Water Conservation Capacity of Soil and Litter Layers of Five Magnoliaceae Plants in Hainan Island, China

Abstract

Magnoliaceae plants have high ornamental value, resulting in their widespread use in landscaping construction, and play a major role in the ecological functions of soil and water conservation. However, the landscape value of magnolias from the perspective of the water conservation capacity of the litter and soil layers is not yet fully understood; this restricts the popularization and application of Magnoliaceae plants in landscaping. In this study, we determined the characteristics of the litter thickness and mass, water absorption process, and soil water-holding capacity associated with five Magnoliaceae plants (Michelia shiluensis, M. crassipes, M. foveolata, M. maudiae, and M. odora). (1) The total litter thickness ranged from 2.29 to 5.58 cm, with M. crassipes achieving the highest value. The total litter mass for M. shiluensis (25.11 ± 2.58 t·ha⁻¹) was largely greater than that for the other magnolias. The mass of the un-decomposed litter (UL) layer was 1.31- to 3.82-fold larger than that of the semi-decomposed litter (SL). (2) The maximum water retention capacity (H_max) and effective water retention capacity (H_eff) of M. shiluensis were markedly larger than those of the other magnolias. (3) The W_max and W_eff of the UL layer were greater than those of the SL layer. (4) The soil bulk density varied from approximately 1.22 ± 0.08 g·cm⁻³ to 1.55 ± 0.08 g·cm⁻³, and the total soil porosity varied from 40.03 ± 3.44% to 46.42 ± 1.02%. The soil bulk density rose with an increasing soil depth, yet the total porosity was reduced. The soil water-holding capacity of the 0–30 cm soil layer varied from approximately 26.23 to 70.33 t·ha⁻¹, with soil near M. crassipes having the greatest water-holding capacity. The soil water infiltration recorded for M. crassipes was significantly higher than that of the other magnolias. The water conservation capacities associated with M. crassipes and M. shiluensis were the largest, which may suggest that these species are better at increasing rainfall interception, lightening splash erosion, and reducing surface runoff. Hence, we suggest that M. crassipes and M. shiluensis should be prioritized in landscaping applications.

1. Introduction

Magnoliaceae are an ancient lineage of relict plants [1] with a fossil record dating back 140 million years to the Cretaceous Period [2]. Once extensively distributed across a range of geological settings, magnolias are now mostly located in the tropical and subtropical zones of Asia and the Americas, including about 18 genera and over 300 species [1,3,4,5]. Magnoliaceae are pioneers among flowering plants [6,7] and are among the most typical examples of the evolution of the primitive angiosperms [8]. Integral to forest ecosystems, these plants hold significant scientific, ornamental, officinal, ecological, and economic value [9].

China serves as a considerable hub for the diverse Magnoliaceae family, having the greatest number of species, encompassing approximately 14 genera and over 165 species [1]. In addition to their use in landscaping, these plants serve as sources of spices, officinal resources, and industrial materials [10]. Previous studies have primarily focused on their geographical distribution, biological characteristics, domestication and introduction, genetic diversity and population structure, growth and physiological characteristics, and medicinal and ornamental applications [1,3,6,11,12,13,14,15,16]. Their beautiful shape and the pleasant appearance of their leaves and flowers make Magnoliaceae plants a treasure trove for landscaping. In their evaluation of ornamental Magnoliaceae, Fang et al. [17] found that Magnolia officinalis, Michelia maudiae, Yulania cylindrica, and Yulania amoena were excellent ornamental trees for landscaping, while Ma [18] found Michelia chapensis and Parakmeria lotungensis preferable. Similarly, Kong [19] recommended Michelia alba, Michelia champaca, and Manglietia hainanensis for landscaping. However, the landscape value of Magnoliaceae plants with regard to the water conservation capacity of the litter and soil layers is not yet fully understood, and only a limited number of studies have explored the differences in these plants’ water conservation capacity. Thus, the water conservation capacity of Magnoliaceae plants that are prioritized in landscaping applications should be further investigated.

The Hainan Tianxiang Magnolia Plant Conservation Research Center, which is devoted to the breeding of Magnoliaceae, was established in 2014. Covering an area of approximately 2000 ha, the center not only collects and preserves Magnoliaceae germplasm resources but also functions as a park space and provides both scientific education and ecological demonstrations. The center is home to six genera, more than 20 species, and more than 50,000 plants, including many plants that are on the IUCN Red List of Threatened Species, such as Manglietiastrum sinicum, Michelia shiluensis, Manglietia megaphylla, Manglietia aromatic, Manglietia forrestii, Tsoongiodendron odorum, Magnolia liliflora, etc. The Hainan Tianxiang Magnolia Plant Conservation Research Center supplies many ornamental seedlings for landscaping purposes, such as Michelia shiluensis, which is a species endemic to Hainan. In particular, Magnoliaceae species from the Michelia genus have the most landscaping applications in gardens, courtyards, campuses, and so on, and the water conservation capacity of the litter and soil layers in the Michelia genus remains poorly understood.

