Fertilisation Effects on Edaphic Properties in Different Soil Types in Relation to Castanopsis sieboldii Sapling Growth

Fertiliser Experiments and Analysis of Nutrient Impacts

The impacts of nutrients and soil type on C. sieboldii growth using 0.05-m² Wagner pots in the Tokyo University ofAgriculture field site (35.642649, 139.630235) were investigated (Figure 1). The outdoor sites were unshaded (open area) and were well ventilated. The growing tree saplings were watered by hand daily with 1 L in the summer, i.e., July to mid-September. The average annual temperature and precipitation at the site are 15.8 °C and 1,598.2 mm, respectively (Japanese Meteorological Agency 2025).

• Download figure
• Open in new tab
• Download PowerPoint

Figure 1.

Experimental pots with humic topsoil (left), humic subsoil (centre), and river sand (right) on 2019 April 16.

The 4 fertilisers used in the tests were N, P, K, and Ca. Based on previous fertiliser tests (Sugiyama et al. 2002), 10 pot treatments were used with the 4 nutrients in 3 replicates: pots unfertilised (control), N pot, P pot, K pot, Ca pot, without N pot (PKCa pot), without P pot (NKCa pot), without K pot (NPCa pot), without Ca pot (NPK pot), and the NPKCa pot.

Firstly, impacts of individual N, P, K, and Ca were determined by calculating the differences in growth values between PKCa, NKCa, NPCa, NPK, and NPKCa to evaluate the individual impacts. Each N, P, K, and Ca addition was considered to have an impact if there was a negative difference between the treatments.

Secondly, the impacts were assessed by calculating the differences in values between unfertilised pots and those with individual N, P, K, and Ca added. Each N, P, K, and Ca pot was considered to have an effect if there were positive differences in growth.

Thirdly, if there were both impacts between PKCa, NKCa, NPCa, NPK, and NPKCa, and between N, P, K, Ca, and unfertilised, the individual N, P, K, and Ca applications were considered to affect C. sieboldii growth. For example, N influenced the number of C. sieboldii leaves when PKCa application growth was less than NPKCa. Growth with N was generally greater than in unfertilised soil.

The compound impact was determined by calculating the difference between PKCa, NKCa, NPCa, NPK, NPKCa, and unfertilised or individual N. If the growth with compound fertiliser was greater than that of the unfertilised or N alone, the compound fertiliser had a positive impact.

Two-way repeated ANOVA (P < 0.05) with elements of soil types, fertilisation, and their interactions were used to confirm statistical differences in all measured values. Tukey’s tests (P < 0.05) were conducted at difference between PKCa, NKCa, NPCa, NPK, and NPKCa, and between unfertilised and individual N, P, K, Ca to determine the individual impacts, and the differences between PKCa, NKCa, NPCa, NPK, NPKCa, and unfertilised/N to gauge compound impacts.

Growth Differences Under Three Soil Types

Growth was measured in unfertilised pots to evaluate the effects of humic topsoil, humic subsoil, and river sand with varying humus content (C-N) on C. sieboldii to mimic soil conditions in the field.

Plant Materials and Experimental Pots

Small plants (90 samples), each approximately 60 cm tall, were used for pot experiments. The weights of both shoots and roots and the lengths of their main roots are presented in Table 1. At planting, the average weight and main root length of the plants were 48.3 g to 51.4 g and 36.8 cm to 44.2 cm, respectively. The experiment began after saplings had been planted and acclimatised for 1 to 2 months to eliminate any influences of plant health or problems with vigour.

The pots used for the experiment were Wagner pots with a surface area of 0.05 m². They were made from white high-impact styrene resin. Each pot measured φ256 (upper diameter) × φ234 (lower diameter) × 297 mm (height). The distance between each other was approximately 30 cm. Gravel (3.1 kg) of 0.5 cm to 1.0 cm diameter was placed in the pot as the base, and 2 kg of river sand was poured onto the gravel in all pots. Additionally, 17.9 kg of river sand, 8.4 kg of humic subsoil, or 9.0 kg of humic topsoil were added to the Wagner pots containing gravel and river sand.

