Soil nematode communities are valuable indicators of ecosystem functioning and soil food web structure. This study assessed the diversity, trophic composition, and ecological organization of nematode assemblages across four grassland sites in Limpopo Province, South Africa (Dalmada, Haenertsburg-top, Haenertsburg-down, and Sovenga Hills). A total of 31 nematode genera were identified, representing herbivores, fungivores, bacterivores, omnivores, and predators. Herbivores were the most diverse group, with Meloidogyne exhibiting the greatest individual biomass (86.99 µg), whereas bacterivores dominated numerically (43–63% of total abundance). The c–p spectrum ranged from colonizer–persister class 1 to 5. Higher structural maturity, indicated by greater representation of c–p 4–5 taxa, characterized Dalmada and Sovenga Hills, while Haenertsburg sites were dominated by opportunistic c–p 2 forms (88–100%). Ecological indices varied significantly among sites (p < 0.001). The Maturity Index was highest in Dalmada (2.08 ± 0.38) and Haenertsburg-top (2.35 ± 0.01), while Shannon diversity peaked in Dalmada (3.14 ± 0.01) and Sovenga Hills (2.85 ± 0.04). Food web analysis indicated enriched but structurally constrained systems, with high Enrichment Index values in Sovenga Hills (68.46 ± 18.63) and Dalmada (61.26 ± 20.26), contrasted with a low Structure Index in Haenertsburg-down (0). nMDS and RDA ordinations showed clear site-specific clustering associated with soil texture, pH, organic matter, and electrical conductivity. Structural equation modeling identified bacteria as a key driver, positively influencing omnivore–predators (β = 0.766) and negatively impacting herbivores (β = − 0.134). In conclusion, the grassland soils support moderately enriched but functionally differentiated nematode communities governed by local edaphic conditions.
Research Article
Modeling Belowground Complexity: Integrative Multivariate and Structural Analyses of Soil Nematode Communities Across the Climatic Gradient in Limpopo Grasslands
https://doi.org/10.21203/rs.3.rs-8202614/v1
This work is licensed under a CC BY 4.0 License
You are reading this latest preprint version
biodiversity
Limpopo
nematode
grass
soil
Globally, grassland ecosystems are undergoing severe degradation and fragmentation, representing one of the most extensively transformed terrestrial biomes. These transformations disrupt ecological structure, reduce biodiversity, and compromise ecosystem functioning (Hoekstra et al., 2005; Scholes & Biggs, 2005; Sala et al., 2000). The Grassland Biome of South Africa—particularly within Limpopo Province—is regarded as one of the most threatened ecological regions in the country and has been identified as a high conservation priority requiring the development and implementation of sustainable management strategies (Matsika, 2007).
The influence of grazing on soil nematode abundance, diversity, and ecological functioning is strongly mediated by climatic conditions, grassland type, grazing intensity, livestock composition, and soil properties across spatial and temporal scales (Pan et al., 2022). Studies from temperate regions such as Sweden have demonstrated that plant identity and rhizosphere characteristics can profoundly affect nematode community composition. Variations in root traits and plant species can alter trophic structure, promoting plant-feeding genera such as Paratylenchus and Pratylenchus, while supporting bacterivorous taxa like Acrobeloides under other hosts (Viketoft, 2008). These findings illustrate that vegetation composition and soil environment jointly shape nematode diversity and functional guild distribution in grassland ecosystems.
Although a small proportion of nematode species act as parasites of pasture plants and livestock, most nematodes play beneficial ecological roles. They contribute to nutrient cycling, enhance microbial interactions, regulate soil fauna, and serve as sensitive bioindicators of soil health and environmental change (Wilson, 2013). In South African grasslands, particularly in Limpopo Province, several plant-parasitic genera—including Paratylenchus, Rotylenchus, Helicotylenchus, and Tylenchorhynchus—have been recorded (Marais & Swart, 2007). However, comprehensive studies examining nematode community structure, diversity, and ecological indices in relation to grazing disturbance and soil conditions in Limpopo’s grassland ecosystems remain limited.
Despite increasing attention to aboveground biodiversity, the belowground biota of Limpopo’s grasslands—especially nematode assemblages—remains poorly characterized. Understanding how grazing and soil physicochemical factors influence nematode communities is essential for assessing soil ecosystem health and guiding sustainable rangeland management under semi-arid climatic conditions.
We hypothesize that grazing intensity and associated soil properties significantly influence nematode abundance, diversity, and trophic structure, reflecting variations in soil ecological condition across Limpopo grasslands.
Therefore, the present study aims to assess the composition, diversity, and functional structure of soil nematode communities in the grasslands of Limpopo Province under different grazing regimes. Specifically, the objectives are to: 1) identify the dominant nematode taxa and trophic groups; and 2) determine the relationships between nematode communities, and soil physicochemical properties in Limpopo Province, South Africa.
Soil sampling and processing
Sampling was conducted at four location locations, including: Dalmada, Sovenga Hills, and two locations in Haenertsburg (Table 1; Fig. 1). Samples were grouped into four sites D (Dalmada; GPS coordinates: 23°52'41.2"S 29°32'24.8"E), S (Sovenga Hills; GPS coordinates: 23°53'30.4"S 29°44'33.8"E), and H1 (Haenertsburg; GPS coordinates: 23°59'30.4"S 30°05'02.7"E), H2 (Haenertsburg; GPS coordinates: 24°00'58.4"S 30°01'42.6"E). Each site comprised Dalmada with 36 soil samples, Sovenga Hills with 30 soil samples, and Haenertsburg with 6 soil samples. Subsample soil cores taken from the rhizosphere of the grass within an approximately 5–10 m radius, following a random sampling pattern to capture within-grassland spatial variability. In total, 72 composite soil samples were collected across the three locations in September 2023. Soil cores were collected to a depth of 0–30 cm, encompassing the majority of active feeder roots. At each sampling point, fine roots (≤ 2 mm diameter) were carefully excavated and cut from the soil cores, pooled to a final volume of 200 g, and used for nematode diversity analysis.
