Abstract
Background Urban areas, many located in lowlands, are often characterised by poor drainage and compacted soil and this can have an impact on the growth of urban trees. Castanopsis sieboldii and Cinnamomum camphora are 2 evergreen tree species distributed throughout southern Japan. Many of these trees are planted in urban areas and their suburbs, and others grow naturally. The differences between these 2 species in their response to soil physical conditions, particularly those towards waterlogging and low aeration conditions, have not previously been studied.
Methods In this study, we analysed and compared the tolerance of Castanopsis spp. and Cinnamomum spp. to stagnant soil water conditions and investigated the effects of soil quality improvement using obsidian perlite to increase air content in soil under stagnant and drained water conditions on the growth of these plants in pot experiments. In the study, Castanopsis and Cinnamomum plants were grown on sand alone or mixed with obsidian perlite, and these were subjected to stagnant or drained water treatments.
Results Under stagnant water conditions, all young Castanopsis trees died within approximately 18 months. In contrast, only one young tree of Cinnamomum spp. grown on sand died under stagnant water conditions, and those grown on sand mixed with obsidian perlite were still alive after 35 months under stagnant conditions. Soil quality improvement using perlite increased the leaf abundance (numbers) in both species.
Conclusions Stagnant water with a reduced soil air phase adversely affected Castanopsis spp. growth, but over the study period of 3 years Cinnamomum spp. was unaffected. When selecting tree species for urban planting, the impact of localised soil conditions on tree physiology may be critical.
Introduction
Cinnamomum camphora (L.) J. Presl (Lauraceae) and Castanopsis sieboldii (Makino) Hatusima ex Yamazaki et Mashiba (Fagaceae) are evergreen trees distributed throughout southern Japan from the centre of the mainland to Shikoku and Kyusyu (Momiyama 1997; Ohba 1997). However, it is not known whether C. camphora is native or introduced. In some regions, C. sieboldii and C. camphora trees are worshipped as deities, and as such, are critical components of urban forests. Historically, the so-called giant C. sieboldii and C. camphora trees have been conserved because of their age, size, and shape. Giant C. sieboldii and C. camphora trees are found in urban and suburban areas of Tokyo presumably conserved for spiritual reasons (Xu et al. 2018).
Castanopsis spp. have localised distributions from sea level to 900 m above sea level (Kobayashi 2008), with many C. cuspidata individuals on lower slopes or in valleys, but sparse at high altitudes (Kobayashi and Hiroki 2003). In contrast, C. sieboldii occurs mainly at higher altitudes, such as on hilltops and ridges. According to Ueno et al. (2009), the distinct distribution patterns of Castanopsis cuspidata and Castanopsis sieboldii may be linked to their water requirements. Watanabe et al. (2021) reported the presence of giant Castanopsis spp. trees on slopes, along the edges of plateaux, and on mountaintops. However, it remains unclear why Castanopsis spp. are common on slopes or the tops of mountains, hills, and plateaux but absent elsewhere. In southern China and Japan, most Cinnamomum spp. are found in low-lying areas (< 400 m) (Suzuki et al. 2016).
Light, temperature, and soil conditions (physical, chemical, and biological) are major drivers of plant growth, and the association between topography and tree growth is known empirically (Kusaka 1965). Field observations and literature suggest that soil conditions are linked to site topography and may affect water depth and longevity of flooding in some locations (Ohta et al. 1985; Montgomery et al. 1997). One further influence may be, for example, the presence of an efficient drainage system (perhaps a pipe network) in the surface materials close to watercourses (Ohta et al. 1985). Floods affect the growth and survival of trees in affected areas because flooding and stagnant water conditions may damage plants (Rotherham 2015), and this potentially affects low-lying urban forests especially, because of reduced soil water drainage and waterlogging on flatlands. Flooding dynamically changes the soil status of a site and whilst the association between soil water drainage and tree growth has been considered (McAlhaney et al. 2020), the impact of stagnant water remains unclear (Rotherham 2015; McAlhaney et al. 2020). However, it is clear that anaerobic conditions may be caused by water stagnation during inundation (Kozlowski 1985). Waterlogging can substantially reduce transpiration and the growth of fine roots in affected trees (Fujita et al. 2021). We predicted that comparing effects of waterlogging on test plants of C. camphora trees (common in lowland areas) and C. sieboldii (rarely found in lowlands but common on upper zones, slopes, plateaux, small mountains, and hills) would show differences in physiological and ecological responses.