Soil Materials for Experiments

The soils used for the fertiliser experiments were humic topsoil, humic subsoil, and river sand. The humic topsoil was Kanto loam topsoil with humus. The humic subsoil was a Kanto loam subsoil widely found in Eastern Japan (Ohkura et al. 1993). The humic topsoil is referred to as Kurobokudo (Hosono and Sase 2015) and classified as Andosols according to the IUSS (2007). The soil is black with a high level of humus, a high phosphate absorption coefficient, and is present in most soil A horizons. It is referred to as the Kanto loam humic soil layer (allophane). The humic subsoil is below the humic topsoil and is present in most of the brown soil horizons (soil B horizon). Brown forest soils are classified as Cambisols by FAO (1998) and are referred to as Kanto loam subsoils (light soils without humus). In general, humic subsoils have a low base saturation percentage (Hatano et al. 2021). River sand consisted of fine (0.02 mm to 0.2 mm) and coarse (0.2 mm to 2.0 mm) sand without nutrients. Humic topsoil was used because it is the natural substrate in which C. sieboldii occurs. Humic subsoil was used for comparison with humic topsoil, which had the same loam soil but without humus. River sand was used as a control to compare the humic topsoil and subsoil.

The concentrations of NH₄⁺, NO₃^–, PO₄^3–, K⁺, Ca²⁺, C, and N were measured to evaluate the soil chemical properties, and liquid, gaseous, and solid phases. This was used to evaluate the physical properties of soil in the Wagner pots during the experimental period. Two soil samples were collected from the surface of each pot for soil chemical analysis in December 2019 for the river sand and in August 2020 for the humic topsoil and subsoil. The soil taken from the Wagner pots was oven-dried at 105 °C for 18 hr. Thereafter, 12 g of dried soil sample was added to 30-mL ultra-pure water in a 50-mL plastic filtration bottle with a 3.5-cm diameter. Filter paper was placed between the two bottles without a lid. The bottle containing the soil and water was shaken for 3 min and left to stand until 20 mL was filtered. Then, the filtered soil water was poured into 50-mL conical tubes and stored in a refrigerator for 1 day. Soil pH was then measured using a pH meter (HM-60G; DKK-TOA Corporation, Tokyo, Japan). The NH₄⁺, NO₃^–, PO₄^3–, K⁺, and Ca²⁺ were measured to determine the N, P, K, and Ca concentrations using laboratory analysers and an ion analyser (IA-300; DKK-TOA Corporation, Tokyo, Japan). C/N was measured using a Vario MAX CN Macro Elemental Analyser (Elementar Analysensysteme GmbH, Langenselbold, Germany). The ratios of the liquid, gaseous, and solid phases were measured using a digital voluminometer (DIK-1150; Daiki Rika Kogyo Co., Ltd, Hongsu City, Japan).

As shown in Table 2, both the humic topsoil and subsoil were weakly acidic, whereas the pH of river sand was nearly neutral. The humic topsoil had high NO₃^– and NH₄⁺. PO₄^3– and K⁺ concentrations were lower in the humic topsoil than in the humic subsoil and river sand. The lowest Ca²⁺ concentration was found in the river sand, and the highest was found in the humic topsoil. The highest C and N contents were found in the humic topsoil, followed by the humic subsoil. River sand contained only small amounts of C and N. Regarding soil physical characteristics, both humic topsoil and subsoil showed clay and clay loam properties, and river sand contained 85% to 100% sand. The solids ratio of the river sand was twice that of the humic topsoil and subsoil. The gaseous ratio was highest in river sand, followed by humic subsoil and humic topsoil. The volumetric water content (liquid ratio) was approximately 50% for humic topsoil and subsoil but was only 1.4% for river sand.

Fertilisers for Experiments

The fertilisers used for these experiments were urea (NH₂)₂CO, which contains 46% N; calcium superphosphate (Ca[H₂PO₄]₂), which contains 17.5% watersoluble phosphoric acid; potassium chloride (KCl), which contains 60% water-soluble potassium; and slaked lime (Ca[OH]₂), which contains 70% CaO equivalent. Each fertiliser element was supplied at 10 g/m² under pure-element conditions: N, 1.09 g; P₂O₅, 3.45 g; K₂O, 0.83 g; and Ca(OH)₂, 0.71 g/year. These were slow-release, granule-type, uncoated fertilisers. The granules were ground into powder and mixed thoroughly with the soil in each pot before application.

Fertiliser was applied 6 times (August 2017, July 2018, and January, April, July, and October 2019) in river sand pots, and 3 times (August 2018, February 2020, and November 2020) in humic subsoil and topsoil (Table 3).