In this study, five Magnoliaceae species from the Michelia genus (Michelia shiluensis, M. crassipes, M. foveolata, M. maudiae, and M. odora) at the Hainan Tianxiang Magnolia Plant Conservation Research Center were surveyed. We quantified the five Magnoliaceae plants’ litter thickness, litter mass, soil porosity, water absorption process, soil water-holding capacity, and soil infiltration and then synthetically assessed the water conservation capacity using the entropy weight method. Our aim was to illustrate the differences in the water conservation capacity of the litter and soil layers of Magnoliaceae plants and to provide a theoretical basis and practical significance for selecting ornamental tree species for landscaping.

2. Materials and Methods

2.1. Study Sites

This study was conducted at the Hainan Tianxiang Magnolia Plant Conservation Research Center (19°48′~19°50′ N, 109°67′~109°69′ E) (Figure 1) in Lanyang Town, Danzhou City, Hainan Province, China. The region is located in the tropics and belongs to a tropical monsoon climate; the annual average temperature is 23.5 °C, and the mean annual rainfall is around 1823 mm. The soils are typical latosols. Various Magnoliaceae plants, such as Michelia shiluensis, Michelia crassipes, Michelia foveolata, Michelia maudiae, Michelia odora, etc., were introduced to the study area, with a planting density of 400 trees·ha⁻¹.

2.2. Litter and Soil Sample Collection

Field sampling was conducted for the five Magnoliaceae Michelia species (M. shiluensis, M. crassipes, M. foveolata, M. maudiae, M. odora) in mid-January 2021. According to the Chinese National Standard (GB/T 33027-2016) [20], three 20 m × 20 m plots were established for each plant species.

Table 1. The basic information of the sampling plots.

Tree Species	Stand Age (Years)	Average Tree Height (m)	Average Diameter at Breast Height (cm)	Tree Density (Trees·ha−1)	Canopy Density	Slope Gradient (°)
M. shiluensis	7	4.85 ± 0.42	13.09 ± 1.01	400	0.24	3~5
M. crassipes	7	2.39 ± 0.31	11.87 ± 1.12 ★	400	0.22	3~5
M. foveolata	7	5.45 ± 0.33	12.37 ± 1.52	400	0.24	3~5
M. maudiae	7	5.91 ± 0.53	9.63 ± 0.92	400	0.22	3~5
M. odora	7	5.01 ± 0.48	11.94 ± 1.21	400	0.34	3~5

Note: Data are mean ± S.D. ^★: Ground diameter of M. crassipes.

shows the basic information of the sampling plots. We sampled the semi-decomposed litter (SL layer) and un-decomposed litter (UL layer) in 0.5 m × 0.5 m quadrats, with five quadrats randomly selected for each plot. We collected a total of 150 litter bags (5 species × 3 plots × 5 quadrats × 2 litter layers). Meanwhile, we recorded the litter thickness. Furthermore, we sampled soil samples according to soil profiles at 0 to 10 cm, 10 to 20 cm, and 20 to 30 cm layers using the cutting ring (100 cm³) method. Three soil profiles were randomly selected within each plot, and for each soil layer, we collected two cutting rings. We obtained a total of 270 soil samples (5 species × 3 plots × 3 soil profiles × 3 soil layers × 2 cutting rings).

2.3. Laboratory Analyses

Upon the receipt of the litter samples, the mass of the fresh litter (m_f) was determined in the field and the amount of litter mass (m₀) was measured via drying at 75 °C in the lab [21]. The indoor water soaking method was used to evaluate the litter water-holding capacity [22,23], which involved soaking the litter in water for a period of time, which was set to 0.25, 0.5, 1, 2, 4, 6, 8, 12, and 24 h. The maximum water-holding capacity of the litter (G_m) was calculated as follows [22,23]:

G₀ = (m_f − m₀)/m₀ × 100

(1)

G_m = (m₂₄ − m₀)/m₀ × 100

(2)

where G₀ is the litter water-holding capacity under a natural environment (%), G_m is the litter’s maximum water-holding capacity (%), and m₀, m_f, and m₂₄ are the dry litter mass, fresh litter mass, and the litter mass after soaking for 24 h, respectively.