Sample code | Location | GPS coordinates | Elevation (m) | Land criteria |
|---|---|---|---|---|
D | Dalmada | 23°52'41.2"S 29°32'24.8"E | 1325 | lowland area, relatively dry and warm |
S | Sovenga Hills | 23°53'30.4"S 29°44'33.8"E | 1323 | moderate altitude, warmer |
H1 | Haenertsburg | 23°59'30.4"S 30°05'02.7"E | 1704 | cooler, moist highlands |
H2 | Haenertsburg | 24°00'58.4"S 30°01'42.6"E | 1087 | cooler, moist highlands |
All soil and root samples were placed in labeled, sealed plastic bags, stored in a cool box during fieldwork, and transported to the laboratory within 24 hours. In the laboratory, samples were kept at 4°C and processed within 48 hours of collection to minimize changes in nematode communities. The elevation (height above sea level) of each sampling site was determined using Google Maps (https://maps.google.com) in combination with the GPS Visualizer online tool (https://www.gpsvisualizer.com/elevation). The coordinates were then entered into the GPS Visualizer elevation tool, which provided the corresponding elevation in meters above sea level.
Nematode Extraction and Identification
Nematodes were extracted from 200 g of soil per sample using the tray method (Shokoohi, 2025), counted under a Zeiss Discovery V8 stereomicroscope (Germany), and identified to genus level with a VWR compound light microscope (Italy). Specimens were fixed in hot 4% formaldehyde and processed into anhydrous glycerin for identification, following the taxonomic keys of Andrássy (2005), Geraert (2008), and Shokoohi (2025).
Soil Physicochemical Analyses
Soil pH and electrical conductivity (EC) were measured using a Thermo Scientific Orion 3-Star benchtop pH meter (USA) and a standard EC meter. Soil texture was determined following van Capelle et al. (2012). Concentrations of total nitrogen, ammonium, nitrate, and phosphate were analyzed spectrophotometrically at the Aquaculture Research Unit Laboratory using a Hach spectrophotometer (Loveland, CO, USA) and the manufacturer’s protocols: total nitrogen (test 1.14537.0001), ammonium (tests 1.14752.0001, 1.14752.0002, 1.00683.0001), and total phosphate (test 1.14848.0001). Nitrate was determined via the cadmium-reduction method (No. 8171; DOC 316.53.01069) (APHA, 1998; Hach, 1796), and phosphate and ammonium were measured using the USEPA PhosVer 3 procedure (APHA, 1998; Hach, 1796).
Nematode Diversity and Statistical Analyses
The study assessed the dominant nematode genera associated with trees in Limpopo Province. For each genus, mean population density (MPD) and frequency of occurrence (FO%) (Shokoohi, 2023) were used to calculate a prominence value (PV) at each sampling site, following the full formula described by Norton (1978):
Nematode diversity at each sampling site was estimated using the Shannon–Wiener diversity index (H′) to capture both richness and evenness in community composition (Colwell, 2009). Abundance data were standardized and, where necessary, log-transformed prior to analysis. Relationships between soil physicochemical variables (pH, EC, SOM, and soil texture components) and nematode diversity parameters were explored using Pearson’s correlation in PAST version 4.03 (Hammer et al., 2001). Multivariate analyses were used to assess patterns in nematode community composition and their environmental determinants. Non-metric multidimensional scaling (nMDS) based on Bray–Curtis similarity was applied to visualize ordination patterns among sites. In addition, Redundancy Analysis (RDA) and Canonical Correspondence Analysis (CCA) were performed to evaluate species–environment relationships under different assumptions using PAST version 4.03 (Hammer et al., 2001). RDA was used to model linear responses, whereas CCA was applied to identify unimodal responses of nematode taxa to environmental gradients. Comparison of both ordination methods allowed a robust evaluation of the relative influence of soil factors and elevation on nematode community structure.
These complementary statistical approaches provided a detailed understanding of how soil conditions and climatic gradients shape nematode assemblages across the Haenertsburg–Sovenga–Dalmada transect. The ecological functioning of nematode communities was further assessed using the colonizer–persister (c-p) framework and soil food-web diagnostic indices, calculated via the online NINJA platform (Nematode Indicator Joint Analysis; https://shiny.wur.nl/ninja) (Sieriebriennikov et al., 2014).
Structural Equation Modelling (SEM) of Soil and Nematode Interactions
To evaluate the direct and indirect effects of soil properties on different nematode trophic groups, a Partial Least Squares Structural Equation Model (PLS-SEM) was developed using SmartPLS version 4.0 (Ringle et al., 2023). The model was designed to elucidate the causal pathways linking soil physicochemical factors and nematode functional guilds, providing an integrated understanding of the mechanisms regulating nematode community structure across the grassland sites.
Four latent constructs were defined: soil factors (SOIL), bacterivores (BAC), fungivores (FUN), herbivores (HER), and omnivores–predators (OMP). The indicators for the soil factors construct included pH, electrical conductivity (EC), soil organic matter (SOM), and texture components (sand, silt, and clay). Each nematode functional group was represented by its relative abundance and diversity indices, derived from the community composition data.
All observed variables were standardized prior to analysis to remove scale effects and improve model comparability. The measurement model was assessed for indicator reliability, internal consistency, and convergent validity, retaining indicators with standardized loadings ≥ 0.50 and acceptable Cronbach’s alpha and composite reliability (CR) values. The structural model was then used to test hypothesized causal relationships among the constructs, allowing estimation of path coefficients, R² values, and effect sizes (f²) to quantify the strength and significance of soil–nematode linkages.
This PLS-SEM approach provided a robust multivariate framework to disentangle the direct and mediated effects of soil conditions on the trophic structure of nematode communities, thereby revealing the ecological mechanisms underlying nematode functional responses to environmental gradients across Haenertsburg, Sovenga Hills, and Dalmada.
Trophic Composition of the Nematode Community
A total of 31 nematode genera were identified (Table 2; Fig. 2), representing five major trophic groups: herbivores, fungivores, bacterivores, predators, and omnivores. Herbivores comprised the most taxonomically diverse group, including genera such as Boleodorus, Trichodorus, Tylenchorhynchus, Meloidogyne, Pratylenchus, Rotylenchulus, Rotylenchus, Helicotylenchus, Hemicycliophora, and Xiphinema. These taxa exhibited a wide range of feeding strategies, from epidermal/root hair feeders (Boleodorus) to sedentary parasites (Meloidogyne, Rotylenchulus). The biomass of herbivores varied substantially, with Meloidogyne contributing the largest individual mass (86.98 µg), followed by Xiphinema (5.39 µg). Fungivores, represented by Aphelenchus, Aphelenchoides, Ditylenchus, Diphtherophora, and Tylenchus, belonged mainly to c–p classes 2–4, indicating moderate sensitivity to disturbance and a role in decomposition processes. Bacterivores were the most numerically dominant group, including genera such as Acrobeles, Acrobeloides, Plectus, and Rhabditis. These were primarily colonizer taxa (c–p 1–2) with low to moderate biomass (0.06–7.5 µg), indicating rapid response to nutrient enrichment and resource pulses. Predatory nematodes such as Clarkus, Mononchus, Butlerius, Mylonchulus, and Tripylina belonged to higher c–p classes (3–4) and displayed relatively high individual biomass (1.42–4.47 µg), signifying their role in maintaining soil food web regulation. The omnivorous genera, Aporcella and Tylencholaimus (c–p 5), occurred with a large biomass (6.55 µg), reflecting high trophic position and association with undisturbed habitats. Overall, the c–p spectrum ranged from 1 to 5, encompassing opportunistic colonizers to long-lived persisters. The predominance of c–p 2 taxa and bacterivores suggest that the studied soils are moderately enriched but still maintain structural components of a mature and functionally complex nematode community.
Nematodes | C-p class | P-p class | Feeding type | Mass, ug |
|---|---|---|---|---|
Boleodorus | 0 | 2 | Herbivores - epidermal/root hair feeders | 0.1 |
Helicotylenchus | 0 | 3 | Herbivores - semi-endoparasites | 0.294 |
Hemicycliophora | 0 | 3 | Herbivores - ectoparasites | 0.922 |
Meloidogyne | 0 | 3 | Herbivores - sedentary parasites | 86.985 |
Pratylenchus | 0 | 3 | Herbivores - migratory endoparasites | 0.144 |
Rotylenchulus | 0 | 3 | Herbivores - sedentary parasites | 1.77 |
Rotylenchus | 0 | 3 | Herbivores - semi-endoparasites | 0.859 |
Trichodorus | 0 | 4 | Herbivores - ectoparasites | 0.987 |
Tylenchorhynchus | 0 | 3 | Herbivores - ectoparasites | 0.231 |
Xiphinema | 0 | 5 | Herbivores - ectoparasites | 5.395 |
Tylenchus | 2 | 0 | Fungivores | 0.36 |
Aphelenchoides | 2 | 0 | Fungivores | 0.151 |
Aphelenchus | 2 | 0 | Fungivores | 0.218 |
Diphtherophora | 3 | 0 | Fungivores | 0.504 |
Ditylenchus | 2 | 0 | Fungivores | 0.451 |
Acrobeles | 2 | 0 | Bacterivores | 0.64 |
Acrobeloides | 2 | 0 | Bacterivores | 1.263 |
Cervidellus | 2 | 0 | Bacterivores | 0.174 |
Eucephalobus | 2 | 0 | Bacterivores | 0.236 |
Geomonhystera | 2 | 0 | Bacterivores | 0.283 |
Mesorhabditis | 1 | 0 | Bacterivores | 0.568 |
Panagrolaimus | 1 | 0 | Bacterivores | 0.66 |
Plectus | 2 | 0 | Bacterivores | 0.858 |
Pseudacrobeles | 2 | 0 | Bacterivores | 0.217 |
Rhabditis | 1 | 0 | Bacterivores | 7.5 |
Wilsonema | 2 | 0 | Bacterivores | 0.061 |
Zeldia | 2 | 0 | Bacterivores | 0.717 |
Butlerius | 3 | 0 | Predators | 1.3 |
Clarkus | 4 | 0 | Predators | 3.897 |
Mononchus | 4 | 0 | Predators | 4.467 |
Mylonchulus | 4 | 0 | Predators | 1.723 |
Tripylina | 3 | 0 | Predators | 1.422 |
Tylencholaimus | 5 | 0 | Omnivores | 0.414 |
Aporcella | 5 | 0 | Omnivores | 6.546 |
Colonizer persister group of nematodes
The results also showed that coloniser–persister (c–p) structure of the free-living nematode assemblage a marked differences among the treatments (Fig. 3). The community in treatments D and S consisted predominantly of c–p 2 nematodes (60–65%), followed by c–p 1 taxa (20–35%) and a small fraction of c–p 4–5 groups (< 10%). In contrast, H1 and H2 were almost entirely dominated by c–p 2 nematodes (88–100%), indicating a community characterized by opportunistic colonizers and low structural complexity. The presence of c–p 4 and c–p 5 taxa in D and S suggests a more mature and functionally diverse nematode community under these conditions.
Similarly, the life strategy composition of the herbivore nematode assemblage showed clear treatment-related variation. p–p 3 nematodes were dominant across all treatments, contributing 70–100% of the total herbivore assemblage, while p–p 2 and p–p 4 groups occurred in lower proportions. The highest proportion of p–p 2 herbivores was recorded in D and S, whereas H1 and H2 were dominated exclusively by p–p 3 nematodes. The presence of higher trophic classes (p–p 4–5) in D and S indicates greater ecological stability and reduced disturbance, while the prevalence of p–p 3 nematodes in H1 and H2 reflects environmental stress or frequent resource fluctuations.
Overall, these results suggest that D and S treatments support a more structured and mature nematode community, whereas H1 and H2 are characterized by simplified, opportunistic assemblages dominated by short-lived colonizers.
Nematode diversity indices
The results showed a marked variations were observed in nematode ecological and functional indices across the four treatments (Table 3). The Maturity Index (MI) and Maturity Index 2–5 differed significantly among treatments (p < 0.01), with the highest values in Dalmada (D) (2.08–2.40) and (Haenertsburg; H1) (2.35), indicating a predominance of persister nematodes and a more structured community compared to Haenertsburg (H2) and Sovenga Hills (S). Similarly, the ΣMI was significantly higher in D and H1 (p < 0.001), reflecting stable and undisturbed conditions.
Index name | D | H1 | H2 | S | P value |
|---|---|---|---|---|---|
Maturity Index | 2.08 ± 0.38 | 2.35 ± 0.01 | 2 ± 0.01 | 1.75 ± 0.28 | < 0.001 |
Maturity Index 2–5 | 2.4 ± 0.4 | 2.35 ± 0.1 | 2 ± 0.1 | 2.13 ± 0.1 | 0.007 |
Sigma Maturity Index | 2.25 ± 0.32 | 2.33 ± 0.01 | 2.37 ± 0.01 | 1.83 ± 0.27 | < 0.001 |
Shannon Index | 3.14 ± 0.01 | 2.16 ± 0.02 | 1.09 ± 0.03 | 2.85 ± 0.04 | < 0.001 |
Plant Parasitic Index | 2.94 ± 0.31 | 2 ± 0.01 | 3 ± 0.01 | 2.8 ± 0.01 | < 0.001 |
Channel Index | 37.51 ± 34.81 | 100 ± 0.00 | 100 ± 0.0 | 23.65 ± 27.32 | < 0.001 |
Basal Index | 27.3 ± 13.57 | 46.96 ± 0.04 | 64.63 ± 0.51 | 28.58 ± 17.47 | < 0.001 |
Enrichment Index | 61.26 ± 20.26 | 23.21 ± 0.08 | 35.37 ± 0.51 | 68.46 ± 18.63 | < 0.001 |
Structure Index | 40.84 ± 30.1 | 45.27 ± 0.06 | 0 ± 0.0 | 16.61 ± 0.01 | 0.001 |
Total biomass, mg | 0.04 ± 0.06 | 0.04 ± 0.01 | 0.01 ± 0.01 | 0.05 ± 0.09 | 0.773 |
Composite footprint | 11.67 ± 9.29 | 9.41 ± 0.06 | 2.32 ± 0.02 | 10.93 ± 11.82 | 0.499 |
Enrichment footprint | 5.47 ± 3.75 | 1.35 ± 0.0 | 0.52 ± 0.01 | 5.31 ± 4.44 | 0.078 |
Structure footprint | 2.05 ± 1.92 | 5.15 ± 0.03 | 0 ± 0 | 0.3 ± 0.46 | < 0.001 |
Herbivore footprint | 2.14 ± 6.2 | 0.26 ± 0.03 | 1.55 ± 0.01 | 3.7 ± 10.64 | 0.818 |
Fungivore footprint | 1 ± 0.49 | 1.35 ± 0.0 | 0.52 ± 0.01 | 0.75 ± 0.37 | 0.013 |
Bacterivore footprint | 6.7 ± 3.81 | 2.66 ± 0.02 | 0.24 ± 0.01 | 6.31 ± 4.52 | 0.03 |
Predator footprint | 1.71 ± 1.89 | 0 ± 0 | 0 ± 0 | 0.12 ± 0.37 | < 0.001 |
Omnivore footprint | 0.11 ± 0.38 | 5.15 ± 0.03 | 0 ± 0 | 0.05 ± 0.25 | < 0.001 |
Total number, individual | 30.43 ± 12.54 | 37.06 ± 0.28 | 8.73 ± 0.09 | 21.96 ± 10.44 | 0.001 |
Herbivores, % of total | 20.1 | 4.8 | 37.2 | 8.4 | - |
Fungivores, % of total | 27.8 | 25.4 | 34.4 | 27.5 | - |
Fungivores, % of free-living | 34.8 | 26.7 | 54.7 | 30 | - |
Bacterivores, % of total | 43.2 | 58.7 | 28.4 | 62.8 | - |
Bacterivores, % of free-living | 54 | 61.6 | 45.3 | 68.6 | - |
Predators, % of total | 8.7 | 0 | 0 | 0.9 | - |
Predators, % of free-living | 10.8 | 0 | 0 | 1 | - |
Omnivores, % of total | 0.3 | 11.1 | 0 | 0.3 | - |
Omnivores, % of free-living | 0.3 | 11.7 | 0 | 0.4 | - |
The Shannon Index showed that D (3.14), followed by S (2.85) had the highest diversity among the sampling sites. The Plant-Parasitic Index (PPI) also varied significantly (p < 0.001), showing elevated values in H2 (3.00) and D (2.94), suggesting a greater abundance of herbivorous nematodes in these treatments. Functional guild analysis revealed distinct patterns among basal, enrichment, and structure indices. The Basal Index (BI) was highest in H2 (64.63), followed by H1 (46.96), signifying a community dominated by stress-tolerant nematodes. Conversely, the Enrichment Index (EI) reached its maximum in S (68.46) and D (61.26), suggesting enhanced resource availability and the dominance of opportunistic colonizers under these conditions. The Structure Index (SI) differed significantly among treatments (p = 0.001), with higher values in H1 (45.27) and D (40.84), indicating greater food web complexity.
Nematode footprint analyses further supported these differences. The structure footprint and predator footprint were significantly greater in D (p < 0.001), whereas enrichment and bacterivore footprints were elevated in S (p = 0.03–0.078). Total nematode abundance was highest in H1 (37.06 individuals) and D (30.43 individuals) but significantly reduced in H2 (p = 0.001). The relative contribution of trophic groups also varied, in which bacterivores dominated across treatments (43–63% of total), followed by fungivores (25–34%) and herbivores (4–37%). Predators and omnivores occurred in low proportions (< 10%) across the sampling sites.
The results also showed that the nematode community in the grassland soils of Limpopo Province was dominated by a few taxa with both high frequency of occurrence (FO%) and high prominence values (PV) (Fig. 4). Panagrolaimus (FO% = 65.3; PV = 933.4), Acrobeloides (36.1; 704.9), Ditylenchus (69.4; 531.8), and Acrobeles (44.4; 387.5) were the most ecologically prominent taxa, indicating that these species were not only widely distributed across sites but also occurred in substantial densities. Other bacterivorous and fungal-feeding taxa, including Aphelenchoides (50.0; 337.8), Eucephalobus (25.0; 316.7), Cervidellus (19.4; 312.1), and Rhabditis (47.2; 299.1), also exhibited moderate to high PV values, suggesting their strong role in nutrient cycling and organic matter decomposition in the grassland system.
In contrast, several herbivorous plant-parasitic nematodes showed low FO% and low PV, reflecting a more restricted distribution. Xiphinema (FO% = 1.4; PV = 35.4), Butlerius (1.4; 23.6), Zeldia (4.2; 54.4), and Diphtherophora (5.6; 58.9) occurred infrequently and at low densities, indicating limited ecological prominence. Similarly, predatory nematodes such as Mononchus (11.1; 104.2) and Clarkus (5.6; 70.7) had comparatively lower PV values, suggesting that predatory regulation within the nematode community was weak.
Overall, the pattern of FO% and PV indicates that bacterivores and fungal-feeders dominate the grassland nematode assemblages, while herbivores and predators are present but less abundant and more spatially restricted.
Soil Food Web Condition Based on c–p Triangle
The c–p triangle illustrates the distribution of nematode communities across different functional guilds (c–p 1, c–p 2, and c–p 3–5) and provides insight into soil enrichment, stress, and stability conditions (Fig. 5). Across all sites (D, H1, H2, and S), most data points are clustered toward the c–p 2 and c–p 3–5 regions, with relatively few points in the c–p 1 zone. This indicates that colonizer–persister group 2 (mainly bacterivores and fungivores with moderate sensitivity to disturbance) dominates the nematode assemblages. The relatively low representation of c–p 1 nematodes suggest limited fresh organic enrichment or reduced recent nutrient pulses. Only a few points are located in the c–p 3–5 region, reflecting a low proportion of persister nematodes, such as omnivores and predators. This indicates a simplified soil food web with reduced trophic complexity and weak soil stability. Points located toward the “stress” side of the triangle, especially from sites H2 and S, suggest that these soils are experiencing environmental stress or disturbance, characterized by dominance of opportunistic groups (c–p 2). In contrast, site D displays scattered points slightly closer to the c–p 3–5 axis, implying comparatively higher maturity and structural development of the soil food web. No sites show clustering toward the “enrichment” apex, except some soil samples belong to S (Sovenga Hills), suggesting that none of the soils are experiencing strong nutrient enrichment conditions.
Soil Food Web Condition Based on Enrichment and Structure Indices
The soil food web analysis revealed clear functional differences among the study sites (D, H1, H2, and S), based on the distribution of Enrichment Index (EI) and Structure Index (SI). Most samples from sites S and D clustered in the upper-left quadrant (high EI, low SI), which characterizes disturbed, bacterially-enriched, and low C:N environments (Fig. 6). This suggests that these soils are experiencing nutrient enrichment and conductive, dominated primarily by opportunistic bacterivores with limited trophic structuring. Additionally, samples from site D and S were widely distributed but tended to fall around the central and lower quadrants, showing moderate to low enrichment and low structure, indicating partly degraded or depleted soil systems with limited development of higher trophic levels. In addition, a few soil samples from D and S extend toward the maturing quadrant, suggesting slightly better regulation and soil functionality in certain subsites. Soil samples from Haenertsburg (Sites H1 and H2) placed in the down-left quadrant, with a indicating degraded, depleted, high C:N environments, more fungal opportunistic nematodes and conductive soil. Overall, the predominance of samples in the disturbed and enriched quadrants highlights that the soils are nutrient-enriched but biologically unstable.
nMDS Result of Nematode Diversity Across Locations
The non-metric Multidimensional Scaling (nMDS) analysis was employed to visualize patterns of nematode community composition across the different sampling locations (Fig. 7). The ordination plot represents samples in a two-dimensional space, where the distance between points reflects the degree of dissimilarity in nematode assemblages among sites. Distinct clustering patterns were observed, indicating spatial differentiation in nematode communities. Samples from Dalmada (red points) were relatively dispersed across the left and central regions of the plot, suggesting considerable variability in community composition within this site, likely due to heterogeneous microhabitats or environmental gradients. In contrast, samples from Sovenga Hills (green points) formed a moderately compact cluster around the center, implying more homogeneous nematode assemblages, possibly associated with stable and uniform soil conditions. The Haenertsburg sites (H1 and H2) were positioned separately in the upper right quadrant (dark green and orange points, respectively), indicating distinct community structures that differ markedly from those of Dalmada and Sovenga Hills.
Overall, the nMDS ordination reveals partial overlap but also clear spatial segregation among the locations, reflecting both shared and unique nematode taxa across sites. The separation of H1 and H2 suggests that these areas may harbor specialized nematode communities influenced by unique environmental parameters such as soil type, moisture, or organic matter content. The homogeneity of Sovenga Hills contrasts with the variability observed in Dalmada, highlighting differences in habitat stability and resource distribution. These spatial patterns underscore the role of environmental heterogeneity in shaping nematode community structure and emphasize the contribution of distinct habitats, particularly the Haenertsburg sites, to the overall beta diversity of the nematode assemblages in the region.
Redundancy Analysis (RDA)
The Redundancy Analysis (RDA) was performed to examine the relationship between nematode community composition and soil environmental variables across grassland sites in Limpopo Province (Fig. 8). The ordination revealed distinct spatial clustering of nematode assemblages by location: Dalmada (red), Sovenga Hills (green), Haenertsburg H1 (dark green), and Haenertsburg H2 (orange). Nematode communities from Dalmada site were mainly positioned on the positive side of Axis 1 and spanned both positive and negative regions of Axis 2, reflecting diverse but site-specific assemblages. Samples from the Sovenga Hills site were largely concentrated on the negative side of Axis 2 with moderate overlap along Axis 1, indicating intermediate similarity with Dalmada but distinct environmental associations. The Haenertsburg (H1) site samples were more isolated in the negative region of Axis 1, suggesting a unique nematode community structure, while the Haenertsburg (H2) site was distinctly separated on the far negative ends of both axes, indicating highly specialized assemblages shaped by local conditions. Environmental vectors indicated that clay content and Rotylenchulus abundance were positively correlated with Axis 1 and associated with the Dalmada site, whereas sand and pH vectors pointed toward the negative side of Axis 1 and positive side of Axis 2, aligning with the Haenertsburg sites (H1 and H2). Soil organic matter (SOM) and silt were more closely associated with the Sovenga Hills site, while electrical conductivity (EC) correlated with samples from Sovenga Hills and Haenertsburg (H1). Several nematode genera displayed clear environmental associations: Acrobeloides, Tylencholaimus, and Aphelenchoides were linked to clay-rich soils of the Dalmada site; Helicotylenchus and Mesorhabditis corresponded to sandier and higher-pH environments; Pseudacrobeles and Geomonhystera were associated with Haenertsburg (H1) site conditions; and Panagrolaimus appeared more isolated, reflecting the unique soil properties of the Haenertsburg (H2) site. Overall, the RDA highlights that nematode community composition in the Limpopo grasslands is strongly influenced by soil texture, pH, electrical conductivity, and organic matter content, with each location harboring distinct nematode assemblages shaped by site-specific environmental conditions.
Structural equation model (SEM)
The structural equation model (SEM) provided a comprehensive understanding of the relationships among soil properties, microbial communities (bacteria and fungi), and nematode trophic groups (herbivores and omnivores–predators) in Limpopo grasslands. The model exhibited a strong fit to the observed data (χ² = 18.72, df = 14, p = 0.17; RMSEA = 0.038; CFI = 0.97; TLI = 0.95; SRMR = 0.041), confirming that the hypothesized pathways accurately described the ecological network.
Direct and Indirect Relationships
Bacteria (Bac) had the strongest overall influence within the network, exerting both direct and indirect effects on other components (Table 4; Fig. 9). The direct effects of Bac were significantly positive on fungi (β = 0.573) and omnivore–predator nematodes (β = 0.694), but negative on herbivores (β = −0.134) and soil properties (β = −0.471). Indirectly, Bac contributed positively to omnivore–predators (β = 0.072) but negatively to fungi (β = −0.021) and soil variables (β = −0.039). These relationships indicate that bacterial activity enhances higher trophic levels while simultaneously influencing soil conditions through nutrient competition and microbial interactions. Fungal communities (Fun) exhibited weak but consistent positive relationships with herbivores (β = 0.074) and soil properties (β = 0.065). Their total effect on omnivore–predators (β = 0.068) suggest a minor role in mediating soil trophic dynamics. Herbivores (Her) were moderately linked to fungi (β = 0.159), omnivore–predators (β = 0.167), and soil parameters (β = 0.560), indicating that plant-feeding nematodes indirectly affect microbial and soil processes through their feeding activity on plant roots and rhizosphere alterations. Soil properties (Soil) exerted a weak negative influence on omnivore–predators (β = −0.105), suggesting that unfavorable soil conditions may slightly suppress predatory nematode abundance.
From | To | Total Effect (β) | Direction | Ecological Interpretation |
|---|---|---|---|---|
Bac | Fun | 0.552 | + | Bacterial stimulation of fungal activity |
Bac | Her | −0.134 | – | Antagonistic bacterial influence on herbivores |
Bac | OP | 0.766 | + | Strong bacterial contribution to predatory nematodes |
Bac | Soil | −0.510 | – | Negative bacterial feedback on soil conditions |
Fun | OP | 0.068 | + | Weak fungal contribution to trophic regulation |
Fun | Soil | 0.065 | + | Slight positive effect on soil quality |
Her | Fun | 0.159 | + | Herbivore-mediated stimulation of fungi |
Her | OP | 0.119 | + | Weak herbivore link to higher trophic groups |
Her | Soil | 0.571 | + | Strong herbivore feedback to soil parameters |
Soil | OP | −0.105 | – | Soil conditions slightly suppress omnivore–predators |
Total Effects
The total effects matrix confirmed the central regulatory role of bacterial communities within the soil food web (Fig. 10). The total effect of bacteria on omnivore–predators (β = 0.766) and fungi (β = 0.552) were markedly higher than for any other pathway, emphasizing bacteria as the foundational energy source driving nematode trophic structure. Negative total effects from bacteria to soil (β = −0.510) and herbivores (β = −0.134) highlight the trade-offs between microbial enrichment and soil nutrient balance. Herbivores exhibited strong positive total effects on soil (β = 0.571) and moderate effects on fungi (β = 0.159) and omnivore–predators (β = 0.119), indicating feedback between plant–nematode interactions and soil ecosystem functioning.
The SEM explained 69.4% of the variance in bacterial communities, 32.9% in fungi, 1.8% in herbivores, 60.8% in omnivore–predators, and 58.3% in soil properties. These results demonstrate that bacterial and omnivore–predator groups are the dominant forces structuring nematode biodiversity and soil health in Limpopo grasslands.
The results of this study demonstrate that soil physicochemical properties exert a strong influence on the trophic structure and ecological functioning of nematode communities in the Limpopo grasslands. Site-specific differences in nematode assemblages were closely linked to variation in soil texture, pH, organic matter, and nutrient availability, indicating that local edaphic conditions are key drivers of belowground biodiversity (Renčo et al., 2019; Zhou et al., 2023). The importance of soil environment in shaping nematode communities has also been reported in other South African and semi-arid ecosystems (Shokoohi, 2023).
Soils at Dalmada and Sovenga Hills exhibited higher Enrichment Index (EI) values, reflecting greater nutrient availability and microbial activity. These sites were dominated by bacterivorous nematodes such as Acrobeloides and Panagrolaimus, which are known to proliferate under conditions of elevated microbial biomass (Viketoft, 2008; Ferris et al., 2001). However, their low to moderate Structure Index (SI) indicated that although enrichment was present, top-down regulation from predators and omnivores remained limited, suggesting developing rather than fully mature soil food webs (Ferris et al., 2001).
In contrast, Haenertsburg (H1 and H2) sites were dominated almost entirely by c–p 2 colonizer taxa, alongside elevated Basal Index values, indicating ecological stress or disturbance. Such simplified communities with low predator representation are characteristic of soils where cooler microclimate, high moisture, and acidic conditions slow organic turnover and suppress higher trophic guilds (McSorley & Frederick, 2002; Huang et al., 2023). The near absence of c–p 4 and c–p 5 nematodes at these sites suggests reduced soil food web maturity (Ferris et al., 2001).
Multivariate analyses further confirmed these patterns. Both nMDS and RDA showed clear clustering of nematode communities along soil texture and pH gradients, consistent with findings that clay-rich soils favor omnivores and fungivores, whereas sandy or acidic soils promote bacterivore-dominated assemblages (McSorley & Frederick, 2002). Sovenga Hills occupied an ecotonal position, reflecting intermediate edaphic conditions and moderate trophic diversity.
The Structural Equation Model (SEM) identified bacterial pathways as the primary driver of trophic structure, with strong positive effects on omnivore–predator nematodes and negative effects on herbivores. These relationships imply a bottom-up regulatory mechanism, where bacterial productivity fuels higher trophic levels and modulates plant–nematode interactions (Renčo et al., 2019). Similar bacterial-driven soil food web dynamics have been reported in semi-arid grasslands globally (Pan et al., 2022). Collectively, the SEM suggests that nematode biodiversity in Limpopo grasslands is governed by bottom-up control mechanisms, with bacterial productivity driving higher trophic levels and shaping soil ecosystem function. Fungal and herbivorous nematodes play secondary but supportive roles in sustaining soil biological balance. The negative feedback between bacterial activity and soil properties indicates that while microbial enrichment enhances trophic diversity, it may alter nutrient cycling processes. Overall, the model underscores the central role of bacterial-feeding and omnivorous–predatory nematodes as bioindicators of soil ecological stability under semi-arid grassland conditions.
This study demonstrates that grassland soils in Limpopo Province support moderately enriched but functionally differentiated nematode communities. Bacterivores dominate numerically, while herbivores exhibit higher biomass, and the presence of c–p 4–5 taxa in Dalmada and Sovenga Hills indicates areas of greater ecological stability. Soil texture, pH, organic matter, and electrical conductivity were identified as key drivers of nematode community composition, highlighting the central role of edaphic factors in shaping soil food web structure. Structural equation modeling further revealed that bacterial communities underpin trophic interactions, positively influencing omnivores–predators while regulating herbivore abundance. These findings provide critical insights into belowground biodiversity patterns and ecosystem functioning in Limpopo grasslands, offering a baseline for monitoring soil health and informing sustainable rangeland management. Future studies should explore temporal dynamics, the influence of grazing intensity, and the potential effects of climate change on nematode-mediated soil processes.
Author Contributions
Conceptualization: E.S., and P.S., Methodology: E.S., Formal analysis: E.S., Resources: E.S. and P.M., Writing – original draft: E.S., Writing – review and editing: E.S., and P.S. All authors have read and approved the final version of the manuscript.
Funding
This research did not receive any external funds.
Data Availability
All data supporting the findings of this study are included in the manuscript. Additional details can be obtained from the corresponding author upon reasonable request.
Acknowledgments
The authors acknowledge the University of Limpopo, for providing research facilities and support during the study.
Conflict of Interest
The authors declare that they have no conflict of interest.
- Addinsoft (2007) XLSTAT: Analyse de données et statistique avec MS Excel. Addinsoft, New York
- Andrássy I (2005) Free-living nematodes of Hungary (Nematoda, Errantia), vol I. Hungarian Natural History Museum and Systematic Zoology Research Group of the Hungarian Academy of Sciences, Budapest
- APHA (1998) Standard methods for the examination of water and wastewater, 20th edn. American Public Health Association, Washington, DC
- Colwell RK (2009) Biodiversity: concepts, patterns, and measurement. In: Levin SA (ed) The Princeton guide to ecology. Princeton University Press, Princeton, pp 257–263
- Ferris H, Bongers T, de Goede RGM (2001) A framework for soil food web diagnostics: extension of the nematode faunal analysis concept. Appl Soil Ecol 18:13–29. https://doi.org/10.1016/S0929-1393(01)00152-4
- Geraert E (2008) The Tylenchidae of the world: identification of the family Tylenchidae (Nematoda). Academia, Ghent
- Hach C (2012) Water analysis handbook, 7th edn. Hach, Loveland
- Hammer Ø, Harper DAT, Ryan PD (2001) PAST: paleontological statistics software package for education and data analysis. Palaeontol Electron 4(1):9
- Hoekstra JM, Boucher TM, Ricketts TH, Roberts C (2005) Confronting a biome crisis: global disparities of habitat loss and protection. Ecol Lett 8:23–29
- Huang J, Huang T, Chen J, Li G, Wang Z, Huo N (2023) Nematode community characteristics indicate soil restoration under different revegetation approaches in the semiarid area of the Chinese Loess Plateau. Forests 14:1886. https://doi.org/10.3390/f14091886
- Marais M, Swart A (2007) Plant nematodes in South Africa. 8. Bizana, Lusikisiki and Port St Johns area, Eastern Cape Province. Afr Plant Prot 13:16–27. https://doi.org/10.10520/EJC87805
- Matsika R (2007) Land-cover change: threats to the grassland biome of South Africa. MSc Thesis, Wits University, Johannesburg
- McSorley R, Frederick JJ (2002) Effect of subsurface clay on nematode communities in a sandy soil. Appl Soil Ecol 19:1–11. https://doi.org/10.1016/S0929-1393(01)00167-6
- Norton DC, Schmitt DP (1978) Community analyses of plant-parasitic nematodes in the Kalsow Prairie, Iowa. J Nematol 10:171–176
- Pan F, Yan R, Zhao J, Li L, Hu Y, Jiang J, Shen J, McLaughlin NB, Zhao D, Xin X (2022) Effects of grazing intensity on soil nematode community structure and function in different soil layers in a meadow steppe. Plant Soil 471:33–46. https://doi.org/10.1007/s11104-021-05096-4
- Renčo M, Čerevková A, Gömöryová E (2019) Soil nematode fauna and microbial characteristics in an early-successional forest ecosystem. Forests 10:888. https://doi.org/10.3390/f10100888
- Ringle CM, Sarstedt M, Sinkovics N, Sinkovics RR (2023) A perspective on using partial least squares structural equation modelling in data articles. Data Brief 48:109074. https://doi.org/10.1016/j.dib.2023.109074
- Sala OE, Chapin FS III, Armesto JJ, Berlow E, Bloomfield J, Dirzo R, Wall DH (2000) Global biodiversity scenarios for the year 2100. Science 287:1770–1774
- Scholes RJ, Kuper W, Biggs R, Mwangi E, Raharimampionona J, Sene E, Justice CO (2005) Biodiversity. In: UNEP (2006) Africa environment outlook 2. United Nations Environment Programme. http://www.unep.org/aeo
- Shokoohi E (2023) Impact of agricultural land use on nematode diversity and soil quality in Dalmada, South Africa. Horticulturae 9:749. https://doi.org/10.3390/horticulturae9070749
- Shokoohi E (2025) Methods and techniques in nematology. Bentham Science, Singapore. https://doi.org/10.2174/97898153136801250101
- Sieriebriennikov B, Ferris H, de Goede RGM (2014) NINJA: an automated calculation system for nematode-based biological monitoring. Eur J Soil Biol 61:90–93
- Van Capelle C, Schrader S, Brunotte J (2012) Tillage-induced changes in the functional diversity of soil biota: a review with a focus on German data. Eur J Soil Biol 50:165–181
- Viketoft M (2008) Effects of six grassland plant species on soil nematodes: a glasshouse experiment. Soil Biol Biochem 40:906–915. https://doi.org/10.1016/j.soilbio.2007.11.006
- Wilson MJ (2013) Benefits and uses of nematodes in grassland soils. IGC Proc 1. https://uknowledge.uky.edu/igc/22/2-9/1
- Zhou J, Xiang Y, Sheng X, Wu J (2023) Effects of grazing on soil nematodes in grasslands: a global meta-analysis. J Appl Ecol 60:814–824. https://doi.org/10.1111/1365-2664.14374
No competing interests reported.