Experimental Periods and Place

As shown in Table 3, the saplings in river sand were planted 2017 June 30 after measuring the plant weight and main root length. The experiment concluded 2022 September 21 (5 years and 3 months later) following measurements of shoot and root weights. The experimental materials in humic topsoil and humic subsoil were planted 2018 June 17 after measuring the plant weight and main root length, and concluded 2022 November 20 (4 years and 5 months later) following measurements of shoot and root weights. The pots were placed at a field site (35.642824, 139.630212) at the Tokyo University of Agriculture, Setagaya Campus, Sakuragaoka, Setagaya-ku, Tokyo, Japan. The pots were arranged in rows for each of the humic topsoil, subsoil, and river sand, but to eliminate any influence on growth because of positioning, the rows and the locations of individual pots within the rows were rotated once a month.

Evaluation of Growth

Eight attributes were measured to compare the C. sieboldii growth under different soil conditions: (1) dry weights of shoots and/or roots; (2) number of leaves; (3) tree height; (4) main trunk diameter close to the ground (MTD); (5) soil and plant analyser development (SPAD) value; (6) length of the first branch; (7) number of first branches; and (8) diameter at the base of the first branch (DBF).

1. The dry weights of shoots and roots were measured after drying using a hot air drier at 105 °C for 18 hr. This was after completion of the experiment on 2022 November 20 in humic topsoil and humic subsoil and 2022 September 21 in river sand.

2. The leaves on all first branches that grew from the main stem were counted as the number of leaves on all branches of the plant in each pot.

3. Tree height was measured from the base to the top of the trunk using a measuring tape.

4. The MTD was measured using an electronic micrometre calliper.

5. The SPAD value, i.e., the difference between the transmittance of red (650 nm) and infrared (940 nm) light through the leaf, was measured using SPAD-502 Plus software (Konica Minolta, Inc., Tokyo, Japan).

6. The length from the base to the tip of the first branch was measured using a measuring tape.

7. The numbers of living first branches were counted.

8. The DBF was measured using an electronic micrometre calliper.

All data were collected at 2- or 6-month intervals and used to calculate averages and differences between data at the first and each collection date, except for the SPAD value.

Two-Way Repeated ANOVA for the Effects of Soil Types and Fertilisation on Castanopsis sieboldii Growth

There were statistically significant differences shown for soil types, fertilisation, and interactions between them on dry weight of shoots and roots, dry weight of roots, number of leaves, MTD, SPAD, DBF, and length of first branch. Significant differences were demonstrated for fertilisation, and interaction between dry weight of shoots, soil types, and fertilisation in tree height, and fertilisation at number of first branches were detected (Table 4).

Individual or Compounds of N, P, K, and Ca on Castanopsis sieboldii Growth

Comparison of Castanopsis sieboldii Growth Under Unfertilised Conditions

Dry Weight (T/R Ratio)

The total dry weights of shoots and roots in the humic topsoil, humic subsoil, and river sand were 86.9 g, 53.7 g, and 63.3 g, respectively, under unfertilised conditions (Figure 2), with no significant differences. Dry weight of shoots was the highest in the humic topsoil, followed by river sand, then humic subsoil. The dry weight of roots was lower than for shoots in river sand (T/R ratio = 1.24). Conversely, the dry weight of the shoots was lower than roots in humic subsoil (T/R ratio 0.59) or topsoil (T/R ratio 0.75).

• Download figure
• Open in new tab
• Download PowerPoint

Figure 2.

Shoot to root ratio under unfertilised conditions in humic topsoil, humic subsoil, and river sand. Letters represent statistical differences at P < 0.05 based on the t-test in shoots. There are statistical differences between different alphabets.

Number of Leaves

The number of leaves under unfertilised conditions increased beyond 100 in humic topsoil and 50 in both humic subsoil and river sand (Figure 3). The highest number of leaves was found in humic topsoil, followed by humic subsoil, and finally river sand. The number of leaves in the humic subsoil or river sand was significantly lower than for humic topsoil (P < 0.05, at 2, 4, 9, 12, and 14 months).

• Download figure
• Open in new tab
• Download PowerPoint

Figure 3.

Number of leaves under unfertilised conditions in humic topsoil (line), humic subsoil (line with white circle marker), and river sand (dotted line). Line arrows indicate the fertilisation date for the river sand. The dotted line arrows indicate the fertilisation date for both humic topsoil and subsoil. Letters represent statistical differences at *P < 0.05 and **P < 0.01 among R (river sand), S (humic subsoil), and T (humic topsoil) based on the t-test. Vertical lines indicate standard deviation.