In addition, despite conservation efforts, particularly in urban areas, the deterioration in the quality of surrounding habitats and especially damage caused by construction, have resulted in the decline or death of giant trees. Previous studies (Marubashi et al. 2019; Watanabe et al. 2021) reported that many giant C. sieboldii trees were damaged. This was based on assessments of plant height, branch length, trunk circumference, and the trunk angle to the ground evaluated as an ‘Indicator of Tree Form’. Surveys across the study area confirm that damaged giant C. sieboldii trees are common in low-lying urban and suburban areas. Marubashi et al. (2019) predicted that pH or mountain tops, which tend to have high potassium content in the soil, may have some effect on the growth of large C. sieboldii trees. Watanabe et al. (2021) also thought that specific soil conditions (nutrient levels, soil hardness, and drainage) might strongly affect the ability of C. sieboldii to grow in lowlands. This observation was because when comparing soil conditions across slopes, plateaux, upper plateau edges, mountaintops, and lowlands, the main differences appear related to the physical properties of soils.
MacDonald et al. (1993) reported interactions between plant growth and soil physical phases. Specifically, they examined the soil conditions around healthy and declining oak trees ( Quercus agrifolia) in urban California and revealed that aeration in the upper soil was critical to tree vigour. Another study (Neira et al. 2015) showed that oxygen diffusion rate better indicates soil oxygen availability than oxygen concentration. Moreover, the diffusion rate is lowest in soils with the highest bulk density and high water content. They also found that in poorly aerated soils, oxygen diffusion rate does not increase following air injection, the latter provided in order to loosen the soil structure. Kämäräinen et al. (2018) compared the soil air composition of conventional and structural tree soils (consisting of natural soil and constructional rubble, respectively) and as used for roadside plantings along residential streets in Helsinki (Finland). They analysed the average oxygen concentration and found it was higher in structural soil than in conventional soil.
The association between soil water conditions and plant growth has been extensively studied regarding revegetation in climatically dry regions (Masuda et al. 1983; Masuda et al. 1991). Yoshikawa et al. (1996) assessed the effects of desiccation on C. camphora and Quercus glauca, whereas Yoshikawa and Wang (1991) evaluated its impacts on Pinus densiflora, Q. glauca, Zelkova serrata, Caragana korshinskii, and Hedysarum scoparium. However, from the review it seems that little is known regarding the interactions between gaseous soil phase conditions and plant growth in urban forest scenarios (Cambi et al. 2018). One example of a study on the growth of Japanese trees under soil compaction conditions showed that aboveground growth of Ginko biloba and Camellia japonica var. hortensis f. otome trees was poor in compacted soil with a density of 20 kg/cm3 (Kitamura and Nodazaka 1974). Directly relevant to this study, the growth of C. camphora is associated closely with soil–air phase ratio and permeability, and root development is assumed difficult at a soil–air phase ratio of < 20% and permeability < 0.01 mm/s (Masuda et al. 1991).
With these considerations in mind, we hypothesized that the different distributions of C. sieboldii and C. camphora, and the reason that many giant C. sieboldii trees in urban areas were damaged, are associated with soil physical conditions. In particular, we suggest the soil/air properties, which vary between high zones and low-lying areas, are particularly significant. In this study, we analysed and compared the tolerance of C. sieboldii and C. camphora plants to stagnant soil water conditions and the effects of soil quality improvement using obsidian perlite. The latter treatment was designed to increase air content in dense urban soils under both stagnant and drained water conditions on the growth and survival of these plans using pot experiments.
Materials and Methods
Materials
A large tree, such as an adult roadside specimen, may have the eco-physiological capability to cope with stagnant soil water conditions (e.g., owing to the increased scale of the root and canopy networks). However, experimentation using large trees requires substantial time, labour, and space, making such work difficult and necessarily long-term. With these constraints in mind, young trees of C. sieboldii and C. camphora of similar sizes and growth were selected for the experiments. The seedlings of these plants were cultivated at a nursery in Oita prefecture, Kyusyu, southern Japan. The average number of leaves, height, number of first branches, diameter at the base of the trunk, and length of the main root of the young trees are shown in Table 1. All samples were planted on the 2017 May 17.
Experimental pots and materials. The beginning day for counting data was 2017 July 19.
Methods
Pots equipped with drainage ducts were used to assess the effects of stagnant soil water on plant growth. Small 0.05 m2 Wagner pots (φ256 × φ234 × 297 mm, ground surface area 500 cm2) were used, to provide consistent soil water conditions. The pots were placed in a field site (35.642824° N, 139.630212° E) at the Tokyo University of Agriculture, Setagaya Campus, Sakuragaoka, Setagaya-ku, Tokyo, Japan.
