in hurricane‐damaged tropical forests revealed a four‐fold increase in canopy openness immediately after a hurricane, but the canopy closed within 3 years (Comita et al. 2009). This result is inconsistent with the temporal pattern of canopy openness observed in JOW‐affected forests. Canopy gaps were rapidly formed by hurricanes or typhoons and then re‐closed via recruitment of understory trees or recovery of the canopy. In contrast, standing‐dead trees killed by JOW might require several years to decay, and changes in light conditions gradually occur as a result. This trend has been observed elsewhere in a forest stand disturbed by pine wilt disease and typhoons (Shimizu 2005); light availability gradually increased after mass dieback of pine trees due to pine wilt disease, whereas light availability immediately increased after typhoon disturbance. However, the range of canopy openness values observed in our study (1.7%–14.1%) remained lower in JOW‐affected forests than in pine wilt disease‐disturbed forest (mean ± SD: 23.3 ± 3.0%; Wu & Ando 2010) and was similar to canopy openness values from warm‐temperate forests without JOW infection (4.6%–14.3%; Enoki & Abe 2004; Tsujino & Yumoto 2007). Although canopy openness might continue to gradually increase after the post‐JOW period, the effect of JOW on canopy openness at the study site should be considered subtle but continuous.

Soil water availability at the forest floor gradually decreased during the study period. Values for soil water content in 2020 were ca. 40.8%–56.5% lower than the initial values observed in 2010. Climate data from the area indicate no significant decreasing trends in total annual or growing season precipitation during the study period (Table S2). Soil moisture readings were taken using a consistent protocol during the study period: after at least 7 days of sunny, dry weather. The litterfall layer might act to retain the soil moisture, but no extreme shift in litterfall production was observed at the study site (Gillman 2016; Kremer et al. 2019; Celegeer et al. 2024). Therefore, it is likely that canopy gaps created by JOW were more susceptible to drying through evaporation, leading to reduced soil moisture.

Current year seedling recruitment increased in the late JOW period, which occurred 3–5 years after the onset of JOW in the study area. Recruitment of non‐current year seedlings increased in the post‐JOW period (after 9 years of JOW infection) following the improvement in current year seedling establishment. This pattern is again inconsistent with results obtained from typhoon and hurricane‐affected forests. For example, seedling densities increased over a single year in a hurricane‐disturbed forest and in a forest with pine wilt disease and typhoon disturbances (Shimizu 2005; Comita et al. 2009). Our results suggest that the rate of seedling recruitment in JOW‐affected forests gradually increases, in line with gradual changes in environmental conditions. The temporal pattern of current year seedling densities differed among the plots, such that the density of current year seedlings remained high in the post‐JOW period in Plot 1 but decreased in the post‐JOW period in Plots 2 and 3. This result is likely a product of differences in species composition and topography among the study plots. Plot 1, which was dominated by Q. serrata, showed high seedling density after masting of Q. serrata in 2018.

Effects of environmental conditions on seedling functional traits

We found significant effects of environmental conditions on seedling functional traits and changes in impact factors during JOW periods. SEM results indicated that seedlings with large seed mass, greater height, and longer roots (such as Q. serrata, Q. variabilis, and Q. glauca) had increased recruitment in areas with low soil C/N ratio and soil S content, as well as high soil Na content. Seedlings with high LDMC, RDMC and SDMC (such as A. palmatum, A. crataegifolium, and C. jamasakura) had higher recruitment in areas with low soil nutrients in the late JOW period. These results indicate that the functional traits of current year seedlings in JOW‐affected secondary forests could be influenced by soil nutrients in the late JOW period. Previous studies have demonstrated that soil nutrients are among vital factors that impact seedling recruitment (Bentos et al. 2013; Martini et al. 2020). Additionally, a study of seasonally moist tropical forests in Panama showed that variations in the relationships of functional traits and vital rates could be explained by soil P (Browne et al. 2023). In the post‐JOW period, factors impacting the functional traits of current year seedlings shifted from soil nutrients to the amount of litterfall. The high amount of litterfall enhanced the recruitment of seedlings with large seed mass, greater height and longer roots (such as Q. serrata, Q. variabilis, and Q. glauca) but limited the establishment of seedlings with high LDMC, RDMC and SDMC (e.g., A. palmatum, A. crataegifolium, and C. jamasakura). Large‐seeded species tend to have high seedling establishment rates (Moles & Westoby 2006). Therefore, seedlings of Q. serrata, Q. variabilis, and Q. glauca may have exhibited increased capacity to establish in areas with thick litterfall. Although distinct changes in litterfall production were not observed after JOW convergence (Celegeer et al. 2024), differences in litterfall properties between deciduous and evergreen trees have been reported, which would alter the amount and quality of litterfall on the forest floor (Kavvadias et al. 2001; Barlow et al. 2007). Evergreen leaves have slower decomposition rates compared with deciduous leaves (Pérez‐Harguindeguy et al. 2000). The slow decomposition of evergreen litter might increase litter accumulation on the forest floor. Because the BA and tree density of evergreen trees are increasing at the study site (Celegeer et al. 2024), the change in litterfall quality may have caused temporal alterations to factors affecting seedling functional traits from the late JOW period to the post‐JOW period.