Tree Height

The greatest tree height was observed in humic topsoil, followed by humic subsoil and river sand at 2, 4, 8, 12, and 14 months (P < 0.05) after the experiments began (Figure 4).

• Download figure
• Open in new tab
• Download PowerPoint

Figure 4.

Tree height under unfertilised conditions in humic topsoil (line), humic subsoil (line with white circle marker), and river sand (dotted line). Line arrows indicate the fertilisation date for the river sand. The dotted line arrows indicate the fertilisation date for both humic topsoil and subsoil. Letters represent statistical differences at *P < 0.05 and **P < 0.01 among R (river sand), S (humic subsoil), and T (humic topsoil) based on the t-test. Vertical lines indicate standard deviation.

Main Trunk Diameter Close to the Ground (MTD)

The largest MTD was in the humic topsoil, followed by those in the humic subsoil and river sand. Significant differences were observed after 8 months of the experiment: humic subsoil < humic topsoil (P < 0.05) (Figure 5).

• Download figure
• Open in new tab
• Download PowerPoint

Figure 5.

Main trunk diameter close to the ground (MTD) under unfertilised conditions in humic topsoil (line), humic subsoil (line with white circle marker), and river sand (dotted line). Line arrows indicate the fertilisation date for the river sand. The dotted line arrows indicate the fertilisation date for both humic topsoil and subsoil. Letters represent statistical differences at *P < 0.05 and **P < 0.01 among R (river sand), S (humic subsoil), and T (humic topsoil) based on the t-test. Vertical lines indicate standard deviation.

Soil and Plant Analyser Development (SPAD) Value

The average SPAD value was 26.0, 24.0, and 22.0 in humic topsoil, humic subsoil, and river sand, respectively: river sand – humic subsoil; river sand < humic topsoil (significance at P < 0.05). However, no differences were observed between humic topsoil and humic subsoil (Figure 6).

• Download figure
• Open in new tab
• Download PowerPoint

Figure 6.

SPAD value under unfertilised conditions in humic topsoil (line), humic subsoil (line with white circle marker), and river sand (dotted line). Line arrows indicate the fertilisation date for the river sand. The dotted line arrows indicate the fertilisation date for both humic topsoil and subsoil. Letters represent statistical differences at * P < 0.05 and ** P < 0.01 among R (river sand), S (humic subsoil), and T (humic topsoil) based on the t-test. Vertical lines indicate standard deviation.

Length of First Branch

The length of the first branch showed significant differences (P < 0.01) in humic topsoil > river sand and humic topsoil > humic subsoil at 4, 9, 14, 26, 33, and 40 months after the start of the experiment. However, there was no significant difference between humic subsoil and river sand (Figure 7).

• Download figure
• Open in new tab
• Download PowerPoint

Figure 7.

Length of first branch under unfertilised conditions in humic topsoil (line), humic subsoil (line with white circle marker), and river sand (dotted line). Line arrows indicate the fertilisation date for the river sand. The dotted line arrows indicate the fertilisation date for both humic topsoil and subsoil. Letters represent statistical differences at *P < 0.05 and **P < 0.01 among R (river sand), S (humic subsoil), and T (humic topsoil) based on the t-test. Vertical lines indicate standard deviation.

Number of First Branches and Diameter at the Base of the First Branch (DBF)

There were no significant differences in the DBF or number of first branches. However, the number of first branches was in the order humic topsoil > humic subsoil > river sand. The DBF ranked as follows: humic topsoil > river sand > humic subsoil (Figure 8 and Figure 9).

• Download figure
• Open in new tab
• Download PowerPoint

Figure 8.

Number of first branches under unfertilised conditions in humic topsoil (line), humic subsoil (line with white circle marker), and river sand (dotted line). Line arrows indicate the fertilisation date for the river sand. The dotted line arrows indicate the fertilisation date for both humic topsoil and subsoil.

• Download figure
• Open in new tab
• Download PowerPoint

Figure 9.

DBF under unfertilised conditions in humic topsoil (line), humic subsoil (line with white circle marker), and river sand (dotted line). Line arrows indicate the fertilisation date for the river sand. The dotted line arrows indicate the fertilisation date for both humic topsoil and subsoil.

Leaf Length and Width

Leaf length and width was greater in humic topsoil than in humic subsoil and river sand. Length × width of leaf and leaf length were greater in humic topsoil than in humic subsoil. Leaf width was greater in humic topsoil than in humic subsoil or river sand (P < 0.05) (Figure 10).

• Download figure
• Open in new tab
• Download PowerPoint

Figure 10.

Leaf length and width under unfertilised conditions in humic topsoil (black triangle), humic subsoil (white circle), and river sand (cross). Values represent the average of 10 leaves. There are significant differences (P < 0.05) between humic topsoil > humic subsoil in leaf length and length × width of leaf; and humic topsoil > humic subsoil and river sand in leaf width. Letters represent statistical differences at *P < 0.05 and **P < 0.01 among R (river sand), S (humic subsoil), and T (humic topsoil) based on the t-test.

Individual Impact of N in Humic Topsoil, Humic Subsoil, and River Sand

As shown in Table 6, individual P, K, and Ca applications had no impact in humic topsoil, humic subsoil, or river sand. Individual N addition affected C. sieboldii growth in humic topsoil with impacts on dry weight of shoots and roots, dry weight of shoots, and SPAD. Individual N had impacts in humic subsoil on dry weight of shoots and roots, dry weight of shoots, number of leaves, tree height, MTD, SPAD, and DBF. In river sand, individual N had no impacts on any measurement features.

Compound Nutrients Including N in Humic Topsoil, Humic Subsoil, and River Sand

As shown in Table 6, compound nutrients in humic topsoil had impacts on all measurements, except tree height, number of first branches, and length of first branch in NKCa and NPCa, on tree height and number of first branches for NPK, and for dry weight of roots and tree height with NPKCa. In humic subsoil, compound nutrients impacted on all measurements, except for tree height and number of first branches on NKCa, number of first branches and DBF for NPCa, number of first branches for NPK, and dry weight of roots for NPKCa. In river sand, compound nutrients had impacts, including on the number of leaves with NKCa, NPCa, NPK, and NPKCa, dry weight of shoots and roots, dry weight of shoots for NPCa and NPK, dry weight of roots for NPK, and SPAD with NKCa.

Comparison of the Effect Between Individual N and Compound Nutrients, Including N

Individual P, K, and Ca had no impacts in humic topsoil, humic subsoil, or river sand. However, adding P, K, and/or Ca with N increased the values of shoots and/or roots, and number of leaves and SPAD compared to unfertilised, low-nutrient river sand (Table 2). No impact was observed in river sand with N fertilisation alone, but combined fertilisation did affect growth. On the other hand, N alone had impacts in humic topsoil and subsoil, perhaps because they both contain more PO₄^3– and Ca²⁺ than river sand (Table 2). With N applied alone to humic topsoil and subsoil (relatively high in P and Ca), the impact on growth was similar to combined fertilisation.

NPCa and NPK gave higher values than individual N in dry weight of shoots and roots and dry weight of shoots in river sand. Moreover, NPKCa resulted in higher DBF than individual N in humic topsoil. NPK showed a higher value for the number of leaves than individual N in humic subsoil.

Nitrogen, being the most important macronutrient for crops with individual impacts usually high (Nie et al. 2009), was found to influence C. sieboldii growth. Compound fertilisers containing N had greater effect than single N fertilisers, and this was particularly true in the river sand treatments, probably because of their inherent low levels of available P and Ca. Compound nutrient impacts were not necessarily greater than those of individual fertiliser applications but occasionally produced higher impacts (Armson 1977; Hatano 1991).

Difference in the Growth Between Humic Topsoil, Humic Subsoil, and River Sand

The results showed that C. sieboldii growth differed between humic topsoil and subsoil. The highest dry weight was observed in humic topsoil, followed by humic subsoil, and river sand. The T/R ratio especially was highest in river sand, followed by humic topsoil and subsoil. This means the amount of roots relative to shoots in river sand was less than in humic topsoil and subsoil. The highest number of leaves, SPAD value, tree height, leaf length and width, DBF, MTD, and length of the first branch were all found in humic topsoil. This was followed by humic subsoil and then, lastly, river sand. Overall, C. sieboldii, with unfertilised conditions, showed highest growth in humic topsoil, followed by humic subsoil and river sand. The poor growth in river sand was due to the low substrate concentrations of available NO₃^–, NH₄⁺, PO₄^3–, and Ca²⁺. The main reasons for the difference between the humic topsoil and subsoil were the C and N concentrations. The C/N ratios measured in humic topsoil and humic subsoil were 13.1 (C: 23.771, N: 1.816) and 11.6 (C: 4.874, N: 0.42), respectively (Table 2). N, C, and P in the humic topsoil were higher, and K was lower than those in the humic subsoil and river sand.