The litter water retention capacity was calculated as follows [22,23,24]:

H_eff = (0.85G_m − G₀) × M

(3)

H_max = (G_m − G₀) × M

(4)

where H_eff is the effective water retention capacity (t·ha⁻¹), H_max is the maximum water retention capacity (t·ha⁻¹), and M is the unit litter mass (t·ha⁻¹).

The physical properties of the soil samples were determined. The soil water content was determined by the oven-drying method, and the soil’s porosity was determined by the cutting ring method according to the Forestry Industry Standard of the People’s Republic of China (LY/T 1215-1999) [25]. The soil water-holding capacity was calculated as follows [26,27]:

S = dp × 10,000

(5)

where S is the soil water-holding capacity (t·ha⁻¹), d is the thickness of the soil layer (m), and p is the soil non-capillary porosity (%).

The soil infiltration rate was determined according to Hu et al. [28].

2.4. Synthetic Assessment of Water Conservation Capacity

We quantified the water conservation capacity of the soil layer and litter layer in five Magnoliaceae plants according to the entropy weight method (EWM). The calculation procedures were as follows: step 1: construct a decision matrix; step 2: raw data standardization; step 3: calculate the feature weight; step 4: calculate the entropy; step 5: calculate the weight parameters; and step 6: calculate the water conservation capacity [26,29,30,31].

$$WCC={\sum }_{I=1}^m{W}_i{p}_{ij}$$

(6)

where WCC is the water conservation capacity index, W_i is the weight parameter of the jth indicator of the ith object, and p_ij is the feature weight of the jth indicator of the ith object.

2.5. Statistical Analysis

The analyses were completed with SPSS v. 18.0 (SPSS Inc., Chicago, IL, USA). A one-way analysis of variance (ANOVA) was used to analyze the differences in the water conservation capacity of the litter layer and soil layer. The least significant differences (LSDs) were measured for multiple comparisons. A figure plot was created using Origin 2021 software (Origin, Origin Lab, Farmington, ME, USA).

3. Results

3.1. The Litter Thickness and Mass Associated with the Five Magnoliaceae Plants

We identified dramatic differences in the litter thickness among the five Michelia Magnoliaceae plants (p < 0.05) (Figure 2A), which were dependent on the plants’ differing biological characteristics (p < 0.05). M. crassipes had the greatest total litter thickness (5.58 ± 0.58 cm), followed by M. shiluensis (3.58 ± 0.45 cm), M. odora (3.57 ± 0.67 cm), M. maudiae (2.43 ± 0.47 cm), and M. foveolata (2.29 ± 0.32 cm). The thickness of the UL layers in the five magnolias was greater than that of their SL layers (p < 0.05).

We also found notable differences in the litter mass in the two litter layers in the five magnolias (p < 0.05) (Figure 2B). M. shiluensis had the highest total litter mass (25.11 ± 2.58 t·ha⁻¹), which was markedly larger than that of M. crassipes (15.76 ± 3.29 t·ha⁻¹), M. foveolata (12.05 ± 1.77 t·ha⁻¹), M. maudiae (6.65 ± 0.59 t·ha⁻¹), and M. odora (6.27 ± 0.44 t·ha⁻¹) (p < 0.05). The litter masses of the UL layer and SL layer decreased in the order of M. shiluensis > M. crassipes > M. foveolata > M. maudiae > M. odora (p < 0.05). The UL mass accounted for 79.41 ± 2.14% of the total litter mass in M. shiluensis, 70.39 ± 3.18% in M. crassipes, 64.09 ± 2.05% in M. foveolata, 59.41 ± 3.11% in M. maudiae, and 56.36 ± 1.96% in M. odora. The UL-layer litter mass was higher than that of the SL layer in all five magnolias (p < 0.05) (Figure 2B).

3.2. G_m, H_eff, and H_max

The G_m in the SL layer was 429.60% ± 15.54% for M. crassipes and was not significantly different from that of M. odora (416.51% ± 16.35%) (Figure 3A). However, both were significantly greater than those of M. maudiae (383.87% ± 7.55%), M. shiluensis (351.91% ± 41.96%), and M. foveolata (282.76% ± 23.55%) (p < 0.05). The G_m of the UL layer varied from 298.20% ± 23.14% to 406.24% ± 20.37%, and no significant difference was found between M. crassipes and M. odora or between M. shiluensis, M. foveolata, and M. maudiae (p > 0.05). However, the G_m of the UL layer of M. crassipes and M. odora was larger than that of M. shiluensis, M. foveolata, and M. maudiae, and the G_m of the SL layer was higher than that of the UL layer in M. shiluensis, M. crassipes, M. maudiae, and M. odora, while the opposite trend was discovered for M. foveolata. The G_m of M. crassipes was 417.17% ± 13.33%, which was roughly equivalent to that of M. odora (411.38% ± 14.59%); both G_m values were significantly greater than those of M. maudiae (345.61% ± 15.13%), M. shiluensis (330.66% ± 20.80%), and M. foveolate (290.48% ± 12.99%) (p < 0.05).