Drivers of seedling establishment

Throughout this long‐term investigation, our study plots have remained unaffected by large herbivores such as deer. Additionally, observations of seed and seedling damage caused by small mammals, including mice and other rodents, were rare. Therefore, this study focused on the effects of environmental conditions and seedling functional traits on seedling establishment. SEM results indicate that environmental conditions significantly affected seedling establishment, and the driving variables of seedling density varied according to JOW period. Our results suggest that high seedling density occurred in areas with high canopy openness caused by JOW‐mediated tree mortality in the late JOW period, consistent with the findings of previous studies (Marod et al. 2004; Chao et al. 2022). Similar positive effects of light availability on seedling density have been reported following other natural disturbances, such as hurricanes, fires, and pine wilt disease (Shimizu 2005; Comita et al. 2009; Luo et al. 2021). The SEM results also indicate that C/N ratio and soil Na content were associated with seedling establishment in the late JOW period. Although soils with a low C/N ratio tend to have high decomposition rates, soils with high Na content (salt stress) are often associated with reduced rates of photosynthesis and plant growth (Parida & Das 2005). Therefore, a high C/N ratio and low soil Na content are more likely to promote seedling survival. In the post‐JOW period, high seedling establishment was found in areas with low litterfall, high nutrient levels, and high soil moisture. Previous studies have indicated that soil macronutrients such as Mg and Ca, along with soil moisture are important for seedling growth and photosynthetic activity (Veenendaal et al. 1996; Engelbrecht et al. 2005; Martini et al. 2020). Hata et al. (2010) reported that litterfall negatively influenced seed germination and seedling growth. Because environmental conditions change after JOW disturbance, most influential variables impacting seedling density might also shift, such that litterfall and soil nutrients become relatively important for seedling establishment in the post‐JOW period. Similar temporal shifts in variables affecting seedling survival have been observed following hurricane disturbance, and canopy openness had a significant effect on seedling survival in the first census interval but not in subsequent intervals (Comita et al. 2009).

The low predictive power of functional traits for seedling performance is consistent with findings in previous studies that predicted seedling growth and survival using functional traits (Kraft et al. 2010; Wright et al. 2010; Visser et al. 2016). These results may reflect the fact that the growth and survival of current year seedlings is more strongly influenced by abiotic conditions, considering the substantial variability in environmental conditions in the study area. Light and soil moisture conditions have continued to increase and decrease since the end of the JOW outbreak in the Kaisho Forest. When environmental conditions become more stable, it is possible that seedling functional traits will play a larger predictive role in seedling dynamics. Further accumulation of long‐term demographic information concerning seedlings and micro‐environmental conditions, along with additional quantification of functional traits for a larger number of species, may improve the accuracy of predictions regarding seedling dynamics in recovering JOW‐affected secondary forests.

Changes in seedling dynamics and environmental conditions following JOW disturbance differed from the changes in forests disturbed by other natural pulse‐like disturbances. More research is needed to determine the effects of diseases on seedling recruitment. In addition to JOW, diseases such as Dutch elm disease, white pine blister rust, and pine wilt disease may significantly impact forest ecosystems. Climate change might alter expansion of those diseases in forests. The present results highlight the need to further investigate seedling regeneration and its drivers following disease‐mediated forest disturbance.

