How Do Nematode Communities and Soil Properties Interact in Riparian Areas of Caatinga Under Native Vegetation and Agricultural Use?

Abstract

Global interest in nematode communities and their ecological relationships as unique and complex soil ecosystems has remarkably increased in recent years. As they have a representative role in the soil biota, nematodes present great potential to help understand soil health through analyzing their food chains in different environments. The objective of this study was to analyze the spatial and dynamic distributions of nematode communities and soil properties in two riparian areas of the Caatinga biome: one with native vegetation and the other with a history of agricultural use (modified). The study was carried out in a semi-arid region of Brazil in Parnamirim, PE. In both areas, sampling grids of 60 m × 40 m were established to obtain data on soil moisture, organic matter, particle size, electrical conductivity, and pH, as well as metabolic activity and ecological indices of nematode communities. There was a greater abundance and diversity of nematodes in riparian soils with native vegetation compared to in the modified area due to agricultural use and the dominance of exotic and invasive species. In both areas, bacterivores and plant-parasitic nematodes were dominant, with the genus Acrobeles and Tylenchorhynchus as the main contributors to the community. In the modified area, soil variables (fine sand, clay, and pH) positively influenced Fu4 and PP4 guilds, while in the area with native vegetation, moisture and organic matter exerted a greater influence on Om4, PP5, and Ba3 guilds. Kriging maps showed the soil variables were more concentrated in the center in the areas with native vegetation, in contrast to the area with modified vegetation, where they concentrated more on the margins. The functional guilds in the native vegetation did not exhibit a gradual increase towards the regions close to the riverbank, unlike in the modified area. The presence of plant-parasitic nematodes, especially of the genus Tylenchorhynchus, indicates the need for greater attention in the management of these ecosystems. The study contributes to understanding the interactions between nematode communities and soil in riparian areas of the Caatinga biome, emphasizing the importance of preserving native vegetation to maintain the diversity and balance of this ecosystem, in addition to highlighting the need for appropriate management practices in areas with a history of agricultural use, aiming to conserve soil biodiversity.

1. Introduction

The Caatinga covers most of the semi-arid region of Brazil and is a biome of great biological importance, considered unique in the world, with part of its biological heritage exclusive to the country [1]. Climatic characteristics include temperatures around 28 °C and scarce rainfall, ranging from 500 to 800 mm, with high spatial and temporal variability [2,3,4]. However, the Caatinga is expected to face greater aridity in the next 20 years, with an increase in duration and intensity of droughts and a reduction in wet events [5].

Anthropogenic activities have had remarkable impacts on the vegetation of the Caatinga, such as the removal and burning of native trees, as well as extensive and high-density farming of goats and cattle, leading to significant losses in soil biodiversity [6,7,8].

Riparian areas, which act as a source of protection for water bodies and have rich soil biodiversity, are being excessively suppressed [9,10,11]. Due to their greater water availability, these areas are preferable for agricultural activities in a semi-arid environment, but after periods of drought, they become unsuitable for agriculture, creating modified niches that are exhausted by intense use. With the soil exposed, exotic vegetation species, such as mesquites (Prosopis juliflora), which physiologically survive under severe water-limited conditions, dominate the area and establish themselves [12].

Ecosystems invaded by exotic plant species often show differences regarding the physical and chemical properties of the soil, resulting in changes in the composition of organisms due to interactions with a limited number of plant species [13]. These variations in soil biodiversity are crucial for understanding the ecological and evolutionary responses of terrestrial ecosystems to current and future environmental changes [14].

Several factors, such as soil type, management, cover, composition, age, nutrition, and litter depth, influence the capacity of a soil ecosystem to perform essential functions such as nutrient cycling, temperature control, and energy transfer. These factors also have an impact on the structure and maturation of the soil food chain, as well as on the distribution and abundance of nematodes in space and time in native or forest ecosystems [10,15,16,17,18].