The H_max differed considerably among the five Magnoliaceae plants (p < 0.05) (Figure 3B). The H_max of M. shiluensis was 79.90 ± 5.62 t·ha⁻¹, higher than that of M. crassipes (64.96 ± 5.16 t·ha⁻¹) and obviously higher than that of M. foveolata (35.26 ± 4.17 t·ha⁻¹), M. odora (25.75 ± 2.59 t·ha⁻¹), and M. maudiae (22.51 ± 2.03 t·ha⁻¹) (p < 0.05). The H_max of the SL layer was 20.32 ± 2.12 t·ha⁻¹ for M. crassipes, 18.30 ± 2.89 t·ha⁻¹ for M. shiluensis, 12.27 ± 1.07 t·ha⁻¹ for M. foveolata, 11.29 ± 1.23 t·ha⁻¹ for M. odora, and 10.40 ± 1.32 t·ha⁻¹ for M. maudiae, exhibiting dramatic differences (p < 0.05). In addition, the H_max values of the UL layer decreased in the order of M. shiluensis > M. crassipes > M. foveolata > M. odora > M. maudiae (p < 0.05), ranging from 12.11 ± 2.01 t·ha⁻¹ to 61.60 ± 5.46 t·ha⁻¹. The H_max of the UL layer was 1.16- to 3.37-fold greater than the H_max of the SL layer in the five Magnoliaceae plants (p < 0.05).

The trend of the H_eff in the five Magnoliaceae plants was similar to that of the H_max, and the H_eff differed remarkably among the five Magnoliaceae plants (p < 0.05) (Figure 3C). The H_eff of M. shiluensis was 54.00 ± 4.97 t·ha⁻¹, which was greater than the 48.47 ± 4.74 t·ha⁻¹ of M. crassipes, the 24.50 ± 4.82 t·ha⁻¹ of M. foveolata, the 20.40 ± 2.04 t·ha⁻¹ of M. odora, and the 15.86 ± 1.48 t·ha⁻¹ of M. maudiae (p < 0.05). M. shiluensis had the highest holding capacity, equal to 5.40 mm of rainfall, with 4.85 mm intercepted by M. crassipes, 2.45 mm by M. foveolata, 2.04 mm by M. odora, and 1.59 mm by M. maudiae. The SL layer H_eff did not differ considerably between M. crassipes (14.90 ± 1.86 t·ha⁻¹) and M. shiluensis (12.56 ± 2.14 t·ha⁻¹) (p > 0.05), but the findings were all significantly higher than those for M. foveolata (8.05 ± 1.51 t·ha⁻¹), M. maudiae (7.26 ± 1.26 t·ha⁻¹), and M. odora (8.77 ± 0.80 t·ha⁻¹) (p < 0.05). The H_eff in the UL layer decreased in the order of M. shiluensis > M. crassipes > M. foveolata > M. odora > M. maudiae (p < 0.05) and varied from 8.60 ± 0.98 t·ha⁻¹ to 41.44 ± 3.43 t·ha⁻¹. The UL layer H_eff was 1.18- to 3.30-fold greater than the SL layer W_eff in the five Magnoliaceae plants (p < 0.05).

3.3. Litter Water-Holding Ratio

The water-holding ratio of the SL-layer litter and the UL-layer litter showed the same tendency, which was that it increased with the immersion time. Within the first 2 h, the litter’s water-holding ratio was at its largest; after that, the process began to slow, and complete saturation was achieved at 24 h (Figure 4). The water-holding ratio of the SL-layer litter was relatively higher than that of the UL layer subjected to the same immersion time. In the experiment for water immersion for 0.25 h, the water-holding ratios of the SL layers reached 6.67 t·ha⁻¹, 7.83 t·ha⁻¹, 4.66 t·ha⁻¹, 7.33 t·ha⁻¹, and 8.85 t·ha⁻¹ for M. shiluensis, M. crassipes, M. foveolata, M. maudiae, and M. odora, respectively, while the water-holding ratios of the UL layers reached 4.92 t·ha⁻¹, 6.33 t·ha⁻¹, 4.76 t·ha⁻¹, 3.35 t·ha⁻¹, and 5.77 t·ha⁻¹ for M. shiluensis, M. crassipes, M. foveolata, M. maudiae, and M. odora, respectively (Figure 4). In general, the water-holding ratio of the UL-layer litter decreased in the following order: M. crassipes > M. odora > M. shiluensis > M. foveolata > M. maudiae. Moreover, the water-holding ratio of the SL-layer litter decreased in the following order: M. crassipes > M. odora > M. maudiae > M. shiluensis > M. foveolata. The water-holding ratio and the immersion times for UL- and SL-layer litter from the five magnolias showed a logarithmic relationship.

