Partial Organic Substitution Improves Soil Quality and Increases Latex Yield in Rubber Plantations

Abstract

Partial organic substitution (POS) is a promising strategy to enhance soil fertility and agricultural sustainability. However, the mechanisms by which varying organic substitution ratios affect soil quality and latex yields in rubber plantations remain unclear. We conducted a two-year field experiment in a rubber plantation with six treatments: no fertilizer (CK), 100% synthetic fertilizer (NPK), and synthetic nitrogen fertilizer substituted with 25% (25 M), 50% (50 M), 75% (75 M), and 100% (100 M) manure. The results indicated that POS treatments significantly increased pH, soil organic carbon (SOC), total phosphorus (TP), total nitrogen (TN), NH₄⁺-N, enzyme activity, and leaf nutrient (C, N, and P) content compared to NPK. Compared with NPK, the soil quality (evaluated through the soil quality index, SQI) increased by 15.30–43.42% under POS across both years, with maximal values observed at 50 M (2020) and 75 M (2021); similarly, the latex yield increased by 2.10–18.60%. SOC, NO₃⁻-N,C:P ratio, TN, and pH are the key factors that influence soil quality and latex yield. Structural equation modeling indicated that fertilization and soil factors collectively explained 82% of the variation in latex yield. These results demonstrated that POS effectively alleviated soil acidity, enhanced soil quality, and improved latex productivity, with 50% manure substitution treatment (50M) identified as the optimal short-term substitution strategy in rubber plantations.

1. Introduction

Rubber trees (Hevea brasiliensis Muell. Arg.) are a crucial natural rubber source and is are extensively grown on approximately 13.79 million hectares of land worldwide [1]. Since their introduction in 1904, rubber plantations have become economically significant in tropical China. In 2023, China reached a natural rubber tree-planting area of 1.13 million hectares, producing 0.89 million tons. Nevertheless, China’s natural rubber production self-sufficiency rate is only 13.34%, which is significantly below the national strategic security threshold of 30% [2]. Enhancing natural rubber production to meet the ever-increasing demand driven by economic development has become a significant challenge [3,4]. Advancements in field management and afforestation techniques have significantly enhanced natural rubber yields. Among the various management strategies, fertilization has been identified as a promising approach for enhancing latex yields. Previous studies have found a positive relationship between fertilizer use and rubber latex production [5,6,7]. Gohet et al. [5] reported that latex yield increased by 5–10% under different fertilizer applications. However, Qu et al. [7] reported that rubber yields increased by 25.5–49.3% after the application of organic fertilizer, while chemical fertilizer had a negligible effect. In addition, the extended use of inorganic fertilizers in rubber plantations has led to a significant depletion of soil fertility and a decline in soil quality (SQ), causing a series of environmental problems, including nitrogen (N) leaching, soil acidification, soil organic carbon (SOC) decomposition, and hardening [8,9]. A global meta-analysis by Yang et al. [10] revealed that prolonged inorganic fertilizer application reduced soil microbial diversity by 18–32% and suppressed carbon cycle-related enzyme activity. Additionally, Li et al. [11] conducted a comprehensive meta-analysis across 48 sites and demonstrated that chemical nitrogen fertilization decreased SOC and total nitrogen (TN) by 3.83% and 11.46%, respectively. Therefore, it is crucial to address the critical issues of enhancing soil fertility in rubber plantations through precise fertilization, maintaining the health of the soil ecosystem, and increasing rubber yields.

Partial organic substitution (POS) fertilization in agriculture is a promising solution for promoting sustainable practices [8]. Reducing the use of chemical fertilizers and adopting organic materials can improve soil health and productivity and minimize negative environmental effects [8,12,13]. The application of manure can enhance soil aggregate formation and improve the soil structure. Additionally, the alkalinity of manure can lead to an increase in soil pH upon incorporation into the soil [14]. This practice not only benefits crop yields but also contributes to the overall sustainability of agriculture [15]. Previous studies have extensively examined the beneficial effects of POS fertilization on soil physicochemical conditions, soil enzyme activity, and storage and utilization of soil nutrients, ultimately leading to increased crop production [12,15,16]. Xu et al. [16] revealed that compared to using chemical fertilizer alone, POS fertilization increased SOC by 14.63–54.53%, TN by 39.58–63.16%, and total phosphorus (TP) content by 20.11–227.78%. Soil enzymes are essential for transforming soil materials and facilitating energy metabolism, which is related to microorganisms and plants and is a paramount biological mechanism linking microbial community dynamics and soil nutrient cycling [17]. Jia et al. [18] reported that the C-associate enzyme levels in 100% manure treatment were 34.0–56.3% higher in the topsoil and 108.7–164.1% higher in the subsoil than in CK and NPK, respectively. In yellow loam soil, the substitution of 10–30% synthetic fertilizers with manure fertilizers increased phosphatase and catalase activities but decreased urease activity [19]. Within the winter wheat–summer maize rotation system, a combination of 25% mineral N fertilizer and 75% organic N fertilizer increased soil enzyme activity and decreased residual nitrate-nitrogen levels [20]. However, different crops respond differently to the substitution of manure with chemical fertilizers. Guo et al. [14] demonstrated that an organic fertilizer substitution of 22–36% of chemical fertilizers could effectively prevent the acidification of red soil. Shu et al. [21] discovered that substituting 30% synthetic fertilizer with manure fertilizer not only maximized tea yield but also enhanced the quality of tea. Similarly, Mei et al. [22] established that substituting 70% of the N content in the NPK treatment with N derived from cow manure was the most favorable strategy for increasing maize yield. Compared to other agroecological systems, soil fertility and yield enhancement within rubber plantation ecosystems exhibit different responses to fertilization. This is attributed to the fact that rubber trees are arbor, and the regular tapping process results in significant nutrient loss, thereby increasing the demand for nutrients. Although previous studies have shown that organic substitution fertilization can effectively increase rubber yield, Lu et al. [6] indicated that POS fertilization significantly increased the latex yield by 11.8–16.4%, compared with application of chemical fertilizer alone. However, this study did not evaluate the response of soil quality to fertilization. Therefore, it is essential to examine the mechanisms through which different fertilizers affect latex production by influencing soil nutrient and enzyme activities.

Hainan Island, the second-largest rubber plantation region in China, has rubber plantations encompassing over 5.4 × 10⁵ ha, accounting for approximately 25% of the island’s total land area [23]. Most rubber plantations in Hainan are second-generation. Rubber trees demand higher soil nutrient levels than many other agricultural systems because of the substantial nutrient depletion from latex production, and long-term rubber cultivation has degraded soil fertility. Although partial organic substitution fertilization has been used in rubber plantations in recent years [16,24,25], the mechanisms by which organic substitution influences soil fertility, tree growth, and latex yield remain unclear. Therefore, this study conducted a two-year field experiment to investigate the comprehensive effects of partial organic substitution fertilization on both latex yield and soil quality in a rubber plantation on Hainan Island. This study selected partial organic substitution fertilization levels (25, 50, 75, and 100%) based on the following considerations: (1) manure releases nutrients slower than chemical fertilizers, with 25–50% substitution ensuring nutrient availability while improving soil structure; (2) higher substitution levels (75–100%) allow us to assess manure’s potential to meet crop requirements, evaluate the effects on soil carbon sequestration, and risk nitrogen immobilization; and (3) following China’s “Zero Growth Action Plan,” a 30–50% substitution optimizes productivity, while the 25–100% gradient helps identify optimal rates for specific crops and soils, reducing pollution risks and guiding nutrient management. This study aimed to (1) evaluate the effects of partial organic substitution fertilization on soil fertility and latex yield, (2) identify the key soil properties that influence latex yield, and (3) develop optimal partial organic substitution fertilization strategies to support the sustainable development of the natural rubber plantation industry in Hainan, China.

2. Materials and Methods

2.1. Study Site

The experimental site was located on farmland in Zhubijiang (19°19′42″–19°24′03″ N, 109°15′43″–109°19′49″ E, 114 m a.s.l.) in Baisha, Hainan Province, China. This area has a tropical monsoon climate, and the year is divided into the rainy season (from May to October) and the dry season (from November to April of the following year). The annual average temperature is 23.4 °C and the average humidity is 82%. Precipitation averages 1966.3 mm annually, with the majority falling from May to October, accounting for 85% of the rainfall. The soil type was Udic Ferrallisols derived from granites, which are classified as Oxisols in the USA Soil Taxonomy System. The initial soil properties (0–20 cm) in 2020 were as follows: soil organic matter of 24.29 g kg⁻¹, total nitrogen (TN) of 1.82 g kg⁻¹, total phosphorus (TP) of 0.35 g kg⁻¹, ammonium nitrogen (NH₄⁺-N) of 16.55 mg kg⁻¹, nitrate nitrogen (NO₃⁻-N) of 3.46 mg kg⁻¹, and soil pH of 4.87.

2.2. Experimental Design

The study site was a mature rubber plantation (480 plants ha⁻¹, clone PR-107) established in 2004 with tapping commencing in 2011. In April 2020, an organic substitution fertilization field trial was carried out, encompassing 18 experimental plots consisting of 360 rubber trees, including six treatments replicated three times: (1) CK: no fertilization, (2) NPK: conventional synthetic fertilizer application, (3) 25M: 25% manure N substitution for synthetic N fertilizer, (4) 50M: 50% manure N substitution for synthetic N fertilizer, (5) 75M: 75% manure N substitution for synthetic N fertilizer, and (6) 100M: 100% manure N substitution for synthetic N fertilizer. Adhering to local fertilizer practices, a total of 2.0 kg·tree⁻¹·year⁻¹ of synthetic fertilizer was applied, containing 14% N (urea-46.3% N), 9% P (superphosphate-16% P₂O₅), and 7% K (potassium chloride-60% K₂O) fertilizers, up to an application rate of 280.0 g N, 140.0 g P₂O₅, and 180.0 g K₂O tree⁻¹·year⁻¹. It is important to note that all fertilization treatments (including NPK, 25M, 50M, 75M, and 100M) had the same overall N application. Furthermore, the POS treatments (25M, 50M, 75M, and 100M) achieved the same levels of P and K as the NPK treatment by supplementing with super phosphate and potassium chloride, respectively. The mean nutrient concentrations of C, N, P, and K in cow manure used in this study were 21.7%, 1.8%, 0.4%, and 0.3%, respectively. The N, P₂O₅, and K₂O contents of NPK were consistent across the 25M, 50M, 75M, and 100M treatments (Table S1). In accordance with the nutrient requirements for rubber tree growth and the characteristics of fertilizer decomposition, cow manure was used as the basal fertilizer in January, and chemical fertilizers were applied as top dressings in April, July, and September at a ratio of 5:3:2.

2.3. Sampling and Analyses

2.3.1. Rubber Yield

Daily fresh latex from each replicate per treatment was weighed every tapping day (every 3–5 days) from April to December each year. These data were then used to calculate the cumulative latex yield of a single rubber tree under different fertilization treatments (kg·tree⁻¹·year⁻¹).

2.3.2. Sampling and Analysis of Leaves

The leaves of rubber trees were collected in September 2020 and 2021. From each treatment, 12–15 rubber trees were selected, and the second and third leaves were collected from the terminal whorl of branches in the canopy of each rubber tree. Thirty healthy leaves were selected and mixed to create a composite sample for each treatment. Fresh leaves were brought to the laboratory and dried at 105 °C for 30 min and then at 65 °C for 48 h until a constant weight was reached, after which the nutrient concentrations were determined [9]. C, N, and P concentrations were determined using the dichromate oxidation, Kjeldahl digestion, and molybdovanadate methods, respectively [26].

2.3.3. Soil Sampling and Analyses

Soil samples were collected at the end of each year’s rubber-tapping season (December 2020 and 2021). Samples from all the plots were collected at a depth of 20 cm using a five-point sampling method. Fresh soil samples were promptly transferred to the laboratory and divided into two parts. One part was refrigerated at 4 °C to analyze soil enzyme activities and mineral nitrogen concentrations. The second portion was air-dried, ground, sieved to 100 mesh, and analyzed for other soil properties. Mineral N (NO₃⁻-N and NH₄⁺-N) was extracted using a 2 M KCl solution and measured using a continuous flow injection analyzer (Proxima1022/1/1, Alliance Scientific Instruments, Frépillon, France). The soil pH was determined using a pH electrode (m:v = 1:2.5). SOC was quantified using the K₂Cr₂O₇-H₂SO₄ oxidation method [26]. TN was extracted using the semi-micro Kjeldahl method, and TP was autoclaved with H₂O₂ and H₂SO₄-HClO₄ and determined using a continuous flow analyzer [9,26].

The activities of soil acid phosphatase (AP), leucine aminopeptidase (LAP), β-1,4-N-acetylglucosaminnidase (NAG), and β-1,4-glucosidase (BG) were fluorometrically analyzed in polystyrene 96-well, 300-µL microplates by evaluating the fluorescence at 360 nm excitation and 450 nm emission using a microplate reader (Synergy, Epoch™ 2, BioTek Instruments, Inc., Winooski, VT, USA), as outlined in DeForest’s [27] methodology.

2.4. Soil Quality Measurement

In this study, overall soil health and fertility were analyzed by calculating soil quality. Soil quality is commonly assessed through the integrated measurement of the physical, chemical, and biological properties that govern soil functions and environmental sensitivity. The soil quality index (SQI) is the preferred method owing to its adaptability and ease of implementation. Therefore, we selected soil pH, NH₄⁺-N, NO₃^—-N, SOC, TN, TP, AP, BG, NAG, and LAP as the total dataset to evaluate the effects of various fertilization methods on the soil quality index. Each soil indicator was transformed into a dimensionless score (S_i) between zero and one during the calculation process, with higher scores indicating better results.

In this study, all the indicators were evaluated using Equation (1), which assessed ‘more is better’ outcomes. The equations provided by He et al. [28] were applied to measure the different indicators and determine their impact on soil quality.

$${\mathrm{S}}_{\mathrm{t}}={\displaystyle \frac{{\mathrm{x}}_{\mathrm{t}}-{\mathrm{x}}_{\mathrm{m}\mathrm{i}\mathrm{n}}}{{\mathrm{x}}_{\mathrm{m}\mathrm{a}\mathrm{x}}-{\mathrm{x}}_{\mathrm{m}\mathrm{i}\mathrm{n}}}}$$

(1)

$${\mathrm{S}}_{\mathrm{t}}={\displaystyle \frac{{\mathrm{x}}_{\mathrm{m}\mathrm{a}\mathrm{x}}-{\mathrm{x}}_{\mathrm{t}}}{{\mathrm{x}}_{\mathrm{m}\mathrm{a}\mathrm{x}}-{\mathrm{x}}_{\mathrm{m}\mathrm{i}\mathrm{n}}}}$$

(2)

where S_t represents the score of the tth indicator, and x_t, x_max, and x_min correspond to the measured, maximum, and minimum values of the tth indicator, respectively.

The weight of each indicator (W_t) was determined by analyzing the ratio of the tth indicator’s commonality to the total commonalities of all indicators using principal component analysis [29]:

$$W_t={\displaystyle \frac{C_t}{\sum _{t=1}^n{C}_t}}$$

(3)

SQI was calculated by multiplying the weighted values by the scored values [29]:

$$\mathrm{S}\mathrm{Q}\mathrm{I}=\sum _{\mathrm{t}=1}^{\mathrm{n}}{\mathrm{W}}_{\mathrm{t}}{\mathrm{S}}_{\mathrm{t}}$$

(4)

where n represents the number of indicators, and S_i and W_i denote the rating and importance of the ith indicator, respectively.

2.5. Statistical Analyses

Before analyzing variance (ANOVA), the data were assessed for normal distribution using the Shapiro–Wilk test. Non-normal data were transformed using natural logarithms. One-way analysis of variance (ANOVA) and independent sample t-tests were used to compare soil properties, enzyme activities, leaf nutrients, and yields across different treatments within the same year and the same treatment over different years. The results were considered significant at p < 0.05.

Furthermore, a two-way ANOVA analysis of variance was conducted to determine the effect of fertilization treatment and sample year on soil properties, leaf nutrients, and latex yield. The corrupt package in R software (Version 4.0.4) was used to analyze the relationship between soil characteristics, leaf nutrients, and latex production using Pearson’s correlation coefficients and a random forest model. Statistical analyses were conducted using SPSS 26, and graphs were generated using the Origin 2021 software and R 4.0.4.

3. Results

3.1. Rubber Latex Yield

During the two-year study period (2020–2021), all fertilization treatments consistently enhanced rubber latex yield compared to the control (CK), demonstrating a yield increase of 2.92–28.09% (Figure 1). The most effective treatments varied by year, with 75M and 100M emerging as the optimal treatments for 2020 and 2021, respectively. When compared with conventional NPK fertilization, the organic substitution fertilization treatments (25M, 50M, 75M, and 100M) showed differential effectiveness: in 2020, yield enhancements ranginged from 2.10% to 12.89%, whereas 2021 saw improvements between 11.77% and 18.60%. Notably, ANOVA indicated a statistically significant effect of temporal factors on latex yield (p < 0.01), while the fertilization treatment (p = 0.185) and the interaction between fertilization treatment and temporal factors (p = 0.611) were deemed insignificant.

3.2. Leaf Nutrients

The leaf C concentrations exhibited interannual variations across fertilization treatment, measuring 441.76–478.20 g·kg⁻¹. Although the values in 2021 showed significant elevation compared to those in 2020, no treatment differences reached statistical significance (

Table 1. Leaf C, N, and P concentrations and their stoichiometry in different fertilization treatments in 2020 and 2021.

Year	Fertilization	Leaf C	Leaf N	Leaf P	Leaf C:N Ratio	Leaf C:P Ratio	Leaf N:P Ratio
		(g·kg−1)
2020	CK	441.76 ± 5.19Ba	13.37 ± 0.38Cb	1.63 ± 0.12Ba	33.07 ± 0.71Aa	272.06 ± 21.56Aa	8.23 ± 0.62Ca
	NPK	445.00 ± 10.84ABb	18.28 ± 0.37Bb	1.76 ± 0.03ABb	24.34 ± 0.37Ba	252.49 ± 9.19ABa	10.37 ± 0.32ABa
	25M	457.00 ± 8.24Ab	18.60 ± 0.61Bb	1.89 ± 0.32Ab	24.59 ± 1.01Ba	248.51 ± 48.84ABa	10.14 ± 2.15ABa
	50M	452.71 ± 7.22ABb	18.50 ± 0.48Bb	1.85 ± 0.09ABb	24.48 ± 0.91Ba	245.69 ± 14.52ABa	10.04 ± 0.67ABa
	75M	456.09 ± 9.76Aa	18.72 ± 0.40Bb	1.99 ± 0.07Ab	24.37 ± 0.65Ba	230.06 ± 12.09Ba	9.44 ± 0.44BCa
	100M	446.51 ± 9.44ABb	20.57 ± 0.10Ab	1.83 ± 0.12ABb	21.71 ± 0.52Ca	244.50 ± 17.19ABa	11.26 ± 0.71Aa
2021	CK	447.43 ± 17.35Ba	14.35 ± 0.14Da	1.65 ± 0.02Da	31.18 ± 1.49Ab	270.80 ± 13.80Aa	8.68 ± 0.07Ba
	NPK	466.72 ± 8.47Aa	20.62 ± 0.13Ba	1.96 ± 0.04Ca	22.63 ± 0.51Bb	238.66 ± 5.08Bb	10.55 ± 0.25Aa
	25M	470.79 ± 9.53Aa	20.81 ± 0.04Ba	2.02 ± 0.02BCa	22.62 ± 0.48Bb	232.70 ± 6.61Ba	10.28 ± 0.10Aa
	50M	475.73 ± 13.71Aa	20.80 ± 0.06Ba	1.98 ± 0.13BCa	22.87 ± 0.71Bb	240.60 ± 15.91Ba	10.53 ± 0.74Aa
	75M	478.20 ± 19.20Aa	20.79 ± 0.08Ba	2.45 ± 0.04Aa	23.15 ± 0.92Bb	194.91 ± 8.76Cb	8.42 ± 0.15Bb
	100M	468.86 ± 9.54Aa	21.09 ± 0.13Aa	2.05 ± 0.02Ba	22.23 ± 0.37Ba	228.99 ± 4.60Ba	10.30 ± 0.07Ab
ANOVA results (F values)
Fertilization		5.178 **	671.536 **	24.529 **	218.523 **	10.176 **	15.082 **
Year		37.844 **	474.731 **	40.907 **	42.010 **	8.944 **	0.382 ns
Fertilization × Year		0.941 ns	16.050 **	4.105 **	3.589 **	0.991 ns	2.055 ns

Values in the table represent each sample plot (mean ± standard errors). CK, no fertilization; NPK, chemical fertilizer application; 25M, 25% manure N substitution for synthetic N fertilizer; 50M, 50% manure N substitution for synthetic N fertilizer; 75M, 75% manure N substitution for synthetic N fertilizer; 100M, 100% manure N substitution for synthetic N fertilizer; Leaf C, carbon concentration of rubber leaf; Leaf N, nitrogen concentration of rubber leaf; Leaf P, phosphorus concentration of rubber leaf. Different uppercase letters indicate significant differences among different treatments in the same year (p < 0.05). Different lowercase letters indicate significant differences between the two years (p < 0.05); ** p < 0.01, ns p > 0.05.

). Leaf N dynamics displayed 12.93–21.23 g·kg⁻¹; the 100M treatment significantly enhanced leaf N concentrations compared to the NPK in both years, with increases of 3.94–12.59% and 1.41–25.46% for P concentrations in 2020 and 2021, respectively, compared to NPK. Notably, organic fertilizer substitution consistently reduced the stoichiometric ratios (C:N, C:P, and N:P ratios) across both years, although these reductions were not statistically significant (Table 1).

The ANOVA results revealed significant effects of fertilization and sampling year on leaf C, N, and P concentrations and C:N and C:P ratios (p < 0.01). Fertilization had a significant influence on the N:P ratio, whereas year and fertilization interactions significantly affected N and P concentrations (p < 0.05). However, significant interactive effects were detected for C concentration, C:P ratio, and N:P ratio (p > 0.05) (Table 1).

3.3. Soil Chemical Properties and Enzyme Activities

The fertilization regimes significantly improved soil nutrient availability across both experimental years, with all treated plots demonstrating markedly higher total and available nutrient concentrations than the CK (Figure 2). In 2020, partial organic substitution fertilization reduced NO₃^–-N by 61.16–81.11% relative to NPK fertilization, while simultaneously increasing the pH (3.38–7.41%), SOC (2.54–23.80%), TN (2.69–13.39%), and TP (7.65–16.24%). This trend intensified in 2021, with organic fertilizer substitution treatments significantly increased the SOC, pH value, NH₄⁺-N, NO₃^–-N, TN, TP, and C: N, N: P, and C:P ratios (except for 25M) by 1.92–5.38%, 5.24–66.23%, 14.41–47.52%, 22.87–35.51%, 0.47–28.18%, 5.81–18.71%, 2.18–22.21%, 9.94–27.15%, and 8.86–16.58%, respectively. ANOVA confirmed the significant effects of fertilization strategy, interannual variation, and their interaction on soil nutrient factors (p < 0.05) (Figure 2).

The partial organic substitution fertilizer strategy consistently enhanced soil enzymatic functionality across both experimental years compared to the NPK treatment (Figure 3). Specifically, activities of AP, BG, NAG, and LAP were increased by 2.30–13.11%, 5.48–19.00%, 9.81–18.22%, and 3.96–21.69% in 2020, and by 10.16–22.20%, 6.28–19.62%, 7.38–17.67%, and 3.90–13.87% in 2021, respectively. Fertilization strategy and interannual variation significantly affected soil enzyme activity (p < 0.01). There was a significant interaction between fertilization and year for AP, BG, and NAG (p < 0.01) but not for LAP activity (p = 0.063).

3.4. Soil Quality Evaluation and Correlation with Rubber Latex Yield

Through calculations (Table S2), it was found that the SQI varied from 0.20 to 0.71 across the two years, with higher values in 2021 than in 2020 (Figure 4a,b). All fertilization treatments improved the SQI compared to CK, showing interannual variation, and partial organic substitution treatments performed better than NPK fertilization. Compared with the NPK treatment, the SQI of the 25M, 50M, 75M, and 100M treatments significantly increased by 15.30%, 39.65%, 27.33%, and 27.76% in 2020, and by 18.60%, 37.54%, 43.42%, and 40.79% in 2021, respectively. These improvements resulted from enhanced soil nutrient availability and enzymatic activity, and the organic matter-added treatments generally had more balanced and relatively higher values for some key properties, such as SOC and TN, indicating improved soil fertility. The influence of normalized quality indicators on the SQI further supports these findings. Radar charts illustrating indicator membership degrees directly reflect individual fertility levels, where higher average membership values signify better indicator fertility. Key soil quality parameters, such as SOC, TN, NO₃⁻-N, and enzyme activities, show significantly greater contributions to the quality index in POS treatments compared to NPK. This confirms that POS fertilization effectively improves soil quality in rubber plantations (Figure 4c,d). A higher SQI indicates better soil quality, providing suitable conditions for plant growth through nutrient supply, good structure for root development, and proper moisture and aeration. Therefore, a positive correlation between SQI and yield was observed, with a stronger correlation in 2021, likely due to the cumulative effects of fertilization optimizing the soil-plant relationship (Figure 4e,f).

3.5. Relationships of Rubber Leaf Nutrient and Latex Yield with Soil Variables

Correlation analysis revealed significant positive associations between leaf nutrient status and both SOC and TN levels, as well as between latex yield and multiple soil chemical parameters (excluding NH₄⁺-N and soil N:P) and enzymatic activity (Mantel’s r > 0.25, p < 0.01) (Figure 5a,b). In interannual variations, leaf nutrients demonstrated particularly strong correlations with TN content in 2020, whereas 2021 exhibited significant positive relationships between both leaf nutrients and latex yield with NO₃^–-N, SOC, TN, and soil stoichiometric ratios (N:P and soil C:P) (Mantel’s r > 0.25, p < 0.01). Among the experimental treatments, the principal determinants of latex yield were NO₃^–-N concentration, pH, TN, SOC, and soil C:P ratio (Figure S1).

Random forest regression modeling demonstrated that soil and foliar N and P dynamics were pivotal regulators of latex yield (p < 0.05; Figure 5c,d). Interannual analysis indicated that latex yield was predominantly influenced by leaf P content, soil C:P ratio, foliar N concentration, LAP activity, leaf C:P ratio, soil C:N ratio, and NO₃^–-N levels in 2020 (p < 0.05, Figure 5c), whereas 2021 production was primarily governed by TP content, leaf C:N ratio, TN concentration, NH₄⁺-N levels, soil N:P ratio, foliar N status, and NO₃^–-N availability (p < 0.05, Figure 5d).

Structural equation modeling (SEM) analysis further indicated that the application of fertilization had a significant positive impact on the overall quality of the soil (p < 0.05). Meanwhile, the fertilization treatment indirectly improved soil quality by positively influencing soil enzyme activities and chemical properties (p < 0.001). Furthermore, fertilization resulted in an increase in latex yield by improving the soil quality (Figure 6a). Standardized total effects analysis revealed that the most important direct effects on latex yield were attributed to soil chemical properties (0.96) and soil quality (0.86), whereas indirect effects were influenced by soil properties (0.86) and fertilization regimes (0.76) (Figure 6b).

4. Discussion

4.1. Effect of the Soil Property, Enzyme Activity, and Soil Quality Under Organic Substitution Fertilization

Reducing chemical fertilizer use in agriculture requires a comprehensive assessment of its impact on crop yield and soil fertility [28]. The current study indicated that partial organic substitution (POS) significantly enhances soil fertility by improving pH, nutrient availability, and enzyme activity, compared with NPK treatment (Figure 2), consistent with previous studies [30,31]. For instance, Li et al. [30] demonstrated that partial organic substitution enhanced soil fertility in a wheat–rice rotation system over seven years. In this study, POS significantly increased SOC by 2.54–35.51% compared with NPK (Figure 2). This can be attributed to combined organic-chemical fertilization enhancing soil aggregation, promoting organic carbon accumulation within macroaggregates, and improving physicochemical protection of SOC [32,33,34]. Concurrently, recalcitrant organic components (e.g., humus, lignin, and polyphenols) undergo humification to form stable aromatic and alkyl structures that resist microbial degradation and enzymatic decomposition in the soil and enhance SOC stability [10]. The high-alkalinity ash present in the organic matter helped neutralize the protons produced during N nitrification and stimulated the process of reducing nitrates through denitrification, ultimately leading to an increase in pH values [35,36]. Additionally, the introduction of organic substitutions can stimulate soil microbial function and enhance microecological conditions [37], reducing N loss through denitrification and leaching [31] and thereby improving N concentration. Microorganisms proliferate in response to manure application, significantly contributing to the regulation of SOM dynamics and enhancement of soil carbon stocks [38,39]. Gross and Glaser [34] reported an average increase of 35% in SOC following manure application.

Enzymes are vital indicators of soil health and nutrient cycling [26]. In this study, POS consistently enhanced soil enzymatic activity across both experimental years compared to the NPK treatment (Figure 3). This might be because manure provides energy and nutrients for soil microbial growth, which leads to the synthesis and release of more enzymes into the soil [9,19]. Moreover, microorganisms can efficiently allocate energy and nutrients for growth by adjusting the expression of enzymes to match available resources [40,41,42], which allows them to effectively obtain the necessary resources in environments with limited resources. When faced with an excess of environmental substrates but limited access to essential resources, microorganisms prioritize the production of key enzymes necessary for acquiring limited resources. However, while organic inputs boost enzyme function in nutrient-poor soils [43,44], there is a threshold beyond which added nutrients no longer increase enzyme activity [40], which is consistent with our observation that the enzyme activities at 100M were lower than those at 75M and 50M (especially in 2021) (Figure 3). The observed phenomenon can be attributed to the C:N:P activity ratio of various ecological enzymes, which exhibits a consistent functional organization pattern. This pattern represents the threshold at which microbial communities respond to disturbances or fluctuations in nutrient availability; beyond this point, the activity does not continue to increase or decrease with changes in nutrient availability [45,46,47]. For instance, Kang et al. [45] discovered that elevated nitrogen addition results in a reduction in the activity of key carbon and nitrogen hydrolases. Similarly, Wang et al. [47] reported that the combined application of manure and chemical fertilizers did not significantly enhance the activities of soil C, N, and P cycling-related enzymes but rather decreased the activity of soil LAP. These findings support the results of the present study.

The effect of fertilization on soil quality can be assessed using an SQI that combines multiple soil properties. In this study, POS treatment significantly improved soil quality compared to the chemical fertilizer treatment by enhancing most soil indicators (Figure 4). This finding aligns with earlier studies [13,37,48]. Organic matter directly supplied bioavailable carbon and nitrogen, elevating soil nutrient levels and availability, thereby establishing a more conductive soil environment within the rubber plantation (Figure 2, Figure 3 and Figure 4a,b). Enhanced plant growth increases root exudates and litter inputs, improves soil structure, and stimulates microbial biomass and metabolite production. The formation of organo-mineral complexes increases soil colloidal content, thereby augmenting nutrient retention capacity [28,49]. These synergistic mechanisms collectively enhance soil quality compared with synthetic fertilizers [37,50]. Furthermore, the rate of manure substitution exhibited a positive correlation with SQI, suggesting that partial organic substitution as a fertilization strategy in rubber plantations represents a promising approach to improving soil quality and reducing dependence on chemical fertilizers.

4.2. Effect of the Leaf Nutrient Under Organic Substitution Fertilization

The nutrient composition of leaves serves as a crucial indicator of the nutritional status of forest trees and is commonly used to evaluate the effectiveness of fertilizers and guide fertilization practices [51]. In this study, leaf C, N, and P concentrations were 441.76–478.20 g·kg⁻¹, 13.37–21.09 g·kg⁻¹, and 1.63–2.45 g·kg⁻¹, respectively (Table 1). With the increase in the number of years of fertilization, the nutrient content in the leaves rises; nevertheless, these values remained below the established optimum ranges for rubber tree growth [52]. Huang and He [52] state that when leaf nitrogen content ranges from 3.2% to 3.4% (dry weight), P content is 0.21–0.23%) (dry weight), and the N:P ratio is between 14.8 and 15.2, the rubber tree’s nitrogen status is normal and latex production capacity robust. This is likely attributable to the inherently low soil fertility at the research site, which is further aggravated by nutrient depletion resulting from frequent latex tapping (Figure 1 and Figure 2) [7,53]. Despite the application of fertilizers, the soil within rubber plantations remains deficient in essential nutrients. Lin et al. [54] studied rubber tree leaves over 15 years in Hainan, showing that 50% of samples had N contents below the normal range, while P contents were mostly within the normal range. This temporal variation is also attributed to the inherently slow decomposition and nutrient mineralization rates of organic manure [55]. Consequently, only a limited fraction of essential nutrients (e.g., C, N, and P) becomes available to trees during the initial application years [43,56,57]. Zhu et al. [58] measured the decomposition of three animal manure types (cattle, sheep, and goat) over a year at four diverse sites across tropical regions, and the findings indicated slower decomposition rates in tropical environments than in temperate climates. Compared with NPK, POS fertilization demonstrated superior efficacy in elevating nutrient concentrations in rubber tree leaves (Table 1; Figure 1). This suggests that organic amendments facilitate a more coordinated nutrient supply, thereby improving the nutrient uptake efficiency in rubber trees. These findings are consistent with those of the previous studies [24,54]. Moreover, the consistent reduction in stoichiometric ratios (C:N, C:P, and N:P ratios) with organic fertilizer substitution indicates a potential shift in nutrient balance within rubber tree leaves. For example, a lower C:N ratio may suggest enhanced nitrogen availability and utilization, which can promote protein synthesis and other physiological processes related to latex production. However, the lack of statistical significance may be due to the relatively small sample size or the high variability within the treatments. Further research with larger sample sizes and more precise analytical methods is needed to confirm these findings.

Additionally, significant correlations between soil properties and leaf nutrients were observed, reflecting the role of soil as the principal nutrient reservoir. For instance, N is fundamental for plant growth and metabolism, governs physiological processes, and, ultimately, latex yield in rubber trees. The study identified a significant positive correlation between soil TN, NH₄⁺-N, and NO₃⁻-N levels and leaf N content. Similarly, SOC enhances soil structure and improves water and nutrient retention capacity. These improvements indirectly promote plant nutrient uptake efficiency, thereby correlating with an enhanced leaf nutrient status. Phosphorus, vital for energy metabolism and bio-membrane synthesis, plays a pivotal role in rubber tree physiology and latex biosynthesis; thus, adequate phosphorus supply is a crucial nutrient balance for rubber trees [59,60]. As a result, the application of organic substitution fertilization, particularly at higher proportions of organic fertilizer (75M and 100M), significantly improved nutrient availability in both the soil and foliage.

4.3. Effect of Soil Quality on Latex Yield Under Organic Substitution Fertilization

Soil quality is a fundamental determinant of crop yield, emphasizing the critical role of optimized fertilization practices in enhancing both soil health and agricultural productivity. In this study, a significant positive correlation (p < 0.05) was observed between SQI and rubber latex production in 2021 (Figure 5d). This finding corroborates previous research, indicating that soil quality improvements can enhance rubber latex yield [22,28,57]. It is noteworthy that this correlation was not statistically significant in 2020 (Figure 5c) because organic fertilizers release nutrients gradually and over an extended period. This temporal pattern aligns with the observations of Tiva et al. [56], who reported no significant first-year yield differences under varying chemical fertilizer regimes. Therefore, we recommend that future investigations evaluating the effects of organic fertilization on rubber tree growth and latex production should incorporate a minimum duration of two years to capture potential temporal dynamics. Random forest regression modeling also showed that soil NO₃⁻-N, pH, and total nutrient pools (SOC, TN, and TP), along with their stoichiometric ratios (C:N:P), were identified as critical determinants of latex yield. This might be because soil-available N, through governing plant growth and metabolism, ultimately affects latex synthesis in rubber trees, whereas soil organic fertilizer enhances soil aggregation and modulates soil structure and nutrient retention, thereby indirectly regulating plant nutrient uptake and influencing latex yield [61,62]. These mechanisms are consistent with a substantial body of evidence confirming that organic fertilizer application enhances soil nutrient availability, promotes soil quality, and increases crop yield [28,37].

Compared with CK, fertilization regimes significantly increased rubber latex yield by 2.92–28.09%. These findings align with the established literature reporting yield improvements of 10–20% following fertilizer application in rubber plantations [5,56,63]. Notably, POS treatments (25M–100M) elevated latex yield by 2.33–13.10% (2020) and 10.85–18.60% (2021) compared with NPK (Figure 1). This aligns with studies documenting superior yields and soil outcomes from integrated organic-chemical approaches over chemical fertilization alone [28,58]. For instance, in 2020, the levels of SOC, TN, TP, and NH₄⁺-N in the 75M and 100M treatments were lower than those in the 50M treatment. Conversely, the NO₃⁻-N content in the NPK treatment was significantly higher than that observed in all organic substitution fertilizer treatments. The significant second-year yield advantage under organic substitution suggests that organic amendments progressively enhance production systems, likely through sustained nutrient release and improved soil functionality. Collectively, these results suggest that organic fertilizers are viable alternatives to conventional NPK regimens, offering comparable or enhanced latex yields, while supporting sustainable nutrient management practices in rubber cultivation.

Notably, the optimal organic fertilizer substitution ratio for maximizing latex yield varies annually, with 75M maximizing yield in 2020 versus 100M in 2021. First, organic fertilizers require microbial decomposition for gradual nutrient release, whereas chemical fertilizers provide immediate nutrient supply. In 2020 (initial stage), the 75M blend leveraged this synergy: chemical inputs met urgent rubber tree needs, whereas organic amendments slowly improved soil aggregate structure, preventing compaction and nutrient loss from pure chemical use [33,58]. By 2021, one year of organic input enhanced soil organic matter, microbial activity, and water/nutrient retention. At this stage, 100M ensured steady nutrient release while avoiding chemical-induced acidification, aligning better with long-term post-soil improvement needs. Second, accumulated organic matter further optimizes soil structure and nutrient retention by 2021. Through continuous decomposition, 100M supplied sufficient nutrients, while improved soil conditions boosted root growth (e.g., expanded absorption area), supporting higher yields [53]. Therefore, time-dependent synergies between organic and chemical fertilizers coupled with dynamic shifts in soil quality and crop nutrient demands drive annual adjustments to optimal fertilization strategies. Additionally, rubber tree nutrient requirements for growth and latex synthesis evolve with age and growth stages, modulated by environmental factors (precipitation, temperature, and soil moisture gradients) that indirectly affect fertilization efficacy via nutrient availability and root uptake [50,63]. Long-term studies are thus needed to quantify the sustained impacts of organic substitution on soil quality and latex yield, and to define optimal ratios for sustainable rubber cultivation in the study area.

It is worth noting that soil quality improvement and latex yield enhancement were not strictly proportional to the quantity of organic fertilizer substitution. Compared with 50M, latex yields showed no significant differences in the 75M and 100M treatments, suggesting a plateau effect. Beyond a certain threshold, diminishing returns are likely to occur, where additional fertilizer contributes only marginal gains in productivity or soil health. This may result from nutrient saturation, microbial activity constraints, or disruption of the soil chemical equilibrium. Moreover, from a logistical and economic perspective, excessive organic fertilizer use escalates costs (including transportation, labor, and application) without proportional yield improvements, thereby diminishing cost efficiency. Bulk organic materials pose storage and handling challenges that further complicate their large-scale adoption. Therefore, after comprehensively balancing agronomic gains with economic and operational feasibility, we identified that a 50% organic fertilizer substitution ratio (50M) is recommended as the optimal strategy for enhancing productivity and soil health in rubber plantations on Hainan Island. Nonetheless, it is important to note that the effects of fertilization on soil quality and crop yield manifest in the long term. Therefore, we will conduct similar experiments in different regions and establish long-term field trials to monitor and evaluate the long-term effects of different fertilization methods on soil quality and latex yield under different climatic and soil conditions, with the ultimate goal of assessing the generalizability of our findings.

5. Conclusions

This two-year study demonstrated that partial organic substitution fertilization in rubber plantations increased soil quality (e.g., soil TN, SOC, TP, NO₃⁻-N, NH₄⁺-N concentrations, and enzyme activities) and enhanced rubber tree growth, thereby promoting rubber latex yield. Integrated organic substitution fertilization mitigated soil acidification and enhanced short-term latex yield. Soil NO₃⁻-N, SOC, C:P ratio, pH, and TN were identified as critical soil parameters governing soil fertility and latex productivity. Among the partial organic substitution fertilization regimes, organic fertilizer substitution rates exceeding 50% (50M, 75M, and 100M) significantly enhanced soil quality and yield production. However, soil quality improvement and latex yield enhancement were not strictly proportional to the quantity of organic fertilizer substitution. Therefore, after comprehensively balancing agronomic gains with economic and operational feasibility, we identified that a 50% organic fertilizer substitution ratio (50M) is recommended as the optimal strategy for enhancing productivity and soil health in rubber plantations on Hainan Island. Given the perennial nature of rubber trees and the lagged effects of fertilization, future research should establish long-term optimal substitution ratios by correlating organic substitution regimes with latex physiological parameters (e.g., bark thickness, laticifer density, latex flow, and dry rubber content).