To examine the effects of drainage or stagnant soil water conditions and low or high soil aeration on the growth and survival of C. sieboldii and C. camphora, they were subjected to the following treatments: (a) pots with drainage, filled with sand (Dr-Sa); (b) pots with stagnant water, filled with sand (St-Sa); (c) pots with drainage, filled with sand and perlite (Dr-P); and (d) pots with stagnant water, filled with sand and perlite (St-P). The drainage ducts were plugged using rubber plugs to create pots with stagnant water. The drained and stagnant soil water conditions were denoted as Dr and St, respectively. Sand alone and sand mixed with obsidian perlite were denoted as Sa and P, respectively. Three replicates of the experiment were set up and evaluated with a 2-way analysis of variance (ANOVA).
Gravel (4 kg) was then placed at the bottom of each plastic pot. The pots labelled Dr-Sa and St-Sa were filled with sand (16 kg) collected from a river, whereas those labelled Dr-P and St-P were filled with sand (14 kg) mixed with soil quality improvement material (0.25 kg obsidian perlite), and the ratio of sand to obsidian perlite was 6:4 (v/v). The grain size of the obsidian perlite, Whitloam-TC, manufactured by Toho-Leo Co., Osaka, Japan, used in the experiment ranged from 4 mm to 25 mm. Perlite is inorganic and produced by burning and firing obsidian at temperatures > 1,000 °C. Moreover, it is extremely light and can maintain gaseous air in a solid mixture. The solid: air: liquid proportions of the mixture were estimated as follows: (1) For the sand treatments, these were by volume of solid material to air and liquid: 55.6%, 44.0%, and 0.4%. (2) For perlite treatments, these were 1.4%, 92%, and 5.6%, solid to air and liquid. These volumes were measured using a digital volumometer (DIK-1150; Daiki Rika Kogyo Co. Ltd., Saitama Prefecture, Japan). Drainage was stopped for stagnant water condition on 2018 May 15. Fertiliser was applied at a ratio of N: P2O5: K2O = 8:8:8 using 5 g/m2 pure N at 2-month intervals from 2017 July 19, 2018 November 28, and 2019 November 12 to provide nutrient supplements to sand mixes. Water was obtained from natural rainfall.
In a growth room at around 20 °C, water contents were measured in pots of sand or sand mixed with obsidian perlite after drying. Sand or sand with obsidian perlite of 2,100 g in total, 6:4 by volume were in the plastic pots of 127 mm diameter and 97 mm height. Sampling was conducted between the 2022 April 1 and 2022 July 31.
During the experiments, the Wagner pots in the field were installed with a WaterScout SM100 soil moisture sensor (Spectrum Technologies Ltd, Bridgend, Wales, UK) and Watch Dog 1400 data logger (ESPEC MIC Corp., Plainfield, Illinois, USA). These instruments measured the number of days during which the plants were under stagnant soil water conditions.
The following 9 attributes were measured to compare C. sieboldii and C. camphora growth under different soil conditions: (1) number of dead plants, (2) dry weight of shoots and roots, (3) number of leaves, (4) soil plant analysis development (SPAD) value, (5) plant height, (6) diameter at the base of the trunk, (7) number of first branches, (8) length of the first branch, and (9) diameter at the base of the first branch. Dead plants were identified by determining trunk colour and presence or absence of leaves. Plants whose leaves had all turned brown and dropped and whose trunks and branches had dried or turned brown were considered dead. The dry weights of shoots and roots were measured after drying (using a hot air drier at 105 °C for 18 h). This was on completion of the experimental treatment after 42 months (2022 September 21) following the start of the experimental treatments. The weights of the dead plants were not recorded during the experiments, as it proved difficult to ascertain when a young tree was fully dead or might continue to grow. The number of leaves on all first branches that grew from the main stem was counted, with the number of leaves on all first branches of a plant summed for each pot. 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 the SPAD-502Plus software (Konica Minolta, Inc., Chiyoda City, Tokyo, Japan) to assess leaf growth. Plant height was measured from the base to the top of the trunk with tape. The diameter of the main trunk base was measured using electronic micrometre callipers. The length from the base to the tip of the first branch was measured using tape, and the base diameter of the first branch was measured using electronic micrometre calliper. In addition, the number of live and dead first branches were counted.
Data Analysis
From 2017 July 19 to 2021 April 27, data were collected at 2-month intervals in the first 2 years, 3-month intervals in the 3rd year, and 6-month intervals in the 4th year. The calculated data were compared with the initial data (on 2018 May 15), and the differences between the starting and collected data for each day were plotted, except for the SPAD values. The SPAD value presented is displayed directly on the machine display data. SPAD data were obtained from 10 leaves per pot and averaged. Statistical analyses were performed using Tukey’s least significant difference test and 2-way ANOVA with 3 replicates using a Bell curve in Microsoft Excel version 3.21 (Redmond, Washington, USA).
Results
Differences in Moisture Content Between the Sand and Sand with Obsidian Perlite Pots
Drainage rates were almost the same for the sand and sand with obsidian perlite pots. After drying in an oven at 105 °C for 2 hrs, the water loss from the sand pots was substantially higher than that from the sand with obsidian perlite pots. However, the water content in the sand and sand with obsidian perlite pots was almost the same at 12.5 hrs after drying. The difference in water content between the sand and sand with obsidian perlite pots was approximately 4% at room temperature, 20 °C (Figure 1).
Water content in pots containing sand alone or sand mixed with obsidian perlite after drying at room temperature, 20 °C. Sand alone or sand mixed with obsidian perlite, with a total amount of 2,100 g, were filled in plastic pots having a diameter of 127 mm and a height of 97 mm. The tests were conducted from 2022 April to July.
Water Contents in St-Sa and St-P Wagner Pots
Water content in the pots subjected to St conditions remained high after drainage was stopped on 2018 May 15, mainly from August to November. However, from 2018 November to 2019 February, the water content decreased. The increase in water content was higher in the St-Sa pots than in the St-P pots from 2018 June to November (Figure 2).
Water content in each pot in the field filled with sand alone or sand mixed with obsidian perlite under drained or stagnant soil water conditions. Two WaterScout SM100 soil moisture sensors were placed in each pot. The final values were the average of the soil moisture values obtained with the 2 sensors.
Number of Living Plants
Castanopsis sieboldii and Cinnamomum camphora seedlings were planted in 4 types of experimental pots, Dr-Sa, St-Sa, Dr-P and St-P, and their growth parameters varied (Figure 3). Regarding the effects of drainage (Dr) and stagnation (St) on C. sieboldii plants grown on sand (Sa) or sand mixed with obsidian perlite (P), the plants in the St-Sa pots began to die in 2019 February (Figure 4A), 9 months after drainage was stopped, and all C. sieboldii plants in these pots died by 2019 November (18 months after stopping drainage). One C. sieboldii plant in a St-P pot was dead in 2019 August (15 months after stopping drainage), but no C. sieboldii plants in the Dr-Sa pots died until 2020 October (29 months after stopping drainage). Moreover, one C. sieboldii plant in a Dr-P pot was dead in 2021 April (35 months after stopping drainage). These results indicate that both Dr-Sa and Dr-P with drainage conditions survived longer than stopping drainage condition for the C. sieboldii. Additionally, C. sieboldii plants grown on sand mixed with obsidian perlite (P) under both drainage (Dr) and stagnant (St) soil water conditions survived longer than those grown on sand (Sa) under similar conditions, as one plant in a Dr-Sa pot died earlier than those in the Dr-P pots and all plants in the St-Sa pots died earlier than those in the St-P pots.
Experimental pots on 2018 December 3.
Number of dead experimental plants (pots) on the (A) Castanopsis sieboldii and (B) Cinnamomum camphora. Vertical dotted line shows the time that drainage was stopped (2018 May 15).
In C. camphora (Figure 4B), plants in the Dr-Sa and St-Sa pots died by 2018 September (4 months after stopping drainage). One plant in the St-Sa pot survived until 2021 April (35 months after the discontinuation of drainage and until the end of the experiment). There were no dead plants in the Dr-P or St-P pots during the 35-month experiment, and stagnant soil water conditions were not fatal for C. camphora plants in these pots. Thus, soil quality improvement caused by mixing obsidian perlite with sand strongly and favourably affected C. camphora growth and survival.
However, a higher rate of increase in water content in the St-Sa pots than in the St-P pots from 2018 June to November did not impact the growth of either C. sieboldii or C. camphora species. No C. sieboldii plants died, with one C. camphora plant dead in the Dr-Sa and St-Sa pots each.
Dry Weights of Plants
The dry weights of C. sieboldii plants in the St-Sa pots were not measured because all C. sieboldii plants in these pots died by 2019 November. C. sieboldii plants in the Dr-P pots had higher dry weights than measured for plants grown in the Dr-Sa and St-P pots (Figure 5A). However, in the case of C. camphora plants, the dry weights of plants in the Dr-Sa and Dr-P pots were higher than those of plants in the St-Sa and St-P pots, respectively. The dry weights of C. camphora plants in the Dr-P and St-P pots were higher than those of C. camphora plants grown in the Dr-Sa and St-Sa pots, respectively (Figure 5B). These results show that soil quality improvement caused by mixing obsidian perlite with sand increased the growth of both Castanopsis and Cinnamomum species. Stagnant conditions damaged both C. camphora and particularly C. sieboldii plants.