3.4. Litter Water Absorption Rates

Both the SL-layer and UL-layer litter water absorption rates were observed to show the same tendency, with most of the absorption occurring at the beginning of immersion and declining rapidly within the first 2 h, gradually slowing between 2 and 12 h, and then exhibiting very little change after 12 h (Figure 5). The SL-layer litter water absorption rate was higher than that of the UL layer at the same time. After 1 h, the SL layer water absorption rates were 7.41 t·ha⁻¹·h⁻¹, 10.16 t·ha⁻¹·h⁻¹, 5.63 t·ha⁻¹·h⁻¹, 8.55 t·ha⁻¹·h⁻¹, and 9.78 t·ha⁻¹·h⁻¹ for M. shiluensis, M. crassipes, M. foveolata, M. maudiae, and M. odora, respectively (Figure 5b,d,f,h,j), while the UL layer water absorption rates were 6.12 t·ha⁻¹·h⁻¹, 8.44 t·ha⁻¹·h⁻¹, 5.78 t·ha⁻¹·h⁻¹, 4.45 t·ha⁻¹·h⁻¹, and 7.48 t·ha⁻¹·h⁻¹ for M. shiluensis, M. crassipes, M. foveolata, M. maudiae, and M. odora, respectively (Figure 5a,c,e,g,i). We observed an exponential relationship between the water absorption rate and the immersion time for UL- and SL-layer litter in the five magnolias.

3.5. Soil Water-Holding Capacity

The soil bulk density rose with an increasing soil depth, yet the total porosity was reduced near all five Magnoliaceae plants, and no significant differences were observed (p > 0.05). Within the 0 to 30 cm soil layer, the soil bulk density ranged from 1.22 ± 0.08 g·cm⁻³ to 1.55 ± 0.88 g·cm⁻³, with the largest value recorded for M. maudiae, followed by M. odora, M. shiluensis, M. foveolata, and M. crassipes (

Table 2. The soil litter water-holding capacity for the five Magnoliaceae plants. The different lowercase letters in each category indicate a significant difference among the five magnolias.

Tree Species	Soil Depth (cm)	Bulk Density (g·cm−3)	Non-Capillary Porosity (%)	Capillary Porosity (%)	Total Porosity (%)	Soil Water-Holding Capacity (t·ha−1)
	0 to 10	1.33 ± 0.10 a	7.38 ± 5.15 a	37.77 ± 3.71 a	45.15 ± 1.61 a	73.77 ± 51.52 a
M. shiluensis	10 to 20	1.33 ± 0.10 a	7.11 ± 6.67 a	31.82 ± 7.02 a	38.93 ± 0.40 a	71.13 ± 66.67 a
	20 to 30	1.38 ± 0.05 a	5.95 ± 3.01 a	37.16 ± 6.04 a	43.11 ± 4.73 a	59.47 ± 30.09 b
	0 to 10	1.16 ± 0.18 a	8.46 ± 1.46 a	38.27 ± 1.80 a	46.73 ± 2.81 a	84.60 ± 14.57 a
M. crassipes	10 to 20	1.19 ± 0.10 a	5.35 ± 1.93 a	42.36 ± 2.18 a	47.70 ± 3.86 a	53.47 ± 19.30 b
	20 to 30	1.30 ± 0.02 a	2.72 ± 1.67 b	42.11 ± 0.96 a	44.83 ± 2.03 a	27.20 ± 16.74 c
	0 to 10	1.26 ± 0.13 a	7.81 ± 5.57 a	39.91 ± 4.98 a	47.72 ± 1.61 a	78.07 ± 55.68 a
M. foveolata	10 to 20	1.36 ± 0.27 a	5.66 ± 3.23 a	36.80 ± 6.55 a	42.47 ± 8.09 a	56.63 ± 32.27 b
	20 to 30	1.29 ± 0.17 a	7.63 ± 3.53 a	37.68 ± 4.64 a	45.31 ± 8.05 a	76.30 ± 35.28 a
	0 to 10	1.51 ± 0.13 a	2.82 ± 1.79 a	39.37 ± 4.33 a	42.19 ± 5.96 a	28.17 ± 17.87 a
M. maudiae	10 to 20	1.59 ± 0.12 a	2.90 ± 1.45 a	37.17 ± 1.24 a	40.07 ± 2.10 a	29.03 ± 14.49 a
	20 to 30	1.54 ± 0.04 a	2.15 ± 0.62 a	35.69 ± 2.47 a	37.84 ± 3.03 a	21.50 ± 6.18 a
	0 to 10	1.44 ± 0.29 a	4.86 ± 2.70 a	35.81 ± 3.69 a	40.67 ± 6.38 a	48.60 ± 26.97 a
M. odora	10 to 20	1.45 ± 0.31 a	5.32 ± 3.97 a	37.19 ± 4.79 a	42.51 ± 8.73 a	53.17 ± 39.75 a
	20 to 30	1.40 ± 0.27 a	5.67 ± 5.09 a	39.90 ± 1.71 a	45.57 ± 6.03 a	56.73 ± 50.91 a

