Response of Soil Nematode Communities and Trophic Structure to Trichoderma atroviride P. Karst., in Olive Groves of Mediterranean Croatia

Abstract

Regenerative agriculture is oriented around restoring soil health through natural processes. In this context, soil biota plays a central role, and bioinoculation represents a potentially effective approach for targeted modification of microbial communities. Among beneficial microorganisms, Trichoderma atroviride is prominent for its biocontrol agent (BCA) activity against plant-parasitic nematodes (PPNs), whereas its effects on free-living nematodes (FLNs) under in vivo conditions remain insufficiently explored. The aim of this study was to assess the response of nematode communities to bioinoculation with T. atroviride as an indicator of soil functional status. A three-year field study was conducted in organic olive orchards at Vodnjan and Nadin on four autochthonous olive cultivars, applying two inoculum doses of T. atroviride: 1 × 10⁶ spores mL⁻¹ (LD) and 1 × 10⁸ spores mL⁻¹ (HD). Bioinoculation increased the diversity of the soil nematode communities at both locations. However, the responses differed between the two inoculum doses. Both doses were associated with an increased abundance of FLNs and a reduced abundance of herbivorous nematodes relative to the control, with LD showing a more consistent and ecologically favourable effect. In combination with biotic and abiotic factors, the LD dose was associated with greater trophic diversity and a more structured soil food web, whereas increasing the inoculum concentration (HD) did not result in additional functional improvement.

1. Introduction

Healthy soils are a fundamental prerequisite for stable and resilient agroecosystems [1]. However, the capacity of soils to maintain stability and resilience is declining rapidly, with more than 60% of European soils exhibiting signs of degradation driven by intensive agricultural practices, including erosion, compaction, and contamination. Soil degradation jeopardises biodiversity, ecosystem resilience, and ultimately food security, while underscoring the urgent need to establish effective systems for soil monitoring and restoration [2]. In addition, Ferreira et al. [3] highlight that the Mediterranean region is particularly vulnerable due to the convergence of seasonal droughts, erosion processes, and shallow soil profiles, which collectively exacerbate soil degradation. Within this context, agroecological and organic approaches represent a sustainable alternative, relying on biologically mediated processes that restore soil fertility. These approaches stimulate soil microbial activity and support functional biodiversity [4,5]. Their further development and wider implementation are therefore essential for advancing regenerative agriculture in alignment with the objectives of the European Green Transition by 2030 [6].

Recent research increasingly contributes to the development of sustainable solutions by placing beneficial microorganisms at the centre of approaches aimed at preserving soil fertility and soil health [7,8]. These organisms are no longer regarded merely as inherent components of soil, but as drivers of functional change, and are therefore increasingly employed in the assessment of soil status and the effectiveness of soil restoration measures [7]. In this context, the conservation and restoration of microbiological balance have emerged as important indicators of sustainable soil management. Accordingly, bioinoculants are defined as deliberately applied beneficial microorganisms that are environmentally friendly and contribute to the enhancement of soil functionality, particularly under adverse environmental conditions [9]. Furthermore, bioinoculants are gaining prominence as an effective solution, as the application of beneficial microorganisms stimulates microbial activity, increases nutrient availability, and enhances plant tolerance to stress, thereby reducing the reliance of agricultural production on chemical inputs [10,11]. The close interconnection between microorganisms, soil, and plants places the rhizosphere at the core of their mutual interactions [10,12], where microorganisms act as mediators in the transfer of nutrients and signalling molecules between soil and plant roots [13,14]. The success of bioinoculation is therefore strongly dependent on microbial diversity and the ecological compatibility of inoculants with prevailing environmental conditions [8].

Beneficial fungus of the genus Trichoderma Pers. is recognised as a versatile, avirulent, and endophytic symbiont with a pronounced capacity for the rapid colonisation of rhizosphere and plant roots, where it establishes beneficial interactions with the host plant [15]. It readily adapts to a wide range of ecological conditions and life strategies [16]. Species of the genus Trichoderma possess multiple biocontrol mechanisms, which can be either direct (mycoparasitism or antibiosis) or indirect (competition for nutrients and space). Both modes of action promote plant growth and development and induce plant defence systems against diseases and pests [16,17,18]. Through these mechanisms, Trichoderma simultaneously suppresses phytopathogenic fungi and PPN. In addition, it enhances plant growth and stress tolerance, thereby contributing to functional soil health and the stability of agroecosystems [13,19,20,21].

Trichoderma atroviride P. Karst. is a filamentous fungus that can be isolated from soils of temperate regions and exhibits optimal growth at 25 °C under laboratory conditions [22]. It is capable of synthesising enzymes such as chitinases and glucanases, as well as secondary metabolites including 6-pentyl-α-pyrone (6PP) and harzianic acid (HA). These compounds exert antimicrobial and signalling functions, activate plant defence mechanisms, and modulate microbial communities in soil [23,24]. Collectively, these processes promote plant growth and resilience and improve the physico-chemical and biological properties of soil, including aggregate stability and the activity of beneficial microbiota [13,21].

The efficacy of Trichoderma spp. is determined by the genetic and ecological traits of individual strains and their adaptation to local abiotic conditions [21]. Consequently, the importance of autochthonous isolates, such as T. atroviride [25], adapted to specific microbiological and climatic characteristics of soils is increasingly emphasised. Such isolates may contribute over the long term to the restoration of functional soil health and the resilience of agricultural systems [12,16]. Efficient production and commercialisation of these preparations are crucial for reducing the use of chemical inputs in agriculture [26,27]. However, the biological effects of Trichoderma application in perennial cropping systems remain insufficiently explored. It is still unclear how Trichoderma inoculation influences soil biological properties, especially FLNs, which, due to their sensitivity and rapid response to environmental change, represent reliable bioindicators of soil condition [28,29].

Nematodes, including FLNs and PPNs, represent the most abundant and taxonomically diverse group of soil organisms, encompassing all trophic levels from bacterivores (Ba) and fungivores (Fu) to herbivores (He), omnivores (Om), and predators (Pr) [30]. Their widespread occurrence, ecological sensitivity, and rapid response to environmental change make them valuable indicators of soil functional status [31,32,33,34]. In addition, their presence and abundance can be readily assessed using relatively simple extraction methods [35]. Nematode Indicator Joint Analysis (NINJA) software (https://nemaplex.ucdavis.edu/, accessed on 29 March 2025) [36] enables standardised analyses and calculation of nematological and ecological indices [37].

Nematode indices represent a valuable tool for the biological assessment of soil health, as they integrate information on nematode community structure and trophic interactions into indicators of soil resilience, stability, and functional balance [29,38]. Likewise, trophic indices serve as diagnostic tools that facilitate the interpretation of changes within the soil food web under the influence of environmental factors and anthropogenic activities [33].