In this context, nematodes stand out as good indicators of environmental quality, as they are abundant organisms in the soil and have different life strategies, covering all stages of the soil food web. In addition, they are sensitive to environmental changes and can be easily observed and quantified using simple laboratory techniques [19,20,21].

The study of the nematode community is carried out through morphological identification, which has led to the development of classifications that inform about feeding habits and life strategies (functional guilds) and established ecological indices [19,20,21]. These indices include the maturity index (MI), plant-parasitic nematode index (PPI), enrichment index (EI), structure index (SI), and channel index (CI), providing descriptive and quantitative information on the soil nematode community and on the conditions of ecosystems, either natural or those impacted by agricultural activities [20].

The metabolic activity of nematodes is estimated by different indices, allowing the evaluation of their contribution to the ecosystem, such as activities of enrichment, structure, plant-parasitic nematodes, bacterivores, and fungivores, as well as the composite metabolic activity [22].

Studying the nematode community and its relationships with the soil environment is of great importance, considering climatic variations, vegetation, and soil management [23,24,25]. Consequently, the need to understand nematode communities in arid and semi-arid soils has become more evident [26,27,28].

In Brazil, there is little information on the state of degradation of riparian areas and the need for immediate restoration [29]. In the Caatinga biome, knowledge about soil nematode communities, including free-living nematodes, is even more limited. Anthropogenic transformations have reduced the abundance and diversity of nematodes in the different ecosystems of the biome [30,31].

The hypotheses of the study are as follows: i. the soil food chain in native vegetation has a defined structure based on the nematode community; ii. the relationships between the nematode community and soil attributes vary according to the different vegetation conditions in the Caatinga; and iii. the spatial distribution of the functional nematode guilds differs between native and modified vegetation near the banks of the Brígida River. The objective was to analyze the modifications of the nematode community, its relationships with soil properties, and the spatial distribution of the variables in two riparian areas in the Caatinga biome, one with native vegetation and the other with a history of agricultural use (modified).

2. Materials and Methods

2.1. Study Area and Soil Sampling

The study was carried out in two distinct areas in the Brígida River Watershed located in the municipality of Parnamirim in the upper Sertão region of the state of Pernambuco, Brazil. These areas are classified as native vegetation (NV) and modified vegetation (MV) and are represented in Figure 1. The climate of the region is classified as BSwh according to Köppen’s classification, characterized as semi-arid tropical, with an average annual temperature around 26 °C, average annual rainfall of approximately 569 mm, and potential evapotranspiration of about 1600 mm. The relief is flat, with slight elevations and a predominance of crystalline geology.

Vegetation in the area is classified as hyperxerophile Caatinga, with stretches of deciduous forest, showing a predominantly shrub–tree physiognomy, with significant presence of herbaceous plants during the rainy season, according to [32].

The native vegetation (NV) area is located on the Alexandria Farm, near the Alexandria dam, at coordinates 8°7′45.08″ S and 39°38′17.45″ W. The soil in this region is classified as sandy loam, and the terrain is less rugged, typical of floodplain or bottomland areas. Vegetation found in this area is characterized as native Caatinga, consisting of arboreal, shrubby, and herbaceous plants. These species were identified through floristic surveys, consultation of taxonomic keys, and with the help of specialists, as well as through comparisons with samples available at the Dárdano de Andrade Lima herbarium of the Pernambuco State Agricultural Research Company—IPA. Among the species found, the following stand out: Myracrodruon urudeuva Allem., Schinopsis brasiliensis Engl., Herissantia tiubae (K. Schum) Brizicky, Sideroxylon obtusifolium (Humb. Ex Roem. & Schult.), Parapiptadenia zehntneri (Harms) M.P. Lima & H.C. Lima, Libidibia ferrea (Mart. ex Tul.) L.P. Queiroz, Poincianella pyramidalis (Tul.) L.P. Queiroz, Aspidosperma pyrifolium Mart., Ziziphus joazeiro Mart., Maytenus rigida Mart. and Alternanthera tenella Colla. Other species were not collected for identification as they did not have a fertile stage (presence of flowