2.1 Site description
The experimental site locates on a rubber plantation in Wushi Farmland, Qiongzhong county, Hainan Island, China (19°4′3″−19°12′42″N, 109°47′6″−110°1′2″E, a.s.l. 200 − 500 m) (Fig. 1). The region has a tropical monsoon climate with the rainy season from May to October and the dry season from November to April of the next year. The annual average temperature, mean annual solar radiation and mean annual precipitation is 23.5°C, 4579 MJ·m− 2·yr− 1, and 2462 mm, respectively (Tang et al. 2013). The soils were classified as Udic Ferrallisols that were derived from granites, which were considered Oxisols according to USA Soil Taxonomy System (Zhang et al. 2007). There were two ages of rubber plantations elected in this study, the young rubber plantation (clone RY 7-33-97, planted at the recommended density of 3 m × 7 m in 2016) and the mature rubber plantation (clone PR-107, planted at the recommended density of 3 m × 7 m in 1999). Initial chemical properties and enzyme activities of soil in two rubber plantations were shown in Table 1.
2.2 Experimental design
In the present study, fresh kudzu was collected from a young rubber plantation. The fresh kudzu samples were subsequently divided into two parts: one part was separated from the stem and leaf, while the other part was kept the whole plant. All samples were fragmentation into less than 2 cm pieces and then dried at 105℃ for 0.5 h and at 65℃ for 48 h until a constant weight was achieved (Zhang et al. 2022). A 10 ± 0.1g (dry weight) of either single kudzu component (whole plant, stem, and leaf) was weighted into nylon mesh bags (nylon bags). The dimensions of the nylon bag were 15 cm × 20 cm (2 mm and 0.5 mm mesh size on the surface and bottom of the nylon bag, respectively), allowing microorganisms to access the bag and preventing the loss of small diameter fragments. The nylon bags were sprayed with deionized water after being weighed to prevent leaf breakage during handling and transport to the field (Lecerf et al. 2008). The initial nutrient concentrations of the kudzu were shown in Table 2.
Table 1
Initial soil chemical properties and enzyme activities of soil in two sites
|
Mature rubber plantation
|
Young rubber plantation
|
P
|
Moisture(%)
|
18.86 ± 3.17a
|
16.44 ± 1.14a
|
ns
|
pH
|
5.60 ± 0.41a
|
5.15 ± 0.11b
|
*
|
SOC(g·kg− 1)
|
21.02 ± 4.14a
|
15.5 ± 1.15b
|
*
|
Soil TN(g·kg− 1)
|
1.93 ± 0.25a
|
1.60 ± 0.06b
|
*
|
Soil TP(g·kg− 1)
|
0.26 ± 0.07b
|
1.31 ± 0.05a
|
**
|
Soil AP(mg·kg− 1)
|
26.47 ± 0.39b
|
51.57 ± 7.92a
|
**
|
Soil NO3−-N(mg·kg− 1)
|
6.11 ± 3.40b
|
17.66 ± 3.58a
|
**
|
Soil NH4+-N(mg·kg− 1)
|
12.07 ± 2.12a
|
14.09 ± 3.55a
|
ns
|
AcP(nmol·g− 1·h− 1)
|
110.6 ± 1.38a
|
78.68 ± 14.52b
|
**
|
BG(nmol·g− 1·h− 1)
|
128.3 ± 6.09a
|
100.59 ± 7.05b
|
**
|
NAG(nmol·g− 1·h− 1)
|
112.71 ± 3.02a
|
107.45 ± 11.1a
|
ns
|
LAP(nmol·g− 1·h− 1)
|
100.56 ± 1.08a
|
97.22 ± 2.77b
|
*
|
POX(nmol·g− 1·h− 1)
|
1.04 ± 0.17b
|
1.87 ± 0.43a
|
*
|
CAT(nmol·g− 1·h− 1)
|
0.63 ± 0.20b
|
0.91 ± 0.05a
|
*
|
Values are shown as mean ± standard errors (n = 3). SOC, soil organic carbon; Soil TN, soil total nitrogen; Soil TP, soil total phosphorus; Soil AP, Available phosphorus; Soil NH4+-N, soil ammonium nitrogen; Soil NO3−-N, soil nitrate-nitrogen; BG, β-1, 4-glucosidase activity; AcP, acid phosphatase activity; NAG, β-1, 4-N-Acetylglucosaminida activity; LAP, L-leucine aminopeptidase; POX, Polyphenol oxidase activity; CAT, catalase activity. Different lowercase letters indicate significant differences among the three kudzu components. C, Initial carbon contents of kudzu; N, Initial total nitrogen contents of kudzu; P, Initial total phosphorus contents of kudzu; Lignin, Initial lignin contents of kudzu. Different lowercase letters indicate significant differences in initial soil chemical properties and enzyme activities between the two rubber plantation sites (P < 0.05). *, P < 0.05; **, P < 0.01; ns, no significance.