AUTHOR CONTRIBUTIONS

MN and Celegeer designed the study; Celegeer, NW, KO, TO, and MN conducted fieldwork; Celegeer and KO carried out laboratory experiments; Data analysis was conducted by Celegeer, assisted closely by MM; MN supervised the study; Celegeer wrote the first draft; all authors contributed to the subsequent manuscript revision.

FUNDING INFORMATION

JST SPRING, Grant Number: JPMJSP2125; The Ecological Society of Japan (Chubu District); Kaisho Forest Center in Aichi prefecture.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflict of interest.

Supporting information

Fig. S1. Heatmap of each soil nutrient concentration in each plot of the Kaisho Forest. The color scale ranges from light blue to dark blue and represents the variation in each nutrient level from lower to higher concentrations. Since the soil nutrient data were collected from each seedling quadrat, the area of the heatmaps for all three plots is standardized to 40 m × 40 m.

PLB-27-1097-s001.jpg (1.5MB, jpg)

Table S1. Mean values of survival rate in each JOW period and mean (±SD) of seedling functional traits for each species in the Kaisho Forest.

Table S2. Mean annual temperature (oC) and annual and growing season precipitation (mm) during 2009 to 2013 and 2018 to 2020.

Table S3. Mean (±SD) of each soil nutrient in each plot of the Kaisho Forest. Soil nutrients were measured using the surface soil. The detailed method for measuring each nutrient was described in the text.

PLB-27-1097-s002.docx (259.1KB, docx)

ACKNOWLEDGEMENTS

We thank the Kaisho Forest Center in Aichi prefecture for permitting research activities for a decade. We thank Dr. H. Kurokawa of Forestry and Forest Products Research Institute for suggestions and assistance in statistical analysis. We also thank Prof. N. Tomaru and Dr. K. Ogawa of the Laboratory of Forest Ecology at Nagoya University for helpful advice regarding our work. Special thanks to all members of the Laboratory of Forest Ecology at Nagoya University who enthusiastically assisted the measurements of environmental conditions, seedling census, and seedling sampling.

Editor: B. Hu

DATA AVAILABILITY STATEMENT

All archived data (Celegeer et al. 2024) are freely available at: DOI: 10.13140/RG.2.2.14737.40802.

REFERENCES

  1. Barlow J., Gardner T., Ferreira L., Peres C. (2007) Litter fall and decomposition in primary, secondary and plantation forests in the Brazilian Amazon. Forest Ecology and Management, 247, 91–97. [Google Scholar]
  2. Beckschäfer P. (2015) Hemispherical_2.0. Batch Processing Hemispherical and Canopy Photographs with ImageJ, User Manual.
  3. Bentos T.V., Nascimento H.E., Williamson G.B. (2013) Tree seedling recruitment in Amazon secondary forest: Importance of topography and gap micro‐site conditions. Forest Ecology and Management, 287, 140–146. [Google Scholar]
  4. Birouste M., Zamora‐Ledezma E., Bossard C., Pérez‐Ramos I.M., Roumet C. (2014) Measurement of fine root tissue density: A comparison of three methods reveals the potential of root dry matter content. Plant and Soil, 374, 299–313. [Google Scholar]
  5. Blonder B.W., Gaüzère P., Iversen L.L., Ke P.J., Petry W.K., Ray C.A., Salguero‐Gómez R., Sharpless W., Violle C. (2023) Predicting and controlling ecological communities via trait and environment mediated parameterizations of dynamical models. Oikos, 2023, e09415. [Google Scholar]
  6. Browne L., Markesteijn L., Manzané‐Pinzón E., Wright S.J., Bagchi R., Engelbrecht B.M., Jones F.A., Comita L.S. (2023) Widespread variation in functional trait–vital rate relationships in tropical tree seedlings across a precipitation and soil phosphorus gradient. Functional Ecology, 37, 248–260. [Google Scholar]
  7. Brunet J., Bukina Y., Hedwall P.O., Holmström E., von Oheimb G. (2014) Pathogen induced disturbance and succession in temperate forests: Evidence from a 100‐year data set in southern Sweden. Basic and Applied Ecology, 15, 114–121. [Google Scholar]
  8. Bunker D.E., Carson W.P. (2005) Drought stress and tropical forest woody seedlings: Effect on community structure and composition. Journal of Ecology, 93, 794–806. [Google Scholar]
  9. Celegeer, Watanabe N., Otani K., Okada T., Toyama C., Nakagawa M. (2024) Dynamics of litterfall production in a forest damaged by oak wilt disease: A case study in a warm‐temperate secondary forest. Journal of Forest Research, 29, 130–136. [Google Scholar]
  10. Chao K.J., Lin Y.C., Song G.Z.M., Liao C.H., Kuo Y.L., Hsieh C.F., Schupp E.W. (2022) Understorey light environment impacts on seedling establishment and growth in a typhoon‐disturbed tropical forest. Plant Ecology, 223, 1007–1021. [Google Scholar]
  11. Comita L.S., Uriarte M., Thompson J., Jonckheere I., Canham C.D., Zimmerman J.K. (2009) Abiotic and biotic drivers of seedling survival in a hurricane‐impacted tropical forest. Journal of Ecology, 97, 1346–1359. [Google Scholar]
  12. de la Riva E.G., Villar R., Pérez‐Ramos I.M., Quero J.L., Matías L., Poorter L., Marañón T. (2018) Relationships between leaf mass per area and nutrient concentrations in 98 Mediterranean woody species are determined by phylogeny, habitat and leaf habit. Trees, 32, 497–510. [Google Scholar]
  13. Diaz S., Cabido M., Casanoves F. (1998) Plant functional traits and environmental filters at a regional scale. Journal of Vegetation Science, 9, 113–122. [Google Scholar]
  14. Duan H., Resco de Dios V., Wang D., Zhao N., Huang G., Liu W., Wu J., Zhou S., Choat B., Tissue D.T. (2022) Testing the limits of plant drought stress and subsequent recovery in four provenances of a widely distributed subtropical tree species. Plant, Cell & Environment, 45, 1187–1203. [DOI] [PubMed] [Google Scholar]
  15. Dudney J.C., Nesmith J.C., Cahill M.C., Cribbs J.E., Duriscoe D.M., Das A.J., Stephenson N.L., Battles J.J. (2020) Compounding effects of white pine blister rust, mountain pine beetle, and fire threaten four white pine species. Ecosphere, 11, e03263. [Google Scholar]
  16. Ecohydrology Research Institute, The University of Tokyo (2020) Annual report of meteorological data at Shirasaka during 2009–2020. http://www.uf.a.u‐tokyo.ac.jp/research_division/data/kishou/21c/annual/2001‐2020/Sshi_annual2001‐2020.pdf [Accessed on 9 July 2023]
  17. Engelbrecht B.M., Kursar T.A., Tyree M.T. (2005) Drought effects on seedling survival in a tropical moist forest. Trees, 19, 312–321. [Google Scholar]
  18. Enoki T., Abe A. (2004) Sapling distribution in relation to topography and canopy openness in an evergreen broad‐leaved forest. Plant Ecology, 173, 283–291. [Google Scholar]
  19. Gillman L.N. (2016) Seedling mortality from litterfall increases with decreasing latitude. Ecology, 97, 530–535. [DOI] [PubMed] [Google Scholar]
  20. Hada Y., Nakamura Y., Noumi Y. (1999) Vegetation of Kaisyo‐no‐mori region – Plant diversity sustained by geomorphic processes. Japanese Journal of Conservation Ecology, 4, 113–123 (in Japanese). [Google Scholar]
  21. Hata K., Iwai N., Sato T., Sawada H. (2017) Species‐specific growth patterns of trees neighboring dead oak trees caused by Japanese oak wilt disease. Journal of Forest Research, 22, 248–255. [Google Scholar]
  22. Hata K., Kato H., Kachi N. (2010) Litter of an alien tree, Casuarina equisetifolia, inhibits seed germination and initial growth of a native tree on the Ogasawara Islands (subtropical oceanic islands). Journal of Forest Research, 15, 384–390. [Google Scholar]
  23. Helluy M., Gavinet J., Prévosto B., Fernandez C. (2021) Influence of light, water stress and shrub cover on sapling survival and height growth: The case of A. unedo, F. ornus and S. domestica under Mediterranean climate. European Journal of Forest Research, 140, 635–647. [Google Scholar]
  24. Jiang F., Cadotte M.W., Jin G. (2022) Size‐ and environment‐driven seedling survival and growth are mediated by leaf functional traits. Proceedings of the Royal Society B, 289, 20221400. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. John G.P., Scoffoni C., Buckley T.N., Villar R., Poorter H., Sack L. (2017) The anatomical and compositional basis of leaf mass per area. Ecology Letters, 20, 412–425. [DOI] [PubMed] [Google Scholar]
  26. Kassambara A., Mundt F. (2017) Package ‘factoextra’. Extract and visualize the results of multivariate data analyses.
  27. Kavvadias V., Alifragis D., Tsiontsis A., Brofas G., Stamatellos G. (2001) Litterfall, litter accumulation and litter decomposition rates in four forest ecosystems in northern Greece. Forest Ecology and Management, 144, 113–127. [Google Scholar]
  28. Kenzo T., Inoue Y., Yoshimura M., Yamashita M., Tanaka‐Oda A., Ichie T. (2015) Height‐related changes in leaf photosynthetic traits in diverse Bornean tropical rain forest trees. Oecologia, 177, 191–202. [DOI] [PubMed] [Google Scholar]
  29. Kinuura H., Kobayashi M. (2006) Death of Quercus crispula by inoculation with adult platypus quercivorus (coleoptera: Platypodidae). Applied Entomology and Zoology, 41, 123–128. [Google Scholar]
  30. Kraft N.J., Metz M.R., Condit R.S., Chave J. (2010) The relationship between wood density and mortality in a global tropical forest data set. New Phytologist, 188, 1124–1136. [DOI] [PubMed] [Google Scholar]
  31. Kremer K.N., Promis Á.A., Mancilla G., Magni C.R. (2019) Leaf litter and irrigation can increase seed germination and early seedling survival of the recalcitrant‐seeded tree Beilschmiedia miersii . Austral Ecology, 44, 86–94. [Google Scholar]
  32. Kuroda K. (2001) Responses of Quercus sapwood to infection with the pathogenic fungus of a new wilt disease vectored by the ambrosia beetle Platypus quercivorus . Journal of Wood Science, 47, 425–429. [Google Scholar]
  33. Kuroda K., Osumi K., Oku H. (2012) Reestablishing the health of secondary forests “Satoyama” endangered by Japanese oak wilt: A preliminary report. Journal of Agricultural Extension and Rural Development, 4, 192–198. [Google Scholar]
  34. Kuroda K., Yamada T. (1996) Discoloration of sapwood and blockage of xylem sap ascent in the trunks of wilting Quercus spp. following attack by Platypus quercivorus . Journal of the Japanese Forest Society, 78, 84–88 (in Japanese with English summary). [Google Scholar]
  35. Lefcheck J.S. (2016) piecewiseSEM: Piecewise structural equation modelling in r for ecology, evolution, and systematics. Methods in Ecology and Evolution, 7, 573–579. [Google Scholar]
  36. Li F., McCulloh K.A., Sun S., Bao W. (2018) Linking leaf hydraulic properties, photosynthetic rates, and leaf lifespan in xerophytic species: A test of global hypotheses. American Journal of Botany, 105, 1858–1868. [DOI] [PubMed] [Google Scholar]
  37. Li Y., Jiang Y., Zhao K., Chen Y., Wei W., Shipley B., Chu C. (2022) Exploring trait–performance relationships of tree seedlings along experimentally manipulated light and water gradients. Ecology, 103, e3703. [DOI] [PubMed] [Google Scholar]
  38. Luo C., Shen Z., Li Y., Han J., Pausas J.G., Xu Q., Zhang Z., Jiang Y. (2021) Determinants of post–fire regeneration demography in a subtropical monsoon‐climate forest in Southwest China. Science of the Total Environment, 766, 142605. [DOI] [PubMed] [Google Scholar]