). The non-capillary porosity of the soil near M. maudiae was 2.62 ± 1.27% lower than that of the other magnolias (p < 0.05), which achieved non-capillary porosities of 6.81 ± 4.59%, 5.51 ± 1.25%, 7.03 ± 0.88%, and 5.28 ± 3.88% for M. shiluensis, M. crassipes, M. foveolata, and M. odora, respectively. The capillary porosity ranged from 35.58 ± 5.57% to 40.91 ± 0.30%, with the largest value recorded for M. crassipes. The total soil porosity was 42.40 ± 1.51%, 46.42 ± 1.02%, 45.16 ± 5.08%, 40.03 ± 3.44%, and 42.92 ± 6.84% near M. shiluensis, M. crassipes, M. foveolata, M. maudiae, and M. odora, respectively (Table 2). Again, no significant differences regarding the soil capillary porosity and total soil porosity were observed (p > 0.05). The soil water-holding capacity near M. foveolata displayed greater values for the 0 to 30 cm soil profile, while M. maudiae had the minimum value. The soil water-holding capacity decreased as follows: M. crassipes (70.33 t·ha⁻¹) > M. shiluensis (68.12 t·ha⁻¹) > M. foveolata (55.09 t·ha⁻¹) > M. odora (52.83 t·ha⁻¹) > M. maudiae (26.23 t·ha⁻¹) (p < 0.05) (Table 2).

3.6. Variations in Soil Infiltration

The soil infiltration rate was highest at the beginning of the infiltration and gradually declined until it reached a steady state, which occurred in tandem with an increase in the infiltration time (Figure 6). Remarkable differences were observed in the five Magnoliaceae plants (p < 0.05). The soil infiltration rate near M. crassipes was significantly greater than that recorded for M. shiluensis, M. maudiae, and M. odora, which exhibited no significant differences, and higher than that for M. foveolata (p < 0.05) (Figure 6). The initial soil infiltration decreased in the following order: M. crassipes (5.49 mm·min⁻¹) > M. maudiae (3.84 mm·min⁻¹) > M. odora (3.33 mm·min⁻¹) > M. shiluensis (2.86 mm·min⁻¹) > M. foveolata (2.24 mm·min⁻¹) (p < 0.05). Moreover, the steady soil infiltration decreased in the following order: M. crassipes (3.18 mm·min⁻¹) > M. odora (2.09 mm·min⁻¹) > M. maudiae (1.90 mm·min⁻¹) > M. shiluensis (1.81 mm·min⁻¹) > M. foveolata (1.35 mm·min⁻¹) (p < 0.05). The soil near M. maudiae was the first to achieve steady infiltration, with a 28 min duration, while the soil near M. shiluensis had the longest steady soil infiltration time (42 min). We observed an exponential relationship between the soil infiltration rate and the infiltration time in UL- and SL-layer litter for the five magnolias.

3.7. Water Conservation Capacity

The water conservation capacities of the five Magnoliaceae plants were determined by the EWM, and the indexes of the litter and soil layers were selected; the litter indexes included the litter thickness, litter mass, G_m, H_eff, and H_max, while the soil layer indexes included the bulk density, capillary porosity, total porosity, soil water-holding capacity, and soil infiltration. The weighted value of each index was assessed, and the values varied from 0.0613 to 0.1427 (

Table 3. Weighted values of water conservation capacity indexes in five Magnoliaceae plants.