The effects of Trichoderma application in perennial cropping systems have most often been evaluated in comparison to nematicides, primarily in terms of efficacy against PPNs. In this context, the influence of Trichoderma application on the abundance and diversity of FLNs has generally been overlooked. Available studies conducted under in vitro or greenhouse conditions have reported significant reductions in PPN populations in soil and on plant roots [20,39,40,41]. For example, in citrus orchards, the effects of Trichoderma album Preuss, Trichoderma harzianum Rifai, and T. viride Pers., on the PPN Tylenchulus semipenetrans Cobb, have been investigated [41,42]. Against root-knot nematodes of the genus Meloidogyne Goeldi, the effects of T. harzianum and T. viride have been reported in peach orchards [20], while the effect of T. asperellum Samuels, Lieckfeldt & Nirenberg has been reported in pineapple plantations [41]. However, controlled experimental conditions do not encompass the complexity of field agroecosystems. More recent in vivo studies demonstrate that species of genus Trichoderma, applied alone or in combination with bacteria and yeasts, not only effectively reduce PPN populations but also exert positive effects on the abundance and diversity of FLNs [42,43,44]. Bridging this in vivo research gap is essential for understanding the effects of Trichoderma species not only on PPNs but also on FLNs, which reflects the functional status of soil and provides insight into interactions among microorganisms, soil, and plants. Interactions in perennial cropping systems, such as olive orchards, are highly dynamic and complex, and such studies can substantially contribute to a better understanding of the effects of bioinoculation in agroecosystems exposed to pronounced environmental variability.

Olive tree (Olea europaea L.) is a typical evergreen species with a long tradition of cultivation in the Mediterranean region [45], adapted to cool, wet winters and hot, dry summers [46]. In Croatia, olive cultivation extends along approximately 1000 km of the Adriatic coast, encompassing about 5.5 million trees distributed across six subregions [47]. Traditional, small-scale olive orchards represent an important component of sustainable Mediterranean landscapes, as they contribute to the conservation of biodiversity, particularly varietal diversity [48]. Preservation of autochthonous cultivars, such as Oblica, Lastovka, Istarska bjelica, and Buža, is crucial for the resilience and sustainability of Mediterranean olive growing [49,50]. Their genetic variability underlines the need for integrated pest management strategies [51], a conclusion supported by similar findings for the Portuguese cultivar Galega vulgar [52]. Climate change is driving increased thermal and water stress, yield reductions [53], and shifts in phenology and in the occurrence of key pests, such as the olive fruit fly and olive moth [54]. Drought conditions also alter the composition of root-associated and rhizospheres microbiomes, in which beneficial microorganisms activate plant resistance mechanisms [55]. Consequently, the conservation of local cultivars and the promotion of natural microbial interactions constitute a cornerstone of olive sector adaptation to climate change.

The aim of this study is to contribute to a deeper understanding of soil nematode biodiversity and its interactions with T. atroviride. Expansion of our knowledge of the role of bioinoculation in improving and maintaining soil health is addressed through testing the following hypotheses: (i) the presence of T. atroviride enhances soil quality by promoting increases in overall nematode biodiversity, as reflected by abundance-based indicators; (ii) inoculum application reduces the proportion of PPN while increasing the proportion of beneficial FLN trophic groups; (iii) different concentrations of T. atroviride exert distinct effects on nematode genus-level biodiversity, soil disturbance indices, and diagnostic food web indices.

2. Materials and Methods

2.1. Site Description: Soil and Climate

A three-year study (2022–2024) was conducted at two sites in coastal Croatia (Figure 1). The research included organically managed olive orchards, with a focus on autochthonous cultivars characteristic of the investigated area.

The first study site was located in Vodnjan (44°57′34.6″ N, 13°50′14.3″ E), Istria County, and included organically managed olive orchards with the autochthonous cultivars Buža (0.21 ha) and Istarska bjelica (0.21 ha). The second study site was located in Nadin, Biljane Donje (44°05′30.3″ N, 15°28′50.7″ E), Zadar County, and included the autochthonous olive cultivars Oblica (0.17 ha) and Lastovka (0.17 ha). The experimental design comprised three treatments, with 20 trees per treatment. The Trichoderma atroviride fungal isolate was isolated from soil collected in the Skradin area (43°49′29.4″ N, 15°55′16.1″ E), Šibenik–Knin County (Figure 1). Inoculum was propagated at the Department of Plant Pathology, Faculty of Agrobiotechnical Sciences Osijek.

The soil at first study site was classified according to the IUSS Working Group WRB [56] as a Calcic Cambisol (Calcocambisol), developed on limestone parent material [57,58]. This soil type is characterised by a slightly alkaline reaction (pHH₂O ≈ 8.2), a moderate calcium carbonate content (CaCO₃ ≈ 4.8%), and medium-to-elevated levels of organic matter (humus ≈ 3.5%), as confirmed by the results of long-term agrochemical analyses conducted within the framework of this study.

The climatic conditions are presented in Figure 2. The mean annual air temperature in 2022 was 16.2 °C, remained the same in 2023 (16.2 °C), and increased to 16.5 °C in 2024. The total annual precipitation amounted to 596.9 mm in 2022, increased to 823.3 mm in 2023, and reached the highest value in 2024, with a total of 913.5 mm. The mean annual soil temperature at a depth of 10 cm averaged 17.5 °C in 2022, decreased slightly to 17.2 °C in 2023, and returned to 17.5 °C in 2024 [59].

The soil at the second study site was classified according to the IUSS Working Group WRB [56] as Rendzina (Rendzic Leptosol) and/or brown soil on limestone (Eutric Cambisol), developed on limestone and dolomite parent material [57,58,60]. Such soils are characterised by a pronounced skeletal content (up to 50%) and a slightly alkaline reaction (pHH₂O ≈ 8.1), variable calcium carbonate content (CaCO₃ 3.7–7.5%), and a medium-to-high humus content (≈2.8–3.0%), as confirmed by the results of long-term agrochemical analyses conducted within the framework of this study.

The climatic conditions are presented in Figure 3. The mean annual air temperature was lowest in 2022 (15.5 °C), increased slightly in 2023 (15.7 °C), and reached the highest value in 2024 (16.3 °C). The total annual precipitation was lowest in 2022 (830.6 mm), highest in 2023 (1171.5 mm), and decreased in 2024 to 989.4 mm. The mean annual soil temperature at a depth of 10 cm averaged 18.9 °C in 2022, decreased slightly to 18.6 °C in 2023, and reached the highest value during the study period in 2024 (19.8 °C) [59].

According to Bašić [58], the investigated sites belong to two different climatological subregions of coastal (Adriatic) Croatia. The Vodnjan site is located within the region J-1, which is characterised by a typical Mediterranean climate with warm, dry summers and mild, wet winters. In contrast, the Nadin (Biljane Donje) site belongs to the region J-2, which exhibits a slightly modified Mediterranean climate, with lower mean temperatures, more pronounced seasonal oscillations, and a higher frequency of frost events. Both locations are suitable for olive cultivation.