Table 2
Initial endogenous nutrients concentrations of three kudzu components
|
Whole plant
|
Stem
|
Leaf
|
P
|
Lignin(%)
|
27.30 ± 1.91a
|
27.63 ± 1.77a
|
26.57 ± 2.52a
|
ns
|
C(g·kg− 1)
|
436.13 ± 15.39a
|
447.76 ± 19.25a
|
423.44 ± 23.45a
|
ns
|
N(g·kg− 1)
|
27.03 ± 0.50b
|
25.77 ± 0.38c
|
35.29 ± 0.14a
|
**
|
P(g·kg− 1)
|
1.63 ± 0.15b
|
1.57 ± 0.08b
|
2.17 ± 0.09a
|
**
|
C: N
|
16.15 ± 0.83a
|
17.39 ± 1.11a
|
12.00 ± 0.68b
|
*
|
C: P
|
269.37 ± 35.18a
|
286.33 ± 29.05a
|
194.84 ± 8.31b
|
**
|
N: P
|
16.64 ± 1.32a
|
16.45 ± 0.641a
|
16.26 ± 0.75a
|
ns
|
Lignin: N
|
8.09 ± 0.67a
|
8.58 ± 0.59a
|
6.02 ± 0.56b
|
**
|
Lignin: P
|
168.92 ± 27.45a
|
176.47 ± 15.65a
|
122.66 ± 15.93b
|
*
|
Values are shown as mean ± standard errors (n = 3). Different lowercase letters indicate significant differences among the three kudzu components. C, Initial carbon contents of kudzu; N, Initial total nitrogen contents of kudzu; P, Initial total phosphorus contents of kudzu; Lignin, Initial lignin contents of kudzu. Different lowercase letters indicate significant differences in initial endogenous nutrient concentrations among three kudzu components (P < 0.05). *, P < 0.05; **, P < 0.01; ns, no significance.
The field experiment under two returning methods in young and mature rubber plantations from October 2019 to July 2020. Specifically, three 20 m × 20 m plots were randomly selected at each rubber plantation site, with a 20 m distance separating the plots. Within each plot, two evenly spaced 8 rows × 3 column sampling grids were erected, and a nylon bag of each type was placed at each intersection point along the grid (Fig. 1). Each plot has 24 nylon bags in mulching and burying mode, which contained the three kinds of kudzu components. A total of 288 nylon bags (3 components × 2 returning method × 6 plots × 2 site× 4 sampling time) were used in this experiment and half of them (144 nylon bags) were attached with metal pins to the soil surface, while the rest (144 nylon bags) were incorporated into the soil at a depth of 10 cm.
2.3 Kudzu sampling and analyses
Nylon bags were retrieved at 30, 60, 90, and 270 days after the field experiment. Each nylon bag was carefully transported to the laboratory in a separate plastic bag to minimize the loss of any small material fragments from the bag. Specifically, the surface soil on each nylon bag was washed with tap water, and the kudzu in the nylon bag was washed with distilled water 3 times, dried at 65°C, weighed, ground (passed 0.25 mm sieve), and determine the carbon (C), nitrogen (N), phosphorus (P) and Lignin concentrations. The C concentration was determined using the dichromate oxidation method (Nelson et al. 1996), the N concentration by a Kjeldahl digestion method (Nelson et al. 1996), and the P concentration by molybdovanadate method (Lu, 1999). The Lignin concentrations were determined using the acid detergent method as described by Zhang et al. (2020).