  39. Marod D., Kutintara U., Tanaka H., Nakashizuka T. (2004) Effects of drought and fire on seedling survival and growth under contrasting light conditions in a seasonal tropical forest. Journal of Vegetation Science, 15, 691–700. [Google Scholar]
  40. Martini F., Xia S.W., Zou C., Goodale U.M. (2020) Seedling growth and survival responses to multiple soil properties in subtropical forests of south China. Forest Ecology and Management, 474, 118382. [Google Scholar]
  41. Meineri E., Klanderud K., Guittar J., Goldberg D.E., Vandvik V. (2020) Functional traits, not productivity, predict alpine plant community openness to seedling recruitment under climatic warming. Oikos, 129, 13–23. [Google Scholar]
  42. Meziane D., Shipley B. (1999) Interacting determinants of specific leaf area in 22 herbaceous species: Effects of irradiance and nutrient availability. Plant, Cell & Environment, 22, 447–459. [Google Scholar]
  43. Moles A.T., Westoby M. (2006) Seed size and plant strategy across the whole life cycle. Oikos, 113, 91–105. [Google Scholar]
  44. Nagashima K., Shimomura T., Tanaka K. (2019) Early‐stage vegetation recovery in forests damaged by oak wilt disease and deer browsing: Effects of deer‐proof fencing and clear‐cutting. Landscape and Ecological Engineering, 15, 155–166. [Google Scholar]
  45. Nakajima H. (2019) Region‐wide mass mortality of Japanese oak due to ambrosia beetle infestation: Mortality factors and change in oak abundance. Forest Ecology and Management, 449, 117468. [Google Scholar]
  46. Nakajima H., Ishida M. (2014) Decline of Quercus crispula in abandoned coppice forests caused by secondary succession and Japanese oak wilt disease: Stand dynamics over twenty years. Forest Ecology and Management, 334, 18–27. [Google Scholar]
  47. Nicotra A.B., Chazdon R.L., Iriarte S.V. (1999) Spatial heterogeneity of light and woody seedling regeneration in tropical wet forests. Ecology, 80, 1908–1926. [Google Scholar]
  48. Padilla F.M., Pugnaire F.I. (2007) Rooting depth and soil moisture control Mediterranean woody seedling survival during drought. Functional Ecology, 21, 489–495. [Google Scholar]
  49. Parida A.K., Das A.B. (2005) Salt tolerance and salinity effects on plants: A review. Ecotoxicology and Environmental Safety, 60, 324–349. [DOI] [PubMed] [Google Scholar]
  50. Pérez‐Harguindeguy N., Díaz S., Cornelissen J.H., Vendramini F., Cabido M., Castellanos A. (2000) Chemistry and toughness predict leaf litter decomposition rates over a wide spectrum of functional types and taxa in central Argentina. Plant and Soil, 218, 21–30. [Google Scholar]
  51. Pinheiro J., Bates D. (2006) Mixed‐effects models in S and S‐PLUS. Springer science & business media, Cham, Switzerland. [Google Scholar]
  52. Pu X., Umaña M.N., Jin G. (2020) Trait‐mediated neighbor effects on plant survival depend on life stages and stage‐specific traits in a temperate forest. Forest Ecology and Management, 472, 118250. [Google Scholar]
  53. R Development Core Team (2023) R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. [Google Scholar]
  54. Saito S., Shibata M. (2012) The forest structure and tree death rate of forest stands damaged by Japanese oak wilt in Yamagata prefecture. Journal of the Japanese Forest Society, 94, 223–228 (in Japanese with English summary). [Google Scholar]
  55. Shimizu Y. (2005) A vegetation change during a 20‐year period following two continuous disturbances (mass‐dieback of pine trees and typhoon damage) in the Pinus‐Schima secondary forest on Chichijima in the Ogasawara (Bonin) islands: Which won, advanced saplings or new seedlings? Ecological Research, 20, 708–725. 10.1007/s11284-005-0087-0 [DOI] [Google Scholar]
  56. Shipley B. (2009) Confirmatory path analysis in a generalized multilevel context. Ecology, 90, 363–368. [DOI] [PubMed] [Google Scholar]
  57. Shipley B., Vu T.T. (2002) Dry matter content as a measure of dry matter concentration in plants and their parts. New Phytologist, 153, 359–364. [Google Scholar]
  58. Takenaka A. (2006) Dynamics of seedling populations and tree species coexistence in a forest: A simulation study. Ecological Research, 21, 356–363. [Google Scholar]
  59. Truog E. (1930) The determination of the readily available phosphorus of soils. Journal ‐ American Society of Agronomy, 22, 874–882. [Google Scholar]