2.2. Experimental Design

The study was conducted using a split-plot design with three treatments and four replicates: control treatment (without application), treatment with two doses of T. atroviride spores at a concentration of 1 × 10⁶ spores mL⁻¹ (LD), and treatment with a higher dose at a concentration of 1 × 10⁸ spores mL⁻¹ (HD). Trichoderma atroviride (isolate TAv1, collection of the Chair for Phytopathology, Faculty of Agrobiotechnical Sciences, University of Osijek) suspension was applied a total of five times during the 2022–2024 period, twice per year (in spring and autumn), at a dose of 2 L per olive tree, applied within a 1 m radius of the tree canopy. The control group received water only, also applied twice per year. The suspension was applied manually using a watering can with a sprinkler attachment. In total, 288 soil samples were collected along the outer edge of canopy according to the following scheme: 3 treatments × 2 sites × 2 sampling events per year × 2 olive cultivars × 3 years × 4 replicates.

2.3. Soil Sampling for Nematological Analysis

Soil sampling for nematological analyses was conducted twice per year, in spring and autumn, during the 2022–2024 period, always prior to the application of T. atroviride. Soil samples for nematological analyses were collected using a soil probe at a depth of 0–30 cm. After sampling, soil samples were transported to the laboratory and stored at 4 °C until nematode extraction, which was carried out shortly after sampling.

Nematodes were extracted from 100 g of fresh soil, corresponding to approximately 100 cm³ of soil volume. Extraction was performed at the Nematology Laboratory, Faculty of Agrobiotechnical Sciences in Osijek using the Baermann funnel method, targeting live, actively moving nematodes [61,62]. Nematodes were counted using a stereo zoom microscope (Olympus BX16, Olympus Corporation, Tokoy, Japan) and prepared for identification by producing semi-permanent and permanent slides using the rapid glycerin method [62]. Subsequently, nematodes were identified to the genus level using a compound microscope (Olympus BX50, Olympus Corporation, Tokoy, Japan).

In each sample, at least 100 nematodes were identified to genus level, or all nematodes were identified in samples containing fewer than 100 individuals. Measurements were performed using Promicra—QuickPHOTO MICRO 3.1 software (Promicra, Prague, Czech Republic). Identification was carried out using standard taxonomic keys [63,64,65,66,67,68,69,70], supplemented with additional species descriptions published in the relevant scientific literature.

Classification of soil nematodes according to life strategies and trophic affiliation enables the calculation of nematode-based indices that quantify community abundance and diversity. The life strategies of FLNs and PPNs were classified along a coloniser–persister (c–p) scale ranging from cp-1 to cp-5 [71], where opportunistic species with short life cycles are assigned to cp-1, while long-lived and more sensitive species are classified within the cp-5 group. PPNs constitute a distinct “pp” group [72].

Nematode trophic groups are determined based on the morphology of the feeding apparatus and include Ba, Fu, He, Pr, and Om [30]. Based on life-history and feeding strategies, the main biological indices used for the assessment of soil health were calculated. Ecosystem stability is assessed using a maturity index (MI) [31,38]. Its modification, MI2-5, is calculated by excluding the cp-1 group, as this group is highly sensitive to stress caused by soil disturbance or increased microbial activity [73]. In addition, the plant-parasitic nematode index (PPI) is calculated based on He, considering their different response to disturbance compared to FLNs [74].

Furthermore, diagnostic soil food web indices according to Ferris et al. [33] were applied: basal index (BI), enrichment index (EI), structure index (SI), and channel index (CI). These indices reflect the soil food web structure and complexity, food availability, and predominance of bacterial vs. fungal decomposition pathways. Calculation of all indices was performed using NINJA software (https://nemaplex.ucdavis.edu/, accessed on 29 March 2025) [36] which enables standardised and reliable analysis of nematode communities [37].

2.4. Statistical Analysis

2.4.1. Univariate Analysis

The collected data for the analysed feeding and nematode indices, trophic groups, and functional groups did not follow a normal distribution, as determined by preliminary statistical analysis (Shapiro–Wilk test and graphical inspection). To properly analyse the effect of T. atroviride application, a generalised linear mixed model (GLMM) with a Gamma distribution and a logarithmic link function (log-link) was applied. This model was selected because it allows the analysis of data that do not follow a normal distribution while simultaneously accounting for the complex structure of the applied experimental design.

Primary inferential statistical analysis, that is, the testing of effects (individual variables and interactions), was performed using the Wald chi-square test, while post hoc comparisons of estimated marginal means were conducted using Tukey’s test. The estimated mean values (marginal means) and 95% confidence intervals obtained on the transformed logarithmic scale were subsequently back-transformed to the original scale using an antilogarithmic transformation to ensure that the presented results reflected actual values in original units of measurement and were as intuitive as possible for interpretation.

2.4.2. Multivariate Analyses

Although univariate analyses are essential for assessing treatment effects on individual properties, they cannot capture complex, interrelated relationships among variables. Therefore, as a complementary approach, a series of permutational multivariate analyses of variance (PERMANOVAs) were conducted for feeding and nematode indices, trophic groups, and functional groups to evaluate the cumulative effect of T. atroviride. Since PERMANOVA does not allow the direct inclusion of baseline values as covariates, changes in variables were analysed as differences between the final and initial states. The statistical significance of fixed effects was determined using a permutation test (999 permutations), which does not assume normality of distribution and is based on empirical reshuffling of samples among groups, thereby ensuring robustness and suitability for complex ecological data.

Visualisation of cumulative effects was performed using non-metric multidimensional scaling (NMDS), which enables representation of multivariate data in multidimensional space while preserving relative distances among samples. Prior to ordination, a distance matrix describing differences between pairs of samples across all analysed properties was constructed. The NMDS algorithm projected these distances into a three-dimensional space (NMDS1–3) with minimal loss of information. Each point on the plot represents a sample, and their spatial proximity reflects similarity to multivariate profiles. In this way, the NMDS plot visually confirms results obtained by PERMANOVA, providing a clear insight into sample grouping according to model effects. To ensure robustness, double stratification (location × cultivar) was applied in analyses, thereby restricting comparisons within related groups and preventing the mixing of unrelated sources of variability, which allows for a more reliable interpretation of treatment effects within the experimental design.

2.4.3. Software Solutions

Statistical analyses were conducted in the R programming environment, version 4.3.0 (R Core Team, 2023). Data import from and export to Excel files were performed using the readxl [75] and writexl [76] packages, while data preparation was carried out using the dplyr [77] and stringr [78] packages.

Univariate analyses (repeated-measures ANOVA) using GLMMs were performed with the lme4 package [79]. Model assumptions were evaluated using the DHARMa package [80], primary inferential statistical analyses were conducted with the lmerTest package [81], and post hoc statistical analyses were performed using the multcomp [82], multcompView [83], and emmeans [84] packages. Permutational analysis of variance (PERMANOVA) was carried out using the vegan package [85]. Data visualisation was performed using the ggplot2 [86] and plot3D [87] packages.

3. Results

For clarity, the main text presents the results of the univariate model GLMM (Type II Wald χ² tests) for significant fixed effects (p < 0.05) and trends (p < 0.1), whereas the results of the multivariate model (PERMANOVA) are shown in their entirety.

3.1. Trophic Groups and Genus Diversity of Genera

Univariate analysis (GLMM) confirmed, in comparison to the control, an effect of treatment with T. atroviride at both investigated locations, Vodnjan and Nadin, χ² = 55.80 (p < 0.001) (

Table 1. Results of generalised linear mixed model (GLMM) testing effects of treatment with Trichoderma atroviride, year, season, location, and olive variety on genus diversity (Type II Wald χ² tests).

Genus Diversity	Fixed Effects	Df	χ2 Value	p
Genus diversity per 100 cm3 of soil	Treatment	2	55.80	<0.001 ***
	Location	1	0.24	0.6
	Season	1	1.71	0.2
	Year	2	0.02	1.0

Only significant effects and trends (p < 0.1) are shown. *** p < 0.001. Baseline reference level: Nadin_Oblica_control_autumn_2022. Baseline covariates were included in the models but are not presented here unless significant.

), on genus diversity per 100 cm³ of soil. The LD dose increased the genus diversity estimated, 5.45 ± 1.59 (p = 0.001), whereas the HD dose showed a weaker but still significant increase compared to the control (Figure 4).

The effect of treatment with T. atroviride on the diversity of nematode trophic groups was manifested as an increase in their abundance compared to the control (Figure 5). An increase in Ba was recorded at both doses (estm. LD: 0.61 ± 0.20; p = 0.002, HD: 0.59 ± 0.20; p = 0.002), as was an increase in the abundance of Fu at both doses (estm. LD: 0.40 ± 0.20; p = 0.05, HD: 0.51 ± 0.20; p = 0.01) and Pr_Om, χ² = 8.43 (p = 0.01). In contrast, relative to the control, treatment with T. atroviride reduced the abundance of He at the LD dose, estm. −15.88 ± 7.65 (p = 0.04), while at the HD dose a trend of reduced abundance was observed, estm. −14.21 ± 7.65 (p = 0.07) (

Table 2. Results of GLMM testing effects of treatment with Trichoderma atroviride, year, season, location, olive variety and interactions on trophic groups (Ba = bacterivores; Fu = fungivores; He = herbivores; Pr = predators/Om = omnivores).

Trophic Groups	Fixed Effects	Df	χ2 Value	p
Ba	Treatment	2	12.88	0.002 **
	Location	11	8.00	0.005 **
	Season	1	24.99	<0.001 ***
	Location × Olive variety	2	8.87	0.01 *
Fu	Treatment	2	7.02	0.03 *
	Location	1	9.93	0.002 **
	Year	2	5.10	0.08
He	Treatment	2	12.46	0.002 **
	Location	1	16.40	<0.001 ***
	Season	1	13.34	0.003 **
Pr/Om	Treatment	2	8.43	0.01 *
	Year	2	6.62	0.04 *
	Location × Olive variety	2	4.95	0.08

Only significant effects and trends (p < 0.1) are shown. *** p < 0.001, ** p < 0.01, * p < 0.05, p < 0.1. Reference level for all effects is Nadin_Oblica_control_autumn_2022. Baseline covariates were included in the models but are not presented here unless significant.

).

In addition to the treatment effect, a seasonal effect (spring vs. autumn) was recorded for the Ba and He trophic groups reflecting opposite seasonal trends in their abundances. In spring, an increase in Ba abundance was observed, estm. 0.45 ± 0.09 (p < 0.001), and a reduced abundance of He, estm. −12.89 ± 3.53 (p < 0.001).

At the Vodnjan location, a lower initial abundance of He was observed, estm. −20.70 ± 8.07 (p = 0.01), and a higher abundance of Ba, estm. 0.55 ± 0.20 (p = 0.005), compared to the ref. combination (Nadin_Oblica_control_autumn_2022).

A two-way interaction (location and cultivar) showed an effect on the composition of the trophic groups, χ² = 8.87 (p = 0.01). In Nadin, the cultivar Lastovka had a higher abundance of Ba, estm. 0.57 ± 0.20 (p < 0.001), and a trend towards a lower abundance of Pr_Om, estm. −4.90 ± 2.67 (p = 0.07), compared to the cultivar Oblica.

The effect of year (2024 vs. 2022) on Pr_Om showed an increase in abundance in 2024 compared to 2022, estm. 3.94 ± 1.59 (p = 0.02).

PERMANOVA analysis (

Table 3. Results of PERMANOVA testing differences in nematode trophic groups across treatments with Trichoderma atroviride, years, seasons and interaction.

Source of Variation	Df	Sum of Squares	R2	F	p-Value
Treatment	2	10.169	0.060	8.54	0.001 ***
Year	2	7.369	0.044	6.19	0.002 **
Season	1	12.625	0.075	21.21	0.001 ***
Treatment × Year	4	2.343	0.014	0.98	0.366
Residual	230	136.882	0.808	—	—
Total	239	169.388	1.000	—	—

Levels of significance: *** p < 0.001; ** p < 0.01.

) showed an effect of treatment with Trichoderma atroviride, F = 8.54 (p = 0.001), year, F = 6.19 (p = 0.002), and season, F = 21.21 (p = 0.001), on the overall diversity of nematode trophic groups (Figure 6). Season explained the largest proportion of total variance (R² = 0.075), 7.5%, followed by treatment (R² = 0.060), 6.0%, and year (R² = 0.044), 4.4%. The interaction between treatment and year showed no effect (p = 0.366).

3.2. Nematode Functional Guilds

The results indicate a differentiated effect of treatment with T. atroviride on PPNs of two functional groups. Relative to the control, an effect of treatment with T. atroviride on the abundance of the PPN pp-2 group was recorded, χ² = 10.82 (p = 0.004), as well as an effect on the abundance of the pp-3 group, χ² = 6.00 (p = 0.05) (

Table 4. Results of GLMM testing the effects of treatment with Trichoderma atroviride, year, season, location, olive variety and interactions on nematode functional guilds according to the plant-parasitic class (pp) [72].

(pp) Guild	Fixed Effects	Df	χ2 Value	p
pp-2	Treatment	2	10.82	0.004 **
	Location	1	20.77	<0.001 ***
	Season	1	14.30	<0.001 ***
pp-3	Treatment	2	6.00	0.05 *
	Location	1	21.38	<0.001 ***
	Season	1	12.32	<0.001 ***

Only significant effects and trends (p < 0.1) are shown. *** p < 0.001, ** p < 0.01, * p < 0.05. Reference level for all effects is Nadin_Oblica_control_autumn_2022. Baseline covariates were included in the models but are not presented here unless significant.

).

Furthermore, an effect of both doses was recorded (estm. LD: 0.91 ± 0.35; p = 0.009, HD: 0.95 ± 0.35; p = 0.007) on the increase in abundance of the pp-2 group (Figure 7).

An effect of season (spring vs. autumn) on PPN abundance was recorded in spring, χ² = 12.32; 14.30 (p < 0.001). For the PPN pp-2 group, an increase in abundance was recorded, estm. 0.66 ± 0.18 (p < 0.001), in spring, whereas for the pp-3 group a decrease in abundance was recorded, estm. −11.56 ± 3.30 (p < 0.001).

At the Vodnjan location, a higher initial abundance of PPN pp-2 was observed, estm. 1.60 ± 0.42 (p < 0.001), and a lower initial abundance of pp-3, estm. −17.00 ± 7.67 (p = 0.03), compared to the ref. combination.

Furthermore, the results obtained (

Table 5. Results of GLMM testing the effects of treatment with Trichoderma atroviride, year, season, location, olive variety and interactions on nematode functional guilds according to the coloniser–persister (c–p) guild.

(cp) Guild	Fixed Effects	Df	χ2 Value	p
cp-1	Treatment	2	5.38	0.07
	Location	1	8.80	0.003 **
	Season	1	16.42	<0.001 ***
cp-2	Location	1	2.81	0.09
	Year	2	10.17	0.006 **
	Season	1	3.12	0.08
	Location × Olive variety	2	9.72	0.008 **
	Location × Olive variety × Treatment	4	12.30	0.02 *
cp-3	Year	2	15.71	<0.001 ***
	Season	1	17.77	<0.001 ***
	Location × Olive variety	2	6.02	0.05 *
	Location × Olive variety × Treatment	4	9.85	0.04 *
cp-4_5	Location × Olive variety	2	9.29	0.01 **
	Location × Olive variety × Treatment	4	8.91	0.06

Only significant effects and trends (p < 0.1) are shown. *** p < 0.001, ** p < 0.01, * p < 0.05, p < 0.1. Baseline reference level for all effects is Nadin_Oblica_control_autumn_2022. Baseline covariates were included in the models but are not presented here unless significant.

) indicate a differentiated effect of treatment with T. atroviride on individual functional groups. Relative to the control, treatment with T. atroviride reduced the abundance of the cp-1 group in HD treatment, estm. −5.46 ± 2.76 (p = 0.05), while in LD treatment a trend of reduced abundance was observed, estm. −5.15 ± 2.76 (p = 0.07).

In comparison to the ref. combination, at the Vodnjan location a lower initial abundance of the cp-1 group was recorded, estm. −5.47 ± 2.79 (p = 0.05), and a higher initial abundance of the cp-2 group, estm. 22.68 ± 7.14 (p = 0.002).

Two-way interactions (location and treatment) and (location and cultivar) showed that at the Vodnjan location, relative to the ref. combination, the application of T. atroviride reduced the abundance of the cp-2 group at both doses (estm. LD: −23.56 ± 9.66; p = 0.02, HD: −23.53 ± 9.66; p = 0.02), whereas in Nadin the cultivar Lastovka, relative to the cultivar Oblica, had a higher abundance of the cp-2 group, estm. 26.58 ± 6.90 (p < 0.001), and a lower abundance of the cp-4_5 group, estm. −18.99 ± 4.56 (p < 0.001).

A three-way interaction (location, cultivar, and treatment) at the Nadin location, for cultivar Lastovka, relative to the ref. combination, showed a lower abundance of the cp-2 group at both doses: LD estm. −26.97 ± 9.66 (p = 0.007); HD estm. −21.12 ± 9.66 (p = 0.02). In addition, an increase in the abundance of the cp-4_5 group was recorded at both doses: LD estm. 16.36 ± 6.46 (p = 0.01), HD estm. 16.52 ± 6.46 (p = 0.01).

The effect of season (spring vs. autumn) was pronounced. In spring, an increase in the abundance of the cp-1 group was recorded, estm. 5.24 ± 1.29 (p < 0.001), accompanied by a decrease in the abundance of the cp-3 group, estm. −7.83 ± 1.86 (p < 0.001).

The effect of year (2024 vs. 2022) on the abundance of the cp-2 and cp-3 groups in 2024 was recorded as a reduced abundance of the cp-2 group, estm. −12.95 ± 4.13 (p = 0.002), and a simultaneous increase in the abundance of the cp-3 group, estm. 7.63 ± 2.41 (p = 0.002), relative to 2022.

PERMANOVA analysis (

Table 6. Results of PERMANOVA testing differences in nematode functional guilds (cp1–cp5, pp2–pp3) across treatment with Trichoderma atroviride, years, seasons and interaction.

Source of Variation	Df	Sum of Squares	R2	F	p-Value
Treatment	2	6.879	0.017	2.13	0.04 *
Year	2	7.120	0.017	2.20	0.02 *
Season	1	20.297	0.050	12.55	0.001 ***
Treatment × Year	4	5.484	0.013	0.85	0.440
Residual	230	372.052	0.903	—	—
Total	239	411.831	1.000	—	—

Levels of significance: *** p < 0.001; * p < 0.05.

) showed an effect of treatment with T. atroviride, F = 2.13 (p = 0.04), year, F = 2.20 (p = 0.02), and season, F = 12.55 (p = 0.001), on the overall diversity of nematode functional groups (Figure 8). Seasonal differences explained 5% of the total variance (R² = 0.050), whereas the effect of treatment (R² = 0.017) explained 1.7% and year explained 1.7% of the total variance. No interaction between treatment and year was observed (p > 0.05).

3.3. Nematological Indices

A global effect of treatment with T. atroviride on nematological indices was not observed (p > 0.05) (

Table 7. Results of GLMM testing the effects of treatment with Trichoderma atroviride, year, season, location, olive variety and interactions on nematological indices.

Index	Fixed Effects	Df	χ2 Value	p
MI	Year	2	6.03	0.05 *
	Season	1	14.56	<0.001 ***
	Location × Olive variety	2	10.47	0.005 **
	Location × Olive variety × Treatment	4	9.53	0.05 *
MI2-5	Year	2	6.54	0.04 *
	Season	1	4.82	0.03 *
	Location × Olive variety	2	11.35	0.003 **
	Location × Olive variety × Treatment	4	12.99	0.01 *
PPI	Location	1	16.64	<0.001 ***
	Season	1	15.33	<0.001 ***

Only significant effects and trends (p < 0.1) are shown. *** p < 0.001, ** p < 0.01, * p < 0.05. Baseline reference level for all effects is Nadin_Oblica_control_autumn_2022. Baseline covariates were included in the models but are not presented here unless significant.

). In comparison to the ref. combination, a lower initial value of PPI, estm. −1.71 ± 8.15 (p = 0.04), was recorded at the Vodnjan location.

An effect of treatment with T. atroviride was recorded in a three-way interaction (location, cultivar, and treatment) at the Nadin location for the cultivar Lastovka compared to the ref. combination, where an increase in MI values was recorded at both doses, LD estm. 3.87 ± 1.55 (p = 0.02) and HD estm. 3.24 ± 1.55 (p = 0.04), as well as an increase in MI2–5 at both doses, LD estm. 4.39 ± 1.61 (p = 0.004) and HD estm. 4.31 ± 1.61 (p = 0.01).

Furthermore, an effect was recorded in a two-way interaction (location and cultivar), whereby at the Nadin location for the cultivar Lastovka, compared to the cultivar Oblica, lower values of MI were observed, estm. −4.46 ± 1.18 (p < 0.001), as well as lower values of MI2–5, estm. −5.19 ± 1.17 (p < 0.001).

For spring compared to autumn’s affected values of nematode indices, lower values of all nematode indices were recorded: MI estm. −1.93 ± 5.08 (p < 0.001), MI2–5 estm. −1.16 ± 5.28 (p = 0.03) and PPI estm. −1.38 ± 3.53 (p < 0.001).

An effect of year (2024 vs. 2022) was recorded, whereby in 2024 the values of MI increased, estm. 1.61 ± 6.60 (p = 0.02), and those of MI2–5, estm. 1.76 ± 6.68 (p = 0.01), relative to 2022.

PERMANOVA analysis (

Table 8. Results of PERMANOVA testing differences in nematological indices across treatment with Trichoderma atroviride, years, seasons and interaction.

Source of Variation	Df	Sum of Squares	R2	F	p-Value
Treatment	2	0.585	0.006	0.72	0.331
Year	2	0.800	0.008	0.99	0.154
Season	1	5.066	0.050	12.48	0.001 ***
Treatment × Year	4	1.523	0.015	0.93	0.139
Residual	230	93.364	0.921	—	—
Total	239	101.339	1.000	—	—

Levels of significance: *** p < 0.001.

) showed an effect of season on the overall variability of the nematode indices, F = 12.48 (p = 0.001), whereby approximately 5% of the total variance was explained (R² = 0.050). An effect of treatment with T. atroviride and year was not observed (p > 0.05), nor was their interaction.

3.4. Nematode-Based Ecological Indices

Although a global effect of treatment with T. atroviride was not observed (p > 0.05), pairwise comparisons within the model showed certain differences (

Table 9. Results of GLMM testing the effects of treatment with Trichoderma atroviride, year, season, location, olive variety and interactions on nematode-based ecological indices (BI, EI, SI, and CI).

Ecological Indices	Fixed Effects	Df	χ2 Value	p
BI	Year	2	8.17	0.02 *
	Location × Olive variety	2	8.00	0.02 *
	Location × Olive variety × Treatment	4	10.80	0.03 *
EI	Season	1	9.41	0.002 **
SI	Year	2	9.61	0.008 **
	Season	1	7.95	0.005 **
	Location × Olive variety	2	10.04	0.007 **
	Location × Olive variety × Treatment	4	10.92	0.03 *
CI	Location	1	8.03	0.005 **
	Season	1	13.47	<0.001 ***

Only significant effects and trends (p < 0.1) are shown. *** p < 0.001, ** p < 0.01, * p < 0.05. Baseline reference level for all effects is Nadin_Oblica_control_autumn_2022. Baseline covariates were included in the models but are not presented here unless significant.

). Relative to the control, the HD dose increased the BI value, estm. 11.29 ± 4.82 (p = 0.02), while simultaneously reducing the EI value, estm. −16.68 ± 6.25 (p = 0.01).

Two-way interactions (location and treatment; location and cultivar) showed that at the Vodnjan location, treatment with T. atroviride reduced BI at both doses, LD estm. −14.26 ± 6.82 (p = 0.04) and HD estm. −15.57 ± 6.82 (p = 0.03), relative to the ref. combination. In Nadin, for the cultivar Lastovka compared to the cultivar Oblica, a higher BI value was recorded, estm. 17.33 ± 4.84 (p < 0.001), and a lower SI value, estm. −26.61 ± 6.91 (p < 0.001).

In a three-way interaction (location, cultivar, and treatment) at the Nadin location for the cultivar Lastovka, relative to the ref. combination, a lower BI value was recorded at both doses with T. atroviride: LD estm. −17.23 ± 6.82 (p = 0.02), HD estm. −13.90 ± 6.82 (p = 0.05). In addition, a higher SI value was recorded at both doses: LD estm. 24.75 ± 9.71 (p = 0.01), HD estm. 21.53 ± 9.71 (p = 0.03).

In spring compared to autumn, lower SI values, estm. −8.94 ± 3.17 (p = 0.007), lower CI values, estm. −20.07 ± 5.47 (p < 0.001), and higher EI values, estm. 8.93 ± 2.91 (p = 0.004) were recorded.

The effect of year (2024 vs. 2022) showed a reduction in BI, estm. −8.25 ± 2.91 (p = 0.006), and an increase in SI, estm. 12.48 ± 4.15 (p = 0.004), in 2024 relative to 2022.

PERMANOVA analysis (

Table 10. Results of PERMANOVA testing differences in soil food web indices (SI, EI, CI, and BI) across treatments with Trichoderma atroviride, years, seasons and interaction.

Source of Variation	Df	Sum of Squares	R2	F	p-Value
Treatment Year	2	3.597	0.006	0.82	0.397
	2	18.186	0.032	4.14	0.001 ***
Season	1	24.169	0.044	11.00	0.001 ***
Treatment × Year	4	3.786	0.007	0.43	0.828
Residual	230	505.037	0.910	—	—
Total	239	554.774	1.000	—	—

Levels of significance: *** p < 0.001.

) showed an effect of year, F = 4.14 (p = 0.001), and season, F = 11.00 (p = 0.001), on soil food web indices. Seasonal variability explained the largest proportion of total variance (R² = 0.044), 4.4%, while the effect of year accounted for 3.2%. The effect of treatment with T. atroviride and its interaction with year was not observed (p > 0.05).

4. Discussion

This study examined the effect of rhizosphere bioinoculation with T. atroviride (LD and HD) on the abundance, composition, and functional structure of soil nematode communities in organically managed olive orchards. The direction and intensity of the effects of T. atroviride doses at two locations (Vodnjan and Nadin) differed depending on abiotic factors as well as biotic factors, namely olive cultivars (Buža, Istarska bjelica, Oblica, and Lastovka). The results also showed that the effects of T. atroviride doses were strongly dependent on interactions (two-way and three-way) and season. In addition, the temporal component (years of investigation) was considered, allowing a longer-term assessment of the effects of different T. atroviride doses on nematode communities in agroecosystems that are subject to environmental variability.

4.1. Effect of T. atroviride Treatment on Biodiversity and Trophic Groups

In the present study, an increase in nematode diversity at the genus level was observed at both locations. However, the response to inoculation with T. atroviride was uneven. A consistent increase in the abundance of FLNs was recorded, whereas the LD dose was associated with a more pronounced reduction in the abundance of He compared to the HD dose. Furthermore, the LD dose more consistently improved trophic diversity and soil food web responses, thereby supporting functional nematode diversity. Taken together, these results indicate that the responses of soil nematode communities to T. atroviride inoculation may be dose-dependent and non-linear. Notably, the LD dose was associated with more balanced trophic interactions, while the HD dose did not necessarily result in additional improvements in functional responses.

In this context, McLean et al. [26] demonstrated that the efficacy of T. atroviride also depends on the inoculum formulation and its ability to persist in the rhizosphere. Their findings emphasised that inoculum stability in soil represents a key determinant of biological performance. Furthermore, it has been shown that the effectiveness of T. atroviride inoculation depends on the soil type, organic matter content, and nutrient availability [27]. Mukhtar et al. [88] emphasised that the optimisation of bioinoculant dosage is a key factor for successful integration into agroecosystems. Such effects indirectly contribute to increased plant growth and health and promote stable rhizosphere interactions. Moreover, Pascale et al. [89] highlighted the role of plant signalling molecules in shaping root-associated microbial communities towards beneficial assemblages. The available literature [26,27] shows partial agreement with the results of the present study. The mechanisms of action described in previous studies [88,89] are compatible with our findings. However, direct comparison at the level of nematode community structure is not possible due to their differing research objectives, which primarily focused on pathogen suppression and plant growth promotion. In this context, the effects of T. atroviride on soil nematode communities appear to be strongly influenced by the plant genotype, soil properties, initial state of the nematode community, and applied inoculum dose. Accordingly, the present field study did not quantify inoculum persistence or rhizosphere colonisation, but instead evaluated the community-level responses of soil nematodes to different inoculum doses under field conditions.

Saikai et al. [42] tested two fungal species, T. asperellum Samuels, Lieckfeldt & Nirenberg and Purpureocillium lilacinum (Thom) Luangsa-ard et al., on coffee (Coffea arabica L.) plantations, assessing their effects on PPN populations and nematode community structure in soil. The fungus P. lilacinum increased the abundance of Fu, particularly the genus Aphelenchus, suggesting that P. lilacinum represents a preferred food source for certain fungivorous genera [90]. The species T. asperellum improved soil health and enriched microbial community diversity, and both fungi showed potential in suppressing Meloidogyne hapla Chitwood, by reducing its density in soil and roots. The effectiveness of two fungal species, T. asperellum and T. harzianum, two bacterial strains of Pseudomonas fluorescens Migula, (strain 1 and 2), and the yeast Rhodosporidium paludigenum Fell & Tallman was also investigated in vivo on citrus. The species R. paludigenum reduced the abundance of Tylenchulus semipenetrans Cobb, while P. fluorescens (strain 1) reduced the abundance of the genus Xiphinema. An increase in predators (Pr; Mononchus) and omnivores (Om; Dorylaimus), as well as FLNs in general, was also recorded following the application of T. asperellum [43]. In a more recent in vivo study, El-Marzoky et al. [44] evaluated the efficacy of yeast–fungal–bacterial coinoculants on sweet orange trees (Citrus sinensis (L.) Osbeck.). Coinoculant R. paludigenum and T. harzianum reduced the abundance of T. semipenetrans, while P. fluorescens (strain 2) in combination with R. paludigenum was most effective in reducing Xiphinema spp. In contrast, Marin-Bruzos et al. [91], under laboratory conditions, reported an opposite effect when testing three Streptomyces species and two Pseudomonas species against Pratylenchus penetrans (Cobb) Filipjev & Schuurmans-Stekhoven, in onion (Allium fistulosum L.) roots, observing a reduction in P. penetrans but also in FLNs.

The reduction in He abundance following inoculation with T. atroviride at the LD dose confirms its BCA potential. This effect can be attributed to direct antagonism mediated by secondary metabolites [40,88].

The increased abundance of the PPN pp-2 group (genera Tylenchus, Paratylenchus, and Gracilacus) at both doses (LD and HD) suggests a possible reorganisation within the plant-parasitic community, whereby T. atroviride inoculation likely promoted the selective activation of genera with faster colonisation strategies and shorter life cycles. Similar results were reported by El-Marzoky et al. [44] who noted that coinoculants of T. harzianum and P. fluorescens (strain 2), as well as T. asperellum and P. fluorescens (strain 1), increased the abundance of the genus Tylenchus, confirming the sensitivity of PPNs to changes in soil microbial communities. At the same time, the same authors reported that coinoculant T. asperellum and P. fluorescens (strain 1) reduced populations of Helicotylenchus, whereas such an effect was not observed in the present study. Comparison of our findings with those of other authors working under in vivo conditions suggests that the observed differences can be attributed to differing ecological contexts, including site characteristics, physico-chemical soil properties, climatic conditions, and specific features of individual perennial cropping systems. In addition, different Trichoderma spp. inoculants and various coinoculant combinations have been investigated in the literature [42,43,44], often with differing research objectives. This explains why the mechanisms of action are partly compatible, while nematode community response dynamics are not fully comparable among studies.

Higher-dose treatment of T. atroviride inoculum reduced the abundance of opportunistic bacterivores of the cp-1 group at both locations, whereas Saikai et al. [42] reported an opposite effect, with P. lilacinum increasing cp-1 abundance. Similarly, El-Marzoky et al. [44] reported an opposite response, where coinoculant T. asperellum and P. fluorescens (strain 1) increased the abundance of the genus Rhabditis, indicating variable responses of opportunistic bacterivores depending on inoculant type and abiotic and biotic conditions. The reduction in cp-1 abundance under HD T. atroviride treatment in this study was accompanied by a shift in the soil food web from an enriched (EI) towards a basal state (BI), suggesting that higher inoculum doses promote more stable and balanced decomposition pathways in soil [33].

The effect of T. atroviride treatment at both doses (LD and HD) was absent for the cp-2–5 groups, whereas Saikai et al. [42] reported that T. asperellum treatment increased the abundance of the cp-3 and cp-4 groups and led to higher values of the nematode indices MI, MI2-5, ΣMI, and SI. A similar effect was observed in the present study, but only within interaction terms involving location, cultivar, and treatment, indicating that nematode community responses to T. atroviride inoculation depend on specific ecosystem types and biotic factors [28].

4.2. Effect of T. atroviride Treatment and Interaction on Functional Groups, Nematological and Ecological Indices

The interaction between location, cultivar, and treatment reflects a structured response of the nematode community to inoculation with T. atroviride. At the Vodnjan location, regardless of cultivar (Buža and Istarska bjelica), a reduction in the abundance of the cp-2 group and BI was recorded at both inoculation doses (LD and HD), indicating reduced activity of opportunistic bacterivores/fungivores and a shift of the community from a stressed towards a more stable and structured system [33]. This pattern of change indicates a uniform and stable initial functional state of the nematode community in both cultivars, as well as a more stable and consistent response to T. atroviride inoculation. Moreover, it also suggests lower soil heterogeneity within the orchard.

At the Nadin location, in the cultivar Lastovka, a reduction in the abundance of the cp-2 group and an increase in the cp-4_5 group were recorded at both doses (LD and HD), accompanied by increased values of maturity indices (MI and MI2–5), lower BI, and higher SI. Such a pattern of change indicates a shift of the community towards a more mature and stable system [33] dominated by K-strategists [92]. Increased microbial activity under the influence of T. atroviride promoted the development of a more complex trophic network in the Lastovka cultivar, whereby some references have proven the effect of root exudates’ interactions in rhizosphere [93,94,95]. This pattern of change indicates a strong response of a nematode community with lower initial stability and, at the same time, a greater potential for reorganisation.

The cultivar Oblica showed a milder response to inoculation, which is consistent with a more uniform and stable initial nematode community and a more favourable functional state. Observed differences are further reflected by greater soil heterogeneity within the orchard, which shapes the initial functional state of the nematode community and consequently modifies its subsequent response to bioinoculation. These results indicate that nematode community responses to T. atroviride inoculation differ depending on local soil conditions, cultivar characteristics, and production practices.

4.3. Effect of Season on Functional Groups, Nematological Indices and Ecological Indices

Season exerted the strongest influence on shaping the response of the nematode community to inoculation with T. atroviride. Nematode community population dynamics are closely linked to seasonal changes in nutrient availability [30,96], soil temperature [97], and the amount and distribution of precipitation [98,99]. The combined action of these factors determines the trophic activity and seasonal shifts of the main functional groups, which exhibit distinct patterns of representation in spring and autumn.

In the present study, during the spring period, an increase in the abundance of Ba, the cp-1 group, and EI was recorded, accompanied by a simultaneous decrease in the abundance of He and the cp-3 group. A reduction in the values of all nematode indices was also observed, including MI, MI2–5, and PPI, together with lower values of SI and CI. This pattern is consistent with the conceptual framework proposed by Ferris et al. [33], who described the dominance of the bacterial decomposition channel and the increased activity of opportunistic bacterivores under conditions of elevated availability of readily decomposable organic matter and favourable abiotic conditions. Similar seasonal patterns were reported by Natalio et al. [100] in spring wheat and winter oat systems, Jiang et al. [101] in maize crops, and Mola et al. [102] in perennial and semi-perennial agroecosystems. These studies confirm that spring conditions systematically promote a temporary “enrichment” of the soil food web because of enhanced microbial activity. In this context, our results indicate that the spring response of the nematode community represents a seasonal yet predictable functional shift within the soil food web, largely independent of crop type and local specificities.

Furthermore, during spring, an increase in the abundance of the pp-2 group was recorded, alongside a simultaneous decrease in the pp-3 group. A similar pattern was described by Natalio et al. [100], who reported an increase in the pp-2 group during spring, followed, contrary to our findings, by a gradual increase in pp-3 over time. In a study conducted by Mola et al. [102], a uniform increase in the abundance of both the pp-2 and pp-3 functional groups was also observed during spring period. This dynamic indicates a seasonal shift within the PPN community, whereby genera belonging to the pp-2 group, acting as colonisers, exhibit greater adaptability to changing soil conditions, while the pp-3 group is more strongly associated with the availability of active root systems. Partial concordance among these results suggests that spring oscillations in PPN communities are primarily driven by colonisation strategies of the pp-2 group, whereas differences in the response of the pp-3 group likely reflect specific plant species traits and soil conditions across agroecosystems.

Analysis of the temporal effect (2024 vs. 2022) revealed significant changes in the composition and functional characteristics of the nematode community that can be associated with the long-term influence of T. atroviride inoculation. In 2024, an increase in the proportion of predatory and omnivorous trophic groups was observed, which is characteristic of more stable and ecologically mature systems. At the same time, the abundance of the opportunistic cp-2 group decreased, while the cp-3 group increased, further confirming a shift towards a more complex and balanced community structure. Increases in maturity indices (MI and MI2–5), together with a reduction in BI and a concurrent increase in SI, clearly indicate the functional reorganisation of soil towards greater structural complexity and enhanced regulatory capacity of the biota. These consistent trends suggest that T. atroviride inoculation promotes gradual ecological maturation of the soil over time, contributing to a more stable and functionally diverse nematode community.

According to Neher et al. [28], the initial composition of the nematode community is an important indicator of previous disturbances and a predictor of its response within agroecosystems. The effect of T. atroviride inoculation differed between the Vodnjan and Nadin locations, indicating that the initial state of the nematode community played a key role in shaping responses to treatment. Differences in the structural and functional composition of nematodes between the locations confirm that the effect of T. atroviride was not universal but dependent on local ecological conditions, cultivar characteristics, and management practices. The changes observed between 2022 and 2024 point to gradual community maturation under the influence of T. atroviride; however, contrasts between Vodnjan and Nadin demonstrate that the direction and magnitude of this response were determined by the initial community state and site-specific ecological conditions.

5. Conclusions

The results of this study demonstrate that bioinoculation with Trichoderma atroviride significantly affected the composition and functional structure of soil nematode communities in organically managed olive orchards. Responses varied depending on local environmental conditions and cultivar-specific characteristics. The standardised inoculum dose (LD) was more effective in enhancing trophic diversity and indicators of soil food web functioning, while the higher inoculum dose (HD) did not result in additional improvements in functional responses.

A reduction in herbivorous nematodes accompanied by an increase in free-living nematodes confirms the biocontrol and bioinoculation potential of T. atroviride and its role in balancing trophic interactions in soil. Differences in responses among sites indicate that the initial state of the nematode community is an important determinant of inoculation outcomes.

Overall, these findings highlight the importance of an appropriate inoculum dose, as higher concentrations do not necessarily lead to improved functional responses of soil biota. Future research should elucidate the mechanisms underlying interactions between Trichoderma spp., soil microbiota, and plant-mediated processes under field conditions.