Grade I Index	Weight	Serial No.	Grade II Index	Weight
		X1	Litter thickness	0.1321
		X2	Litter mass	0.1427
Litter layer	0.6152	X3	Gm	0.0816
		X4	Heff	0.1221
		X5	Hmax	0.1367
		X6	Bulk density	0.0698
		X7	Capillary porosity	0.0807
Soil layer	0.3848	X8	Total porosity	0.0747
		X9	Soil water-holding capacity	0.0613
		X10	Soil infiltration	0.0983

), and the weighted values were ordered as follows: litter mass > Hmax > litter thickness > Heff > soil infiltration > Gm > capillary porosity > total porosity > bulk density > soil water-holding capacity. The values for the litter layer were 1.60 times heavier than those of the soil layer.

The synthetically assessed value of the water conservation capacity of the litter layer was greater than that of the soil layer near M. shiluensis and M. crassipes, whereas the opposite tendency was observed for M. foveolata, M. maudiae, and M. odora. M. crassipes indicated the highest water conservation capacity (0.3422), followed by M. shiluensis (0.2932), M. foveolata (0.1790), M. odora (0.1455), and finally M. maudiae, which had the lowest (0.0543) (

Table 4. Synthetical assessment of water conservation capacity in five Magnoliaceae plants.

Tree Species	Water Conservation Capacity		Comprehensive Evaluation Value	Rank
	Litter Layer	Soil Layer
M. shiluensis	0.2391	0.0541	0.2932	2
M. crassipes	0.2394	0.1028	0.3422	1
M. foveolata	0.0515	0.1275	0.1790	3
M. maudiae	0.0178	0.0365	0.0543	5
M. odora	0.0637	0.0818	0.1455	4

).

4. Discussion

The water conservation capacity of litter and soil layers plays a major role in regulating the water cycle [23,26,31,32,33] and can have a remarkable impact on the forest ecosystem water budget [34,35,36]. The litter layer has a role in intercepting rainfall, lightening the effects of raindrops splashing, and reducing surface runoff via soil infiltration [36,37,38]. Previous studies have indicated that a thick litter layer protects the soil structure from rain while effectively improving the soil’s water conservation capacity [22,35,39,40]. The litter thickness and mass vary among species, influencing their hydrological characteristics [22,29,35,39]. In this study, we observed that the litter thickness varied from 2.29 to 5.58 cm, and the litter mass ranged from approximately 6.27 to 25.11 t·ha⁻¹. M. crassipes and M. shiluensis achieved the best results, suggesting that these species significantly reduce evaporation compared to other Magnoliaceae plants. The UL-layer litter thickness beneath the five magnolias was higher than the thickness of the SL layer, indicating that the litter decomposition rate was slow in these plants. This may be attributed to the thick, leathery quality of their leaves, which have a relatively slow decomposition rate. Moreover, the significant differences in the litter mass between the SL and UL layers in the magnolias were also observed (p < 0.05), with M. shiluensis exhibiting an exceptionally large litter mass, which indicates that it is favorable for rainfall interception. The UL-layer litter mass accounted for 56.36%~79.41% of the total litter mass, which was higher than that of the SL layer, in agreement with our predecessors’ research [22,23,24,41,42]. However, the rainfall intercepted by M. shiluensis litter is likely to evaporate back into the atmosphere before infiltration [33,37,43]. This may mean that we need to conduct a long-term in situ survey of the litter water-holding capacity under natural rainfall [23,29].

The maximum water-holding capacity G_m defines the greatest effectiveness per unit of litter mass with regard to water retention [22,24,44]. Variations in G_m are affected by the forest structure, litter composition, and litter decomposition characteristics [43,45,46]. In this study, the G_m varied from 290.48% ± 12.99% to 411.38% ± 14.59%, meaning that the litter absorbed about 2.90 to 4.11 times its own dry weight in water. M. foveolata litter had the smallest G_m, whereas M. crassipes had the largest. This variation may have been due to the diversification of the litter’s composition, which may have included differing quantities of leaves, dead branches, and seeds [23,29,41]. In addition, the G_m of the SL layer was larger than that of the UL layer, concurring with the findings of Tu et al.’s study [29]. This may be attributed to a higher degree of fragmentation in the SL layer and the presence of fewer dead branches in the study area [23,29,41,42].

