Long-term changes in soil biological activity and other properties of raised beds in Longan orchards

Site description

Longan growing areas in Vietnam are concentrated in a number of provinces in the North and Mekong River Delta provinces. Statistical data in 2017 show the whole country had about 73.3 thousand ha planted to trees with production estimated at 550,000 t year⁻¹ (Hung et al., 2019). In the Mekong River Delta provinces, the total Longan growing area is 24,913 ha, achieving a productivity of 227,624 tons, and accounting for 40% of the total yield of the country. Of these, four regions have the largest Longan growing areas in descending order as follows Vinh Long (6,129 ha), Dong Thap (5,515 ha), Soc Trang (3,552 ha), and Can Tho (2,512 ha) (General Statistics, Office of Vietnam, 2020). The most common productive cultivars grown in the Mekong River Delta provinces include “Tieu da bo”, “Xuong com vang”, and “Xuong com rao” with a mean productivity of 10 t year⁻¹. Traditional production systems use intensive and conventional chemical farming techniques to increase fruit production to 20–25 t year⁻¹. Further, off-season flowering treatment techniques to control harvesting time are regularly used in the Longan growing areas in the Mekong River Delta (Hung et al., 2019).

The study area is located in An Binh commune, Long Ho district, Vinh Long province (10°17′37.5″N 105°58′55.7″E). This region has a remarkable historical background for Longan cultivation in Vinh Long and in the VMD as well. The study site has an altitude 0.6−1.2 m above sea level; receiving 1,450–1,504 mm year⁻¹ annual precipitation with a flooding season between September and November (Thanh et al., 2019). The Mekong River’s influence during the Holocene retrograde marine sedimentation processes left behind marine sandy soils and saline soils from saltwater intrusion (Ta et al., 2021). The dominant climate in the region has been defined as tropical climate with monsoon. Twenty soil samples from different ages of Longan orchard beds cultivated longer than 15 years were collected and analyzed. Information regarding the age of the orchards and bed ages are shown in Table S1.

Soil sample collection

Soil samples of Longan orchards in Vinh long province, Vietnam were collected in July, 2023 after the final Longan harvest with a total of 20 soil samples from four different groups of Longan orchards based on age of the raised bed. Samples were grouped by the age of the raised bed: Group 1) 15–25-year-old Longan orchards (L1–L5); group 2) 26-37-year-old Longan orchards (L6–L10); group 3) 38–45-year-old Longan orchard (L11–L15); and group 4) 46–60-year-old Longan orchards (L16–L20). Each raised bed age group consisted of soil drawn from five different Longan orchard farms corresponding to five different soil samples. Due to budget limitations, only 20 soil samples were collected for this study. Twenty is considered a sufficiently large enough sample to be representative of the area of study area. At each Longan orchard farm, soil samples were collected using soil auger (Seymour 21306 AU-S6, USA) from the upper layer depth of 0–20 cm, under the canopy of Longan trees from 10 different spots within each Longan farm. The soil was combined to constitute one soil sample from several locations in a zig-zag pattern for homogeneity and kept in clean nylon bags. Then the samples were transported in hard plastic boxes containing liquid nitrogen from the field to the laboratory to preserve the original physical structure of the soil. In the lab, the collected fresh soil samples were divided into two subsamples. One subsample was used to analyze immediately soil microbiology and enzyme activities including the numeration of beneficial bacteria such as nitrogen-fixing bacteria, phosphorus solubilizing bacteria, potassium solubilizing bacteria, calcium solubilizing bacteria and silicate solubilizing bacteria and soil enzyme activities such as β-glucosidase, urease, phosphomonoesterase, and phytase. The other subsample was air-dried at 4 °C so that the soil moisture of all samples reached 10–12%. Then, the air-dried sample was ground through a 0.5-2 mm sieve (G.I. Sieves, Rachana, India) to homogenize the sample as well as to remove the roots and plant debris and used for analysis of soil physical and chemical properties including moisture content, soil texture, soil porosity, and bulk density. Soil chemical properties included tests for pHw, electrical conductivity (EC), soil organic matter (SOM), total nitrogen (TN), total phosphorus (TP), total potassium (TK), available nitrogen (NH₄⁺, NO₃⁻), available phosphorus (AP), exchangeable potassium (K⁺), exchangeable calcium (Ca²⁺), available silicate (SiO₂) and some trace elements such as copper (Cu), zinc (Zn), boron (B) and manganese (Mn).

Soil physical property determination

Soil bulk density was determined using undisturbed soil samples that were collection using a soil auger equipped with Royal Eijkelkamp soil sample rings with a volume of 98.125 cm³ (C53; Royal Eijkelkamp, Giesbeek, The Netherlands). Soil was dried in a 105 °C oven for 24 h and bulk density calculated as the mass of oven-dried soil divided by the total volume (Mtyobile, Muzangwa & Mnkeni, 2020; Van, Ngoc & Hung, 2022). Soil porosity was then computed by dividing the soil bulk density to soil particle density (Mtyobile, Muzangwa & Mnkeni, 2020). The soil porosity calculation formula is shown in the following equation: ${\displaystyle \text{Soil porosity}=\left(1-\left(\text{Soil bulk density}/\text{Soil particle density}\right)\right)\ast 100}$ Soil texture was determined according to the sieve (sand particles (0.02–2 mm); silt (0.002–0.02 mm) and clay (<0.002 mm) method. Robinson’s pipette method was carried out on the basis of Stoke’s law (Grossman & Reinsch, 2002).

Soil chemical property determination

Soil pHw and EC were measured in a prepared soil slurry suspension at a ratio of 1:2.5 (w/v) (soil:water) with a pH and EC meter (Thomas, Haszler & Blevins, 1996). Soil organic matter was determined by Walkley-Black method (Nelson & Sommers, 1982). Total nitrogen was determined using the Kjeldahl distillation method which converts all organic nitrogen in the sample to N-NH₄⁺ form (Horneck & Miller, 1998). Soil samples were digested with H₂SO₄ and HClO₄ to determine total phosphorus and total potassium. Then, the sample was mixed with phosphomolybdate using the reducing agent ascorbic acid and measured colorimetrically on a spectrophotometer (Shimadzu-UV–VIS UVmini-1240, Japan) at 880 nm wavelength. Total soil potassium was measured using an atomic absorption spectrometer and a flame photometer (Shimadzu AA-7000 Series Consumables, Shimadzu, Tokyo, Japan) (Jackson, 1958). Soil N-NO₃⁻ and N-NH₄⁺ were extracted by 2M KCl. The N-NH₄⁺ content is determined by showing the formation of a blue compound due to the reaction between ammonium and salicylate and hypochlorite ions in the presence of sodium nitroprusside. The N-NO₃⁻ concentration was measured by converting NO₃⁻ to NO₂⁻ in a VCl₃ solution. In an acidic environment, the NO₂⁻ solution combines with sulfanilic acid and α-naphthylamine reagents to produce a pink solution that can be detected by spectrophotometer, at 540 and 650 nm wavelengths, respectively (Shimadzu-UV–VIS UVmini-1240; Shimadzu, Tokyo, Japan) (Bremner, 1996). Soil available phosphorus was determined according to the Olsen method (Olsen & Sommers, 1982). Soil exchangeable potassium and Ca²⁺was extracted with 0.1M BaCl₂ and the elemental content was measured on an atomic absorption spectrometer (Shimadzu AA-7000 Series Consumables) (Thomas, 1982). Soil available boron was dissolved in hot distilled water to react with Azomethine H in a buffer solution to create a yellow mixture, then the sample was measured on a spectrophotometer at 420 nm wavelength (Shimadzu AA-7000 Series Consumables) (Wolf, 1971). Soil soluble silicon content was determined on the UV-Visible spectrophotometer (Shimadzu-UV–VIS UVmini-1240) by the silicon molybdenum blue reaction between a wavelength of 400 nm and 800 nm (Hallmark, Wilding & Smeck, 1982). Trace elements such as total soil Copper, Zinc, and Manganese were determined according to the Mehlich 3 method (Mehlich, 1984).

Soil biological property determination

Statistical analysis

One-way analysis of variance (ANOVA) was used in data analysis. Results are presented as means ± standard deviation (SD), and significant differences were verified by computing the Tukey’s Test at the level of 5%. Statistical tests were conducted with the Minitab (version 16.2) software. In R (R Core Team, 2023), a principal component analysis (PCA) was performed using the built-in facto extra package from the replicated data. The suitability of data for PCA was tested by using the Bartlett test after standardization. Pearson’s correlation was conducted using the ‘cor’ function and a correlation matrix was generated using the ‘corrplot’ package.

Soil physical properties of different Longan orchards

The majority of soil samples collected from various Longan orchards representing different raised bed ages exhibited soil textures in the silty clay and silty clay loam classes, as depicted in the USDA soil textural triangle (Fig. 1 and Table 1). These Longan orchard soils typically contained mean percentages of sand, silt, and clay at 5.7%, 60%, and 35%, respectively. The soil textures ranged from medium to fine-textured soils which are typical for the alluvial soils found found in the middle of the Vietnam Mekong Delta, and are well suited for fruit orchard cultivation (Le, 2003). Additionally, the particle density of these Longan orchard soils ranged from 2.43 to 2.46 g cm⁻³, within the normal range for mineral soils in the cultivated layer. The soil moisture content in the surface layer of Longan orchards was quite low, ranging from 24.7% to 29.2% as the soil was sampled at harvest time. The soil bulk density of these 20 different Longan orchard soils varied between 1.15 to 1.31 g cm⁻³, a density which does not affect root growth. The soil porosity of the soil samples was between 46.8% and 53.8%, ranking from low to moderate which is a typical porosity range for clay soil.

Significant variations were observed for soil bulk density, soil porosity, soil moisture content, sand and clay among orchards grown in raised beds of different ages (Table 1) (p < 0.05). Lower soil moisture content, soil porosity and sand content and significantly higher soil bulk density and clay content were decoupled as the Longan orchard beds increased in age. These variations suggest differing levels of soil compaction, probably the long-term impacts of farming practices such as pest and crop control, irrigation, and soil surface management as observed during soil surveys and local farmer interviews. The significant reduction of soil moisture content, porosity, and sand content with increasing soil bulk density, and clay content over increasing duration of Longan cultivation indicates increased soil degradation and soil compaction. The Longan farmers in this study have applied traditional chemicals using intensive management practices in their orchards including burning all the Longan residues leaving behind a bare soil surface. This practice as well as regular herbicide applications are increasing soil degradation, erosion and compaction (Duan et al., 2020). These conditions are reflected in the loss of soil organic matter. In this study the soil organic matter was significantly reduced with an increase in the orchard’s raised bed age. Many previous studies have found that soil porosity increases as the soil organic matter content increases (Nabayi et al., 2021; Fukumasu et al., 2022; Dang & Hung, 2023). The chemical, physical and biological processes in soil are strongly regulated by soil organic matter (Hoffland et al., 2020; Zhai et al., 2022). Improvements soil organic matter can help to increase soil porosity, water-holding capacity, and aggregate stability (Nabayi et al., 2021). A decrease of soil water content, soil porosity, sand content while increasing soil bulk density, and clay content was observed in the older raised beds and may be attributed to use of unsustainable cultivation practices and no incorporation of additional organic materials into the soil (Dang & Hung, 2023). Moreover, in this current study, we found that with an increase in ages of raised beds the clay content of soil was increased, but sand content was decreased, implying that the soil textures have changed over the cultivation period. This finding does not align with the findings of (Quang, 2013; Murano, Takata & Isoi, 2015; Dang & Hung, 2023), who demonstrated that the age of raised beds in fruit tree orchards does not affect the soil texture, and asserted that soil texture is a permanent property unaffected by management practices. The soil texture of the raised bed was identified as silty clay and silty clay loam (Table 1). These orchard soils having an origin from alluvial island deposits with silt clay are highly suitable for fruit plants (Tran et al., 2020). An exploratory data analysis using Pearson’s Correlation revealed the relationship between the ages of raised beds and physical soil properties was significant (Table S2). Notably, a strong positive correlation was found between raised bed ages and bulk density (r = 0.666**), while negative correlations were observed with soil moisture content (r = −0.570**) and soil porosity (r = −0.666**). These findings suggested that older raised beds in Longan orchards are associated with higher degrees of soil compaction, which leads to lower soil porosity and reduced soil water content storage capacity in the rhizosphere. Several previous studies have shown that soil compaction significantly impacts soil structure, soil bulk density, penetrometer resistance, soil aeration, water infiltration, and hydraulic conductivity, and subsequently crop growth and root growth are hampered (Raper & Mac Kirby, 2006; Chan et al., 2006; Radford et al., 2007; Hula, Kroulik & Kovaricek, 2009; Horn et al., 2019; Keller et al., 2019). The adverse effects of soil compaction further result in decreased plant emergence, plant establishment, and plant height (Sidhu & Duiker, 2006; Millington et al., 2016; Shaheb, 2020). In severe cases, soil compaction significantly affects crop growth, development, yield, and farm income (Botta et al., 2010; Chamen, 2011; Godwin et al., 2017; Shaheb et al., 2018; Colombi & Keller, 2019). Therefore, implementing appropriate reclamation practices are critical if the soil’s physical environment is to be improved, thereby enhancing soil health for optimal fruit growth, yield, and quality. Dung et al. (2022); Dung et al. (2023) showed that using leguminous plants as a cover crop or rice straw mulch helped to increase the soil organic matter, soil nutrients, and porosity in citrus orchard raised beds in the Mekong Delta region of Vietnam.

Soil chemical properties of different Longan orchards

The chemical properties of different Longan orchard soils among different raised bed age groups from 15 to 60-year-old are shown in Table 2. There was a tendency of decreasing soil pH by raised bed age at 0–20 cm soil depth, however, no clear significance was found among the groups of raised bed ages (p > 0.05). The soil pH value ranged from 4.86 to 5.2, classifying it as very strongly acidic, which is considered low for agricultural soils. Soil organic matter, total phosphorus, total potassium, available P, available Si and exchangeable Ca²⁺ were found to differ significantly between raised bed age groups (p < 0.05). The studied soil nutrient elements strongly decreased with the age of the raised beds, with reductions in the 46–60 year raised bed group ranging from 43.76% to 63.73%. While available N (NH₄⁺, NO₃⁻), available B, Zn and Mn were significantly higher in young raised beds (p < 0.05). pH, EC, total N, and exchangeable K were low, Cu was below detectable range in all 20 soil samples.

Longan prefers soil with a pH ranging from about 6.0 to 6.5 but is tolerant of acidic soil. A range of pH values from 4.76 to 5.20 among the investigated Longan orchards may adversely affect the growth and yield of Longan. The low pH may be attributed to various factors, including plant uptake of base cations, nutrition leaching, accumulation of hydrogen ion (H⁺), and unbalanced fertilizer supply (Quang & Guong, 2011). The previous study revealed that farmers in this area frequently overuse inorganic fertilizers (urea, KH₂PO₄, (NH₄)₂HPO₄, KCl) with large quantities of inorganic P fertilizers in their orchards to enhance fruit productivity (Le & Ngo, 2022). A low pH condition can cause a deficiency of nutrients such as calcium, magnesium, phosphorus, and molybdenum as well as disturb the mineralization and nitrification process. In the results, the older age group revealed a decrease of 50% in soil organic matter, total P, total K, available P, exchange Ca²⁺, and available Si compared to the younger age group, consequently leading to adverse effects on plant growth and an imbalance in soil nutrients under low pH conditions. The soil pH values of all collected soil samples in this current study were classified as ranging from low to very strongly acidic in the 0–20 cm soil depth, indicating a spectrum of acidification over the long-term period for Longan cultivation. This phenomenon is attributed to the prolonged overuse of N based chemical fertilizers by Longan farmers in the study site, aligning with the findings of Ge, Zhu & Jiang (2018).

Similar to findings by Wang et al. (2013) and Hu, Jin & Wu (2017), the elevated levels of available phosphorus, NH₄⁺, NO₃⁻, B, Zn and Mn in all soil samples, irrespective of the ages of raised beds, were consistent with the prolonged high application of N, P, K and other micronutrient fertilizers in the study area (Hou et al., 2021). Typically, from the household survey we can see that the rate of N chemical fertilizer applied in Longan orchards is much higher than the recommended application rate for Longan trees. This practice can lead a substantial accumulation of available nitrogen in the 0–20 cm soil depth (Wang et al., 2013). Considering the P over fertilization and its strong residual effects, the application rate of P chemical fertilizer should be reduced in most Longan orchards of the study area (Bruun, Mertz & Elberling, 2006). Local farmers have recognized that Longan trees are highly susceptible to B, Zn, and Mn deficiency based on our survey, and their increased attention to B, Zn, and Mn fertilization could explain the high B, Zn, and Mn accumulation in the topsoil of Longan orchards. Copper (Cu) is an essential element for plant growth, however, no Cu was detected in soil analyses. This suggests there may be some benefit to Longan farmers in applying Cu to their soils. The repeated high applications of urea, KH₂PO₄, (NH₄)₂HPO₄, KCl as main chemical fertilizers for Longan without Ca input have caused a significant depletion of soil Ex-Ca which aligns with the study by Li et al. (2017) on the apple orchard soils. Reducing N and K chemical fertilizer inputs while manure application may serve as an alternative approach to remediate soil Ex-Ca depletion, thereby preventing bitter pit that occurs in fruit tree orchards, as suggested by Li et al. (2017).

The findings of this research are consistent with previous studies investigating the impact of raised bed age on physicochemical properties in fruit orchard soil. Research conducted on soil morphological and physico-chemical characteristics of raised-bed soil profiles under Nam Roi pomelo tree cultivation in Hau Giang province, Vietnam similarly showed that prolonged pomelo cultivation affected these soil characteristics, resulting in restricted growth and reduced yield of pomelo. Particularly, the results illustrated that low pH values (4.0–6.0), along with decreasing organic matter and available P content observed gradually changed with increasing depths (Dung et al., 2020). According to (Dang & Hung, 2023), a young bed (10 years old) and an old bed (42 years old) showed a significant difference of available P, soil organic matter, exchangeable cations, available water-holding capacity, and soil porosity with a decrease from 15% to 20% from young beds to older beds. Similarly, research on long-term cultivation of kiwi plantations found significant decreases in soil organic carbon, total N, and calcium concentrations, cation exchange capacity (CEC), water holding capacity, pH, sand content, and N mineralization and nitrification rates (Shan et al., 2020). Our findings indicate that soil organic matter, total P, total K, available P, exchangeable Ca²⁺, and available Si were significantly reduced with increasing ages of Longan cultivation. Similarly, many other previous studies on a county-scale spatial variability of soil pH and SOC, as well as that of available N, P, and K (0–40 cm depth), have consistently reported low levels of soil organic matter and other chemical properties under long-term apple cultivation (Guo et al., 2015; Zhang et al., 2016). Moreover, the relatively low levels of soil organic matter detected in this study can be attributed to several factors, including the highly weathered nature of soils, the warm and humid climate, and the consequent accelerated breakdown of soil organic matter by soil microbes (Slade & Wells, 2022).

Comparisons between soil chemical properties and the ages of raised beds of Longan using Pearson correlation analysis showed s showed that only soil organic matter and soil total N revealed a correlation with raised bed age (Table S3). Specifically, the soil organic matter indicated a strongly negative relationship with raised bed age (r = −0.579**, p < 0.01), with the age of raised beds in Longan orchards accounting for 34% of the variation in soil organic matter measurements (R² = 0.34). Similarly, soil total N also exhibited a highly negative relationship with raised bed age (r = −0.47* (p < 0.05)). The raised bed ages of Longan orchards accounted for 22% of the variation in total N measurements (R² = 0.22). This finding indicated that the long-term use of raised beds for cultivating Longan caused a significant reduction in the content of soil organic matter and total N. Therefore, there is a need for a more sustainable management approach for aged Longan orchards in Vinh Long, VMD, to overcome the low soil fertility and quality, to secure the Longan yield and incomes of Longan farmers.

Biological parameters of different Longan orchard soils

Soil enzyme activities of different Longan orchard soils

The results presented in Table 4 show the activities of selected soil enzymes in Longan orchards of four different raised bed ages. Data reveal a significant pattern of decreased enzyme activity with increasing age of the raised beds among the 4 Longan orchard soil groups. Notably, the β-glucosidase enzyme activity decreased tenfold as the raised bed age increased from 15–20 years to 50–60 years, while the activities of three other enzymes decreased by half. Obviously, the younger the raised bed age, the higher enzyme activity in the soil. Our results showed that all four studied enzyme activities including β-glucosidase, urease, phosphomonoesterase, and phytase significantly decreased with the increasing age of the Longan orchards (Table S4). This could be due to the soil organic matter decreasing also with the age of Longan orchards (Table 2). This result aligns with the findings of Rakshit et al. (2018), which demonstrate that organic matter acts as a substrate for soil enzymes. Consequently, soil enzymes are synthesized by soil microbes and their activity is strongly regulated by the presence of organic matter, serving as a source of soil organic carbon. In a study by (Dang & Hung, 2023), it was found that soil pH, electrical conductivity, available phosphorus, total nitrogen, soil organic matter, exchangeable cations (K⁺, Na⁺,…), cation exchange capacity, bulk density, and soil porosity of the older Longan orchards significantly decreased. Additionally, perennial Longan gardens exhibited low ventilation and high compaction which also resulted in reduced enzyme activity.

Table 4:

Soil enzyme activities of different Longan orchard soils in the Mekong River Delta of Vietnam.

Raised bed age(Year)	Soil enzyme activities (µg g−1 h−1)
	B-glucosidase	Urease	Phosphomonoesterase	Phytase
15–25	33.4a	32.9a	32.9a	7.41a
26–37	13.3c	18.1bc	18.1bc	5.66b
38–45	29.4b	23.8b	23.8b	5.59b
46–60	3.31d	16.5c	16.5c	4.92b
F	*	*	*	*
CV (%)	63.5	32.6	43.4	21.3

DOI: 10.7717/peerj.18396/table-4

Notes:

* Means in the same rows with different letters are significantly (p < 0.05) different according to Tukey’s test; n = 5.

The β-glucosidase enzyme, which is widely distributed in the soil, plays a crucial role in the carbon cycle, by cleaving of cellobiose into glucose molecules. This activity is closely associated with the quantity and quality of soil organic matter (Almeida, Naves & Da Mota, 2015). Urease is the hydrolytic enzyme that degrades urea, and is widely regarded as an effective proxy for nitrogen (N) mineralization in soil (Cordero, Snell & Bardgett, 2019). Phosphomonoesterase, and phytase are specific enzymes that mineralize organic phosphorus from phosphomonoesters, and inositol phosphates, respectively (Turner, 2008). Our study indicates that the changes in the activity levels of soil enzymes- β-glucosidase, urease, and phosphomonoesterase, and phytase-followed the same trends as soil organic matter, total N, and total P, respectively. This is likely because these enzymes are readily induced by their respective substrates. These results suggest that the activities of these enzymes are highly dependent on soil organic matter, total nitrogen and total phosphorus in soil, consistent with the findings of (Turner, 2008; Nannipieri et al., 2012). The activities of these soil enzymes, along with soil organic matter, total nitrogen and total phosphorus exhibited a significant decrease with the increasing age of the Longan orchards. The highest enzyme activities were observed in the younger raised beds of Longan orchards, indicating robust soil nutrient mineralization and ample availability of soil nutrients in these younger beds. During the initial phase of raised bed cultivation, Longan trees exhibit vigorous growth. Further nutrient accumulation in the fruit demands an adequate supply of available nutrients. This phenomenon may be attributed to the trees releasing root exudates, which stimulate the soil microbial activity in the rhizospheric zone. Consequently, the trees can enhance the uptake of limiting soil nutrients that may otherwise be limited (Hamilton III & Frank, 2001). A previous study by Xue, Yao & Huang (2006) on tea orchard soil in China reported that urease enzyme activity was notably higher in younger tea orchards, whereas in the aged tea orchards (>50 years old) it was significantly lower. Further investigations by Dong et al. (2009) and Zhou, Zhang & Downing (2012) revealed that even in younger tea orchards (6 years old) urease enzyme activity strongly decreased (Dong et al., 2009; Zhou, Zhang & Downing, 2012).

In our current study, we observed a significant positive correlation between the β-glucosidase enzyme and respective soil parameters such as soil organic matter ( r = 0.92 ***, R² = 0.846), total nitrogen (r = 0.76***, R² = 0.578), exchangeable Ca²⁺ (r = 0.57 **, R² = 0.33), soil moisture content (0.74***, R2 = 0.55), soil particle density (r = −0.73***, R² = 0.53), and soil porosity (r = 0.52*, R² = 0.27) (Fig. 2). Previous research has indicated that the primary direct driver of β-glucosidase enzyme in soil is organic matter content (Stott et al., 2010; Miralles et al., 2012). Furthermore, β-glucosidase activity in the soil has been found to be sensitive to a range of soil management practices and different soil types and textures. In degraded soils with low organic matter content, there is a reduction in β-glucosidase activity, leading to a diminished availability of simple sugars for microbial populations (Stott et al., 2010). Soil moisture content also regulates the activity of β-glucosidase enzymes in soil and influences the biochemical processes of soil C transformation (Zhang et al., 2011). Mulching with organic materials, such as crop residues, on the soil surface is a proper management practice for conserving soil moisture content and temperature, and can lead to an increase in β-glucosidase enzyme activity (Sarkar, Paramanick & Goswami, 2007; Murungu et al., 2011; Zhang et al., 2011; Mukumbareza, Muchaonyerwa & Chiduza, 2015; Adetunji et al., 2020). Zhang et al. (2007) found that mulching in dryland soils enhanced the water holding capacity by up to 8%, reducing soil evaporation by around 13%. Moreover, Almeida, Naves & Da Mota (2015) showed that the activity of the β-glucosidase enzyme is not only associated with the accumulation of soil organic content but also with the quality of vegetation residues and litter. The relationship between β-glucosidase enzyme and other soil particle density and soil porosity may be indirectly linked to soil organic matter. Moreover, β-glucosidase enzyme activity tends to increase with exchangeable Ca concentration, suggesting that soils developed from limestone parent material favor β-glucosidase enzyme activity. This can be attributed to calcium’s role in enhancing soil structure through soil aggregation and stimulating microbial activity (Arenberg & Arai, 2019).

The urease enzyme in soil is closely linked to nitrogen content and mineralization, as evidenced by numerous previous studies. For instance, Solangi et al. (2024) conducted a study on agricultural soil in China, revealing significant correlations between urease activity and microbial biomass phosphorus (r = 0.743**), available phosphorus (r = 0.498*), and a negative relationship with available K (r = − 0.686**). Additionally, urease activities in soils are associated with various soil properties such as organic matter content, texture, pH, silt, clay, sand contents, total nitrogen, and cation exchange capacity (CEC) (Speir, Pansier & Cairns, 1980; Dash et al., 1981; Reynolds, Wolf & Armbruster, 1985; Singh, Kumar & Dahiya, 1991; Dharmakeerthi & Thenabadu, 1996; Özdemir, Kızılkaya & Sürücü, 2000; Zhang et al., 2014). However, in acidic tea soils, no relationship between urease activity and OC or texture was found to be dependent on polyphenol content of soil (Wickremasinghe, Sivasubramaniam & Nalliah, 1981).

Our results showed no correlations between urease and soil organic matter, nitrogen, phosphorus and other properties, but a positive correlation with exchangeable Ca (r = 0.47*, R² = 0.22) (Fig. 2) was observed. This suggests that multi-enzyme activity may be better correlated with soil fertility than a single enzyme (Dick & Tabatabai, 1992). These findings are consistent with those of Xue, Yao & Huang (2006) which showed no correlations between urease, proteinase and organic matter in tea orchards of different ages in China. It is well known that urease is primarily involved in the transformation of soil N. Our study demonstrates that long-term heavy chemical fertilizer application in Longan orchard soil ecosystems do not affect the urease enzyme activity involved in soil N cycles.

Phosphomonoesterase and phytase are two enzymes involved in the mineralization of organic phosphorus into inorganic phosphorus in soil. While there is no information regarding the variation of these enzyme activities with different orchard soil ages, previous studies have shown that phosphomonoesterase enzyme activity decreases with depth down to 1 m. This decline is associated with reduced fine-root mass density (Cabugao, 2020), a lack of substrate availability (Stone & Plante, 2014) and shifts in microbial biomass C (Hou et al., 2015). In our current study, phosphomonoesterase activity was positively correlated with soil moisture (r = 0.75***, R² = 0.56), soil porosity (r = 0.63**, R² = 0.46), soil organic matter (r = 0.85***, R² = 0.72), total nitrogen (r = 0.68***, R² = 0.46), total phosphorus (r = 0.47*, R² = 0.22), and exchangeable Ca (r = 0.45*, R² = 0.20). Conversely, it was negatively correlated with soil bulk density (r = −0.68**, R² = 0.46), and soil particle density (r = −0.72***, R² = 0.52). These physical and chemical soil properties can be used to predict phosphomonoesterase activity and aligns with Cabugao et al. (2021), findings of positive correlations between soil moisture content, total soil P, and phosphomonoesterase activity. Additionally, in our study, phytase activity was significantly correlated with total phosphorus (r = 0.70***, R² = 0.49), available phosphorus (r = 0.64**, R² = 0.41), soil moisture ( r = 0.46*, R² = 0.21), soil porosity (r = 0.53*, R² = 0.28), and negatively with soil bulk density (r = −0.53*, R² = 0.28). Previous studies have shown that phytase activity was positively correlated with soil moisture, with optimal activity observed at 100% saturation, highlighting the importance of soil moisture for phytase production and microbe survival (Criquet et al., 2004; Arenberg & Arai, 2019). Furthermore, George et al. (2007) found that soil pH and temperature are critical factors regulating phytase activity. Our findings suggest that phytase activity by soil microbes can be optimized by increasing organic carbon substrate, pH, temperature, and total nitrogen (Sadaf et al., 2022).

Collectively integrated interaction among Longan soil properties, raised bed age, Longan tree age, and Longan yield

PCA analysis results showed that most indicators, including SOM, PSB, and PoSB, are correlated with Dim 2. The antagonistic relationship between bulk density, soil moisture, and porosity in Dim 2 suggested that higher bulk density leads to more aggregate compaction, reducing soil porousness (Fig. 2). Additionally, higher NO₃⁻ content in Dim 1, which is plant-accessible, was associated with higher yields in Longan trees, consistent with Cela et al. (2013) who demonstrated that soil-available-N content could accurately predict relative corn yield under irrigated, high-yielding conditions. Therefore, these findings suggest that the recommended chemical fertilizer formula for Longan needs updated, as soil properties and soil fertility have significantly changed over the past 20 years, while farmers still adhere to the outdated recommendations. This update will assist in reducing the over-application of nitrogen fertilizers, which can adversely impact the environment and human health. Notably, higher uptake of exchangeable K⁺ has an antagonistic relationship with exchangeable Ca²⁺, contributing to Dim 1 (Fig. 2). In Dim 1, NFB, exchangeable Ca²⁺, pH_w, andavailable B are negatively correlated with yield, NO₃⁻, EC, exchangeable K⁺, and available Mn and Zn. This aligns with studies showing that available nitrogen, particularly NO₃⁻, is antagonistic with NFB, where mineral N fertilizer application suppresses N-fixing bacteria and enriches disease-related microbial functions (Lekberg et al., 2021; Zhou et al., 2021; Wassermann et al., 2023). The higher soil pH associated with exchangeable Ca²⁺ may not be ideal for Longan, which prefers slightly acidic soil. High available B content can cause phytotoxicity, despite boron (B) being an essential micronutrient (Miwa & Fujiwara, 2010; Shireen et al., 2018; Wang et al., 2024) but becomes toxic in higher quantities in soil for Longan trees. Therefore, Longan farmers should avoid continuous addition of B into their soils. Integrating soil biological parameters and enzyme activities reveals crucial insights into Longan orchard soils. Microbial counts, particularly NFB and PSB, exhibit notable correlations with SOM and phosphorus content, highlighting the role of microbial communities in nutrient cycling and availability. Enzyme activities, such as β-glucosidase, urease, phosphomonoesterase, and phytase, decrease with increasing raised bed age, indicating gradual degradation of soil organic matter and nutrient turnover due to compaction and reduced porosity. This underscores the importance of soil management practices to sustain optimal soil functioning for Longan cultivation. The PCA results illustrate the interconnectedness of biological parameters and enzyme activities with soil physicochemical properties, providing a holistic perspective on soil health and its implications for Longan orchard management.

Pearson’s correlation analysis (Fig. 3) confirms the relationships between the main soil properties indicated by the PCA. Correlation coefficients between variables and their significance were used to evaluate soil health and fertility following Longan tree cultivation and to specify their significant relationships. Notably, significant positive correlations were observed between soil moisture and total P (p < 0.01) and N (p < 0.001), available P (p < 0.05), and NH₄ (p < 0.05). These findings reflect the role of well-moisturized soils in enhancing the solubility and mobility of soil nutrients, thereby elevating nutrient absorption by plant roots (Wang et al., 2023). Additionally, SOM was significantly correlated with increases in soil nutrients, including P (p < 0.05) and total N (p < 0.001), and NH₄⁺ (p <  0.05). Other soil physical properties, such as soil porosity, pH, and EC, significantly influenced the availability of N, P, and B, as well as the exchange of Ca. In contrast, direct negative correlations were observed between the soil bulk density and soil porosity (p < 0.001), organic matter (p <  0.001), N (p < 0.001), and total P (p <  0.01). The availability of P (p < 0.05), and the abundances of soil microbes, including PSB populations (p < 0.05), were consistent with previous studies (Zhang et al., 2022; Zhou et al., 2024).

The status of organic matter in the soils due to Longan tree cultivation is a crucial factor in enhancing the abundance of key soil microorganisms, such as PSB (p < 0.05), which play vital roles in nutrient cycling. Additionally, the activity of phosphomonoesterase exhibited a strong correlation with SOM (p < 0.001), ultimately increasing soil available P. This scenario suggests that the Longan tree cultivation provides essential nutrients and energy sources, and improves soil structure and creates a favorable environment for microbial activities, including PSB (Bünemann, Prusisz & Ehlers, 2011). Increased SOM was also significantly correlated with the content of ß-glucosidase, an enzyme critical for degradation of complex carbohydrates into glucose (Yang et al., 2023). However, negative correlations were found between the age of the raised beds in logan tree cultivation and soil porosity (p < 0.01), SOM (p < 0.05), and N total (p < 0.05). This may be attributed to management practices such as regular tillage over time, which can increase soil compaction and deplete organic matter, leading to decreased soil porosity and total N (Iqbal et al., 2021; Dung et al., 2022). Despite these challenges, Longan yields significantly benefit from other soil properties, such as soil available Zn. This essential nutrient regulates growth hormones and enzymes, influencing metabolic processes including photosynthesis and protein synthesis (Broadley et al., 2007; Saleem et al., 2022).