Dry weight of shoots and roots at end of experiment on the (A) Castanopsis sieboldii and (B) Cinnamomum camphora (2022 September 21).
Number of Leaves
Regarding the effects of drainage (Dr) or stagnation (St) on C. sieboldii growth, stagnant water conditions and soil quality improvement affected the number of leaves (Figure 6A). The change in the number of leaves was almost the same in all pots until the drainage ducts were blocked in 2018 May (8 months after the start of the experiments in 2017 July). However, the change in the number of leaves among plants subjected to different treatments differed after the drainage ducts were blocked in 2018 May. The change in the number of leaves in plants in the Dr-P pots was higher than that in plants in the St-P pots in 2019 March (10 months), 2020 May (24 months), 2020 August (27 months), and 2020 October (29 months) after plugging drainage ducts with rubber plugs in 2018 May (P < 0.05). Additionally, the change in the number of leaves in plants in the Dr-Sa pots was greater than that in plants in the St-Sa in 2019 February (9 months), 2019 March (10 months), 2019 April (15 months), 2019 November (18 months), and 2020 May to August (24 to 27 months)(P < 0.05).
The change in the number of leaves on the (A) Castanopsis sieboldii and (B) Cinnamomum camphora. Vertical dotted lines show the time that drainage was stopped using rubber plugs in stagnating pots on the 2018 May 15. Fertilizer (N:P2O5:K2O = 8:8:8) at 5 kg·10 a-1 was at 2-month intervals from 2017 July 19, 2018 November 28 (arrows on either end), and 2019 November 12 (arrow). Sd-s and Pd-s indicate significant differences at P < 0.05 using Tukey’s test between drained and stagnating condition in sand or in soil improved with obsidian perlite, respectively; dS-P and sS-P indicate significant differences at P < 0.05 using Tukey’s test in sand or in soil improved with obsidian perlite in drained and stagnant conditions, respectively.
The effects of soil quality improvement caused by mixing sand with obsidian perlite (P) on the growth of C. sieboldii plants under soil water drainage (Dr) conditions are informative. The change in the number of leaves was greater in plants in the Dr-P pots than in those in the Dr-Sa pots. There were marked differences between plants in the Dr-P and Dr-Sa pots in 2020 October, 29 months after drainage was stopped. Regarding the effects of soil quality improvement on the growth of C. sieboldii plants under stagnant (St) soil water conditions, we observed significant differences in growth and survival between plants in the St-P and St-Sa pots in 2019 February, March, and November (9, 10, and 18 months, respectively) after stopping drainage (P < 0.05).
In the case of C. camphora plants (Figure 6B), there were significant differences between the growth and survival of plants in the Dr-Sa and St-Sa pots in 2018 September (4 months after drainage was stopped; P < 0.05), indicating that soil water drainage (Dr) and stagnation (St) differentially affected the growth of C. camphora plants grown on sand alone. There was a significant difference (P < 0.05) in plant growth and survival between C. camphora plants grown on sand alone (Sa) and sand mixed with obsidian perlite (P) under soil water drainage (Dr) or stagnation (St) in plants in 2019 November (18 months), 2020 May, August, and October (24 to 29 months), and 2021 October (41 months)(P < 0.05) after stopping drainage.
These results indicate that C. sieboldii plants are relatively less tolerant to soil water stagnation compared with C. camphora plants. However, soil quality improvement using obsidian perlite was more effective for the growth and survival of C. camphora plants than for C. sieboldii plants. Nevertheless, in pots with stagnant soil water (St-Sa and St-P), the change in the number of leaves for both C. sieboldii and C. camphora plants increased in 2018 May, 2019 March to May and 2020 November to May. These results seem to be influenced by fertiliser addition following the cessation of drainage.
SPAD Values
The SPAD values showed apparent differences between soil water drainage (Dr) and stagnation (St) conditions both for plants grown on sand alone (Sa) and sand mixed with obsidian perlite (P). For C. sieboldii, there was a statistically significant difference (P < 0.05) in SPAD values between plants in the Dr-Sa and St-Sa pots in 2018 November, 2019 February to May, and 2021 April, and between those in the Dr-P and St-P pots from 2019 February to March, 2019 November, and 2020 May (Figure 7A). For C. camphora plants, we observed a statistically significant difference in SPAD values between plants in the Dr-Sa and St-Sa pots from 2018 July to 2019 November and 2020 October, and between those in the Dr-P and St-P pots from 2018 September to 2019 March, 2019 November, 2020 May, and 2020 October (Figure 7B).
Soil and plant analyzer development (SPAD) value on each collection date in the (A) Castanopsis sieboldii and (B) Cinnamomum camphora. NS: non-significant difference; different letters indicate significant differences on each experimental date at P < 0.05 using Tukey’s test. Vertical dotted line indicates when drainage was stopped using rubber plugs in stagnating pots on 2018 May 15.
Soil quality improvement with obsidian perlite (P) affected SPAD values under both soil water drainage (Dr) and stagnation (St) conditions. For C. sieboldii, we observed a statistically significant difference (P < 0.05) between plants in the Dr-Sa and Dr-P pots in 2020 October and 2021 April, and between those in the St-Sa and St-P pots in 2019 March and 2019 August (Figure 7A). For C. camphora, we observed a statistically significant difference between plants in the St-Sa and St-P pots in 2018 July, 2019 February and May, and 2021 April, and between those in the Dr-Sa and Dr-P pots in 2018 September, 2019 February and November, and 2020 May (Figure 7B).
Plant Height
As some plants in the experimental pots died, it was difficult to ascertain statistical differences. However, for C. sieboldii, the plant height in the Dr-Sa pots (i.e., of those grown on sand under soil water drainage conditions) was higher than that of those in the St-Sa and Dr-P pots from 2018 May to 2021 April (0 to 27 months)(Figure 8A). For C. camphora, the heights of plants in the Dr-P and St-P pots were greater than the plant heights of those in the Dr-Sa pots. The differences in the heights of plants in the Dr-Sa and St-Sa pots and between those in the Dr-P and St-P pots were not apparent from 2019 November to 2021 April (18 to 35 months) after drainage was stopped (Figure 8B). The differences in plant heights indicated that stagnant soil water conditions decreased C. sieboldii growth on sand, and soil quality improvement using obsidian perlite increased C. camphora growth under both soil after drainage and stagnation conditions. The differences between soil water drainage and stagnation conditions were statistically insignificant for C. camphora growth.
Change in tree height on the (A) Castanopsis sieboldii and (B) Cinnamomum camphora. Vertical dotted line indicates when drainage was stopped using rubber plugs in stagnating pots on 2018 May 15. There were no statistical differences at P < 0.05 using Tukey’s test.
Diameter at the Base of the Trunk
There were no differences in the diameter at the base of the main trunk among the 4 experimental sets, except for that in C. camphora plants in the St-Sa and Dr-Sa pots in 2019 May (12 months) after drainage was stopped (Figures 9A and 9B).
Change in diameter at the base of trunk on the (A) Castanopsis sieboldii and (B) Cinnamomum camphora. Vertical dotted line indicates when drainage was stopped using rubber plugs in stagnating pots on the 2018 May 15. There were no statistical differences at P < 0.05 using Tukey’s test except 2019 May in (B) Cinnamomum comphora. Sd-s indicate significant differences between drained and stagnating condition in sand at P < 0.05 using Tukey’s test.
Number of the First Branches, the Length of the First Branch, and the Diameter at the Base of the First Branch
The number of the first branches (Figure 10), the length of the first branch (Figure 11), and the diameter at the base of the first branch (Figure 12) were similar in all experimental pots for both C. sieboldii and C. camphora.
Change in number of first branches on the (A) Castanopsis sieboldii and (B) Cinnamomum camphora. Vertical dotted line indicates when drainage was stopped using rubber plugs in stagnating pots on 2018 May 15. There were no statistical differences at P < 0.05 using Tukey’s test.
Change in length of first branches on the (A) Castanopsis sieboldii and (B) Cinnamomum camphora. The values are the difference between initial length and the current length at each experimental date. Vertical dotted line indicates when drainage was stopped using rubber plugs in stagnating pots on the 2018 May 15. There were no statistical differences at P < 0.05 using Tukey’s test.
Change in the diameter of the base of the first branches on the (A) Castanopsis sieboldii and (B) Cinnamomum camphora. Vertical dotted line indicates when drainage was stopped using rubber plugs in stagnating pots on the 2018 May 15. There were no statistical differences at P < 0.05 using Tukey’s test.
Discussion
Tolerance to Stagnant Soil Water and the Distribution of C. sieboldii and C. camphora
Growth under drained or stagnant soil water conditions (Dr-Sa, St-Sa, Dr-P, or St-P) differed between C. sieboldii and C. camphora. Stagnant soil water conditions following flooding were likely to affect C. sieboldii growth adversely, but little effect was observed on the growth of C. camphora.
Fujita et al. (2020) investigated root responses to waterlogging (i.e., stagnant soil water conditions) in 5 Japanese afforestation species (Pinus thunbergii, Acer mono, Quercus serrata, Alnus hirsuta, and Fraxinus mandshurica). They suggested that P. thunbergii is the most sensitive, followed by A. mono and Q. serrata. Alnus hirsuta and F. mandshurica are relatively tolerant. Colin-Belgrand et al. (1991) reported that the rate of root adaptation is much greater in Quercus robur than in Q. palustris or Q. rubra, probably indicating a higher flooding tolerance in Q. robur. Dreyer (1994) compared the sensitivity to waterlogging and associated root hypoxia of seedlings of 3 woody species (Q. robur, Q. rubra, and Fagus sylvatica). These studies indicate that Q. robur shows only a slight stress response to partial waterlogging, whereas Q. rubra exhibits an intermediate response. Fagus sylvatica displays the strongest adverse response to both treatments. Although Quercus species belong to the same plant family as Castanopsis spp. (Fagaceae), tolerance to flooding and soil water stagnation may vary among closely related species. The mechanism underlying tolerance to soil water stagnation has been discussed extensively (Yamamoto et al. 1995; Visser et al. 2000; Iwanaga and Yamamoto 2007).
Based on the experimental results discussed in this study, C. sieboldii distribution in the urban forests of Tokyo is discussed. Essentially, only a few giant C. sieboldii trees have been observed on riversides and in the lowlands (Watanabe et al. 2021), which raises the question of the impact of waterlogging on the survival of these trees. Owing to abundant springs and surface water runoff, the soil water content is high at the foot of the mountains (Ohta et al. 1985; Montgomery et al. 1997; Uchida et al. 2001). High water content reduces air content in soil; thus, trees growing at the foot of mountains or in lowlands are often flooded with stagnant soil water over their roots. Castanopsis sieboldii young trees grown in sand under stagnant soil water conditions showed little growth or died in our experiments. Young trees appear to die shortly after germination owing to reduced soil air phase. Following flooding, soil air content may remain low for a long time, and it is likely that this affects the establishment, survival, and growth of young C. sieboldii trees and, thus, the distribution of adult trees. In this study, we show that C. camphora grows under stagnant soil water conditions for 3 years or more, providing a probable explanation for the growth of giant C. camphora trees in lowland urban forest zones in southern China and southern Japan (Suzuki et al. 2016).
Considering the warm temperate climate of Japan is prone to heavy rainfall, plant tolerance to flooding or stagnant water conditions is relatively more important than tolerance to desiccation. The affected trees are often observed along urban roadways.
Effects of Soil Quality Improvement on Castanopsis sieboldii and Cinnamomum camphora Growth May Be Linked to Planting Techniques in the Urban Forest
Soil quality improvement by mixing sand with obsidian perlite enhanced the growth of both C. sieboldii and C. camphora plants in terms of the number of living plants, their dry weights, the number of leaves, and SPAD values, reflecting growth enhancement under drained and stagnant soil water conditions. These results indicate that soil quality improvement using obsidian perlite effectively in enhances plant survival and growth in both species.
The soil conditions in urban forests are significant factors that determine tree growth and survival. In this regard, Cambi et al. (2018) reported that in Q. robur, compaction had a greater effect on the root system than on shoots, particularly on root development in deeper soil layers (i.e., root system depth and main root length). In this study, young trees grew faster in the soil with improved quality (sand plus obsidian perlite) than in that containing sand alone. The addition of obsidian perlite increased the proportion of the soil air (gas) phase by up to 92.9%. Soil quality improvement with obsidian perlite increased soil aeration and enhanced plant survival and growth, which implies that both C. sieboldii and C. camphora plants are adversely affected by low aeration in sand. There was no apparent difference in soil moisture content between sand alone and sand mixed with perlite when dried at room temperature for a week. During summer to autumn each year, the soil moisture content in Wagner pots filled with sand alone was higher than in pots filled with sand and perlite. C. sieboldii and C. camphora plants grown in pots filled with sand and perlite pots did not die during this season in 2018. The low soil moisture content in pots filled with sand and perlite did not affect the growth of both C. sieboldii and C. camphora plants.
The effects of the physical properties of soil on the growth of crops, herbaceous plants, and fruit trees have been reported. For example, Watanabe and Kodama (1965) reported the effects of soils with different bulk densities and aeration conditions on early-stage crop growth, which indicates that optimal soil density and aeration are necessary for crop growth.
It has long been known that oxygen is necessary for root and shoot growth (Kramer 1940; Wilson and Kramer 1949; Neira et al. 2015). Furthermore, the adverse effects of stagnant soil water on plant growth have been linked to the lack of oxygen, and most higher plants cannot grow under hypoxic conditions. Hagan (1950) reported that soil aeration directly affects passive water absorption in the roots of transpiring soil-grown tomatoes in N- and CO2-rich environments. In another study on the effects of soil physical attributes on fruit tree growth, Morita and Yoneyama (1951) reported that when the oxygen level dropped below 5%, the leaves of pear, persimmon, and apple trees turned yellowish. The effects varied between species. Morita and Nishida (1952) studied the association between soil physical properties and fruit-tree growth and noted that peach and persimmon seedlings were affected by soil oxygen concentrations. With soil oxygen content of ≥ 7%, peach seedling growth was normal and comparable to that of controls. However, when the oxygen concentration decreased to < 5%, the growth of peach seedlings decreased substantially. With a soil oxygen concentration of < 3%, persimmon root development remained poor, although the plants survived. Moreover, when oxygen concentration was ≥ 5%, the growth of roots was near normal.
Due to compaction caused for example by walking, the setting of block materials or pavements, soils along urban streets exhibit low levels of air (Masuda et al. 1981; Berrang et al. 1985; Ohnuki and Matsumoto 1992). Masuda et al. (1983) reported physical properties of Zelkova serrata with root growth reduced in compacted soils with a high bulk density. Moreover, Kämäräinen et al. (2018) reported that aeration in conventional soils is sufficient for healthy tree growth. This study indicates that aeration strongly affects the growth of Castanopsis and Cinnamomum spp. in compacted soils. The typical habitats of street trees and heavily-used urban parks in urban forests are characterised by compacted soils. In this context, it is clear that the enhanced aeration of the compacted ground is particularly beneficial. In addition, soil temperature and nutrient availability affect plant growth and tolerance to soil water stagnation (Kramer 1940). The contents of water, nitrogen, organic matter, and oxygen in soil vary with soil type (Karsten 1939). In urban forestry, the association between soil temperature, nutrient availability and soil aeration and their effects on growth require investigation.
In pots with drainage, 1 young tree grown in sand, and 1 grown in sand mixed with obsidian perlite died at our experiment. These plant deaths can be attributed to low nutrient levels in sand and obsidian perlite materials. In this study, we applied 2-way ANOVA with 3 replicates. In future studies, we will investigate the effects of soil water drainage or stagnation conditions on plants grown on sand alone or sand mixed with obsidian perlite, with or without fertilisers, and with additional replicates.
Conclusions
The physiological responses and mortality of young C. sieboldii and C. camphora confirmed their tolerance to soil water stagnation and the effects of soil quality improvement (mixing sand with obsidian perlite). The number of living plants, their dry weights, the number of leaves, and SPAD values increased for both species when the soil quality was improved. However, a higher number of C. sieboldii plants died, and their dry weights and the number of leaves decreased more under soil water stagnation than under drainage conditions. In contrast, all C. camphora plants grown on sand alone or mixed with obsidian perlite lived for > 3 years, and the number of leaves was almost the same under drained and stagnant soil water conditions.
Castanopsis sieboldii and Cinnamomum camphora are evergreen trees distributed across central and southern Japan. Tolerance to soil water stagnation may impact on the distribution of these species. In this case, the selection of trees for roadside planting in urban forestry may need to favour species tolerant to soil water stagnation, such as Salix spp. and Taxodium distichum. In particular, Castanopsis sieboldii does not tolerate stagnant soil water conditions. The chemical and physical aspects of the soil must be considered when planting or maintaining these trees. However, to date, this has not been the case. In addition, soil quality improvement techniques are necessary for the sustainable planting of C. sieboldii and C. camphora in urban forests. Finally, a problem with giant trees in the urban forests in Tokyo is that relatively old specimens (conserved for the reasons mentioned earlier) have been progressively urbanised. This change over recent decades has altered soil conditions to the extent that, in many situations, trees have either died or are now in poor condition. The ongoing project on the history, distribution, conservation, and survival of the giant trees of Tokyo will provide valuable information for future urban forestry strategies and will be relevant to other urban centers globally.
Acknowledgements
This work was supported by JSPS KAKENHI (grant number: JP19K06114) and collaborative projects with global partner universities by the Research Institute of Agricultural and Life Science, Tokyo University of Agriculture. We would like to thank Marubashi Kaori (Tokyo Metropolitan Oshima High School) and the members of the Laboratory of Landscape Plants and Arboriculture for their assistance with field data collection.
- © 2025 International Society of Arboriculture
Literature Cited
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.