The decomposition proportion (%) was calculated according to the Olson decay model (Olson, 1963):
$${M_t}={M_0} \times {e^{ - kt}}$$
1
where Mt is the remaining mass at time t (in days), and M0 is the initial mass, k is the decomposition rate constant calculated using the least-squares method, t is the decomposition time of kudzu.
The mass remaining (%), denoted as R was calculated according to the formula (Wang et al. 2021):
$$R=\frac{{{M_{\text{t}}}}}{{{M_0}}} \times 100\%$$
2
The decomposition rate of kudzu (Dec, g·d− 1), was calculated according to the formula (Wang et al. 2021):
The time for decomposition rate of 50% (T50%) and 95% (T95%) decomposition of the kudzu samples was estimated based on the k values estimated from the equation:
$${T_{50\% }}= - \frac{{\ln (1 - 0.50)}}{k}$$
4
$${T_{95\% }}= - \frac{{\ln (1 - 0.95)}}{k}$$
5
The nutrient release proportion (NRP, %) of kudzu, was calculated according to the formula (Wang et al. 2021):
$$NRP= - \frac{{{X_t} \times {M_t}}}{{{X_0} \times {M_0}}} \times 100$$
6
where Mt is the remaining mass at time t, and M0 is the initial mass; Xt is the concentrations of C, N, or P in kudzu sampled at time t, and X0 is the initial concentrations of C, N, or P in kudzu.
2.4 Soil sampling and analyses
Soil samples were collected simultaneously below each nylon bag (0‒10 cm depth) when nylon bags were taken out from the field. Meanwhile, soil samples were collected where without nylon bags at the same depth in each site as the control (CK). All soil samples were immediately transported to the laboratory. The soil samples were subsequently divided into two parts: one part was stored at 4°C for the measurements of soil ammonium-nitrogen (NH4+-N), nitrate-nitrogen (NO3−-N), and soil enzyme activities. The other part was air-dried and was sieved to 0.149 mm to determine other soil chemical properties.
Soil pH was measured using a suspension of 1:2.5 soil: water with a pH electrode (PHS-3E, INESA, China). The SOC content was determined using the dichromate oxidation method (Nelson et al. 1996). Soil TN, total phosphorus (TP), and available phosphorus (AP)were measured following the methods of Lu (1999). Soil NH4+-N and NO3−-N were extracted with a 2 M KCl solution and quantified with a continuous flow injection analyzer (Proxima1022/1/1, Elians Scientific Instruments, France). Activities of soil enzymes (AcP, BG, NAG, LAP, CAT, and POX) were measured fluorometrically with polystyrene 96-well, 300-ml microplates as described in DeForest’s method (2009).
2.5 Statistical analyses
Before performing analysis of variance (ANOVA), all data were checked for normality using the Shapiro–Wilk’s test and non-normality data were natural logarithm transformed. A two-way repeated-measures ANOVA was conducted to identify the effects of returning methods and decomposition time on the remaining mass of green manure and lignin loss rate. Similarly, three-way ANOVA was conducted to identify the effects of returning methods, study sites, and decomposition time on kudzu nutrient release. One-way ANOVA and Tukey's tests were used to detect significant differences in initial endogenous nutrients (including C, N, P, and lignin) among kudzu components, and soil environment (including the soil enzyme activity and chemical properties) among the four treatments. All statistical analyses were performed using SPSS 26.0 (IBM, Chicago, IL, USA).
A Pearson's correlation coefficients (r) and random forest model between initial kudzu endogenous nutrients and initial soil factors and kudzu decomposition (decomposition rate, lignin decomposition, C, N, P nutrients release proportions) were generated using the corrplot package in R software (R version 4.0.4). Similarly, correlation analysis was constructed to determine the effects of the decomposition of three kudzu components on soil chemical properties and soil enzyme activities under four treatments.