  60. Tsujino R., Yumoto T. (2007) Spatial distribution patterns of trees at different life stages in a warm temperate forest. Journal of Plant Research, 120, 687–695. [DOI] [PubMed] [Google Scholar]
  61. Umaña M.N., Arellano G., Swenson N.G., Zambrano J. (2021) Tree seedling trait optimization and growth in response to local‐scale soil and light variability. Ecology, 102, e03252. [DOI] [PubMed] [Google Scholar]
  62. van Lierop P., Lindquist E., Sathyapala S., Franceschini G. (2015) Global forest area disturbance from fire, insect pests, diseases and severe weather events. Forest Ecology and Management, 352, 78–88. 10.1016/j.foreco.2015.06.010 [DOI] [Google Scholar]
  63. Veenendaal E.M., Swaine M.D., Agyeman V.K., Blay D., Abebrese I.K., Mullins C.E. (1996) Differences in plant and soil water relations in and around a forest gap in West Africa during the dry season may influence seedling establishment and survival. Journal of Ecology, 84, 83–90. [Google Scholar]
  64. Violle C., Navas M.L., Vile D., Kazakou E., Fortunel C., Hummel I., Garnier E. (2007) Let the concept of trait be functional! Oikos, 116, 882–892. [Google Scholar]
  65. Visser M.D., Bruijning M., Wright S.J., Muller‐Landau H.C., Jongejans E., Comita L.S., De Kroon H. (2016) Functional traits as predictors of vital rates across the life cycle of tropical trees. Functional Ecology, 30, 168–180. [Google Scholar]
  66. Watanabe N., Okada T., Tomaru N., Nishimura N., Nakagawa M. (2016) Forest dynamics of stands damaged by Japanese oak wilt in the Kaisho Forest, Aichi prefecture. Journal of the Japanese Forest Society, 98, 273–278 (in Japanese with English summary). [Google Scholar]
  67. Wilson P.J., Thompson K.E.N., Hodgson J.G. (1999) Specific leaf area and leaf dry matter content as alternative predictors of plant strategies. New Phytologist, 143, 155–162. [Google Scholar]
  68. Wright I.J., Cannon K. (2001) Relationships between leaf lifespan and structural defences in a low‐nutrient, sclerophyll flora. Functional Ecology, 15, 351–359. [Google Scholar]
  69. Wright S.J., Kitajima K., Kraft N.J.B., Reich P.B., Wright I.J., Bunker D.E., Condit R., Dalling J.W., Davies S.J., Díaz S., Engelbrecht B.M.J., Harms K.E., Hubbell S.P., Marks C.O., Ruiz‐Jaen M.C., Salvador C.M., Zanne A.E. (2010) Functional traits and the growth–mortality trade‐off in tropical trees. Ecology, 91, 3664–3674. [DOI] [PubMed] [Google Scholar]
  70. Wu C.P., Ando M. (2010) Stand structure and dynamics during a 10‐year period of pine wilt disease forests in Kyoto City. Journal of the Japanese Society of Revegetation Technology, 35, 440–447. [Google Scholar]
  71. Yin J., Lin F., De Lombaerde E., Mao Z., Liu S., Ye J., Fang S., Wang X. (2023) The effects of light, conspecific density and soil fungi on seedling growth of temperate tree species. Forest Ecology and Management, 529, 120683. [Google Scholar]
  72. Zhou L., Thakur M.P., Jia Z., Hong Y., Yang W., An S., Zhou X. (2023) Light effects on seedling growth in simulated forest canopy gaps vary across species from different successional stages. Frontiers in Forests and Global Change, 5, 1088291. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Fig. S1. Heatmap of each soil nutrient concentration in each plot of the Kaisho Forest. The color scale ranges from light blue to dark blue and represents the variation in each nutrient level from lower to higher concentrations. Since the soil nutrient data were collected from each seedling quadrat, the area of the heatmaps for all three plots is standardized to 40 m × 40 m.

PLB-27-1097-s001.jpg (1.5MB, jpg)

Table S1. Mean values of survival rate in each JOW period and mean (±SD) of seedling functional traits for each species in the Kaisho Forest.

Table S2. Mean annual temperature (oC) and annual and growing season precipitation (mm) during 2009 to 2013 and 2018 to 2020.

Table S3. Mean (±SD) of each soil nutrient in each plot of the Kaisho Forest. Soil nutrients were measured using the surface soil. The detailed method for measuring each nutrient was described in the text.

PLB-27-1097-s002.docx (259.1KB, docx)

Data Availability Statement

All archived data (Celegeer et al. 2024) are freely available at: DOI: 10.13140/RG.2.2.14737.40802.

Articles from Plant Biology (Stuttgart, Germany) are provided here courtesy of Wiley

RESOURCES