The importance of high C and N content in topsoil is clear from the literature. Humic topsoil is an andoloil that has high humic content and phosphate absorption (Adachi 1973; Hamasaki 2007). The humic content differed between topsoil, subsoil, and river sand (Yamane et al. 1978), and humic acid contains large amounts of N (Berg and McClaugherty 1989). Oak forests, which consist mainly of Castanopsis cuspidate, maintained C/N = 20 as litter decomposition progressed (Chairul and Yoneda 2002). The C/N ratio increased to 38.1 and 24.7 as the fallen leaves of Chamaecyparis obtuse and Betula maximowicziana increased after one year (Kawada 2000). The soil physical properties were demonstrated by the solidphase ratio: 38.9% to 45.6%; liquid-phase ratio: 47.1% to 47.2%; and gaseous-phase ratio: 7.3% to 14.0%. The soil chemical properties included C: 147.2 g/kg to 162.1 g/kg; N: 11.7 g/kg to 12.7 g/kg; C/N: 12.5 to 13.9; and pH (H₂O): 5.0 to 6.5 in the C. sieboldii forest at the Institute for Nature Study in Tokyo (Miyajima et al. 2016).

Kimura and Kunimura (1961) reported that kaki (persimmon, Diospyros kaki Thunb.) growth was high in loam (i.e., humic topsoil) but less so in clay or fine-gravel soils. In addition, organic material (oak leaf litter) promoted kaki growth, and similar results were found for C. sieboldii.

Impact of Leaf Litter on Castanopsis sieboldii Growth Under Field Conditions

This study demonstrated that N and compound nutrients including N affected C. sieboldii growth. It is therefore informative to consider the importance of litter in the relevant nutrient cycling. Under field soil conditions, the main source of N is leaf litter, which contains low concentrations of P, K, and Ca, but which is high in C. The fallen leaves of Quercus serrata, which belongs to the same family as C. sieboldii, contained N: 0.5% to 1.0%; P: 0.06%; K: 0.2% to 0.6%; and Ca: 1.0% to 1.5%. Microelements such as Mg (magnesium), Zn (zinc), Mn (manganese), and Fe (iron) are also present in humus, along with K and Ca (Armson 1977; Berg and McClaugherty 1989). Strojan (1978) measured the concentrations of Ca, K, Mg, N, P, S (sulfur), Cd (cadmium), Cr (chromium), Cu (copper), Fe, Mn, Na (sodium), Ni (nickel), Pb (lead), and Zn in leaf litter. McClaugherty et al. (1985) reported that mineralisation rates correlate with decomposition rates of native dominant foliage litter. According to Kawada (2000), the chemical components of humus were N = 1.56%, 1.83%, and 2.16%; P = 0.045%, 0.054%, and 0.068%; K = 0.73%, 0.18%, and 0.13%; and Ca = 0.67%, 0.76%, and 0.77% in fresh leaves, after 5 months, and after 1 year, respectively.

The growth of C. sieboldii under different humic contents and C-N ratios in the soil was compared, and whilst humic topsoil and subsoil ratios can be similar, humic content varies with the amount of leaf litter. However, these experiments were also performed with river sand as a control with contrasting soil physical properties and low nutrient content. Watanabe and Maruta (1992) reported that at sites where litter was artificially removed, surface soil had less total C and N than otherwise and was harder. Consequently, tree growth was reduced. Soil hardness, porosity, total C, total N, EC (electric conductivity), and exchangeable bases (Ca, Mg, and K) resulted in major differences between managed and abandoned secondary forests (Tsuji and Hoshino 1992).

Based on these results, humus containing compound nutrients (primarily those based on N, P, K, and Ca, and micronutrients) improves C. sieboldii growth. Plant leaves and branches produce litter, from which humus and soil organic matter develop and release nutrients (Cui et al. 2022). A major component of humus is N (Dincher et al. 2020; Matkala et al. 2020), and in a natural soil, most available plant N is derived from soil organic matter (Strong and Mason 1999). Soil N mineralisation is positively correlated with litter production, and N is returned to the soil via the litter. Soil N mineralisation is negatively correlated with litter C:N and C:P ratios (Manzoni et al. 2010). It is argued that N derived from the humus promotes C. sieboldii growth.

However, soil nutrients, especially available N and P, are low in many urban soils and particularly those under street trees (Schaffert et al. 2022), with low N and C levels also reported (e.g., Zhu et al. 2017; Salomon et al. 2020). Future investigations should therefore address effects of humus content and leaf litter on C. sieboldii growth.