The maximum water retention capacity H_max depends on the natural water content of the litter mass and also the natural rainfall [37,47]. In our study, the H_max recorded for M. shiluensis litter was remarkably higher than that of the other magnolias. This may have beeen due to M. shiluensis having the larger litter mass, which may have bolstered its effectiveness compared with the other magnolias. This is consistent with previous studies, which suggested that the H_max depends on litter storage [22,35,41]. We also found that the H_max of the UL layer was 1.16 to 3.37 times higher than that of the SL due to a greater litter mass. These findings align with those of Li et al. [23], Wu et al. [41], and Tang et al. [42].

The H_eff defines the effective water retention capacity, which is used to evaluate the rainfall absorption potential [29,43,48]. The H_eff is also influenced by the litter’s natural water content, litter mass, and rainfall intensity [29,33,43]. Our study observed that the H_eff differed remarkably among the five magnolias (p < 0.05), and the H_eff of M. shiluensis litter was higher than that of the other plants. M. shiluensis litter had a greater capacity for reducing surface runoff and exhibited the greatest water retention capacity and a better rainfall interception ability [35,40,49]. As such, this plant contributes to improving soil and water conservation [29,41].

Litter water-holding processes reflect the hydrological characteristics of the forest’s litter [22,45]. The rapid water absorption rates observed at the beginning of the experiment gradually decreased until the 12 h mark and remained almost unchanged thereafter, likely due to the drier litter showing lower matric water potential at the beginning of the experiment. Previous studies also observed the same dynamic change in the trend of the litter water-holding ratio recorded [22,26,29,31]. In addition, the UL layer water-holding ratio was higher than that of the SL layer, which may have been due to the greater litter mass of the UL layer [41,42].

The soil water-holding capacity of the forest ecosystem is a key part of water conservation and also plays an important role in regulating hydrological processes [23,29,35,41,50]. The soil’s water-holding capacity directly affects the surface runoff and soil infiltration [23,26,41,42]. The soil’s hydrophysical properties are important factors and likely depend on the soil’s bulk density, soil structure, and soil porosity [40]. In our study, the soil bulk density varied from 1.22 g·cm⁻³ to 1.55 g·cm⁻³, and the total soil porosity varied from 40.03% to 46.42%. Generally, the soil bulk density rose as the soil depth increased, while the total soil porosity showed a gradual decrease; this finding was consistent with previous research [23,29,41,42]. A lower soil bulk density was associated with greater soil porosity and improved the water-holding capacity associated with M. crassipes (70.33 t·ha⁻¹), which achieved a capacity 1.03-, 1.27-, 1.33-, and 2.68-fold higher than that recorded for M. shiluensis, M. foveolata, M. odora, and M. maudiae, respectively. The soil infiltration is another important indicator of soil’s hydrological properties [41,42]. The soil infiltration near M. crassipes was significantly greater than that near the other magnolias, indicating that M. shiluensis litter has a greater capacity to reduce surface runoff and enhance infiltration, contributing to improved rainfall interception [33,41,42]. Further studies should focus on the soil’s microbiological properties to improve our understanding of differences in the water-holding capacity among different plant species, specifically with regard to soil enzyme activities and microbe communities.

The water conservation capacity of the litter layer was 1.60 times larger than that of the soil layer, contradicting the findings of research by our predecessors Bai et al. [26] in 2021 and Tu et al. [29] in 2023. The soils associated with M. crassipes and M. shiluensis had the greatest water-holding capacity, as demonstrated by the larger water-holding and interception capacities of the litter and soil layers, which allowed M. crassipes and M. shiluensis to conserve more water. Our results suggest that, from a water conservation perspective, M. crassipes and M. shiluensis should take priority in landscaping and other ecological services.

5. Conclusions

In addition to their strong ornamental value, Magnoliaceae plants play a major role in soil and water conservation. We studied the water conservation capacity of the litter and soil layers near five Magnoliaceae plants and found that the total litter thickness and total litter mass in M. crassipes and M. shiluensis were largely greater than those for the other magnolias. The H_max and H_eff were markedly larger in M. crassipes and M. shiluensis, meaning that M. crassipes and M. shiluensis litter had far greater water-holding capacities than the litter of the other magnolias, with the UL layer having a higher capacity than the SL layer. In addition, a lower soil bulk density was associated with greater porosity near M. crassipes and M. shiluensis, representing a higher soil water-holding capacity and higher soil infiltration, thus reducing the loss of surface water. The water conservation capacity of the litter layer was 1.60 times larger than that of the soil layer. M. crassipes and M. shiluensis had the greatest water-holding capacities, offering the greatest potential for improvement in this area. In summary, we suggest that M. crassipes and M. shiluensis should be prioritized in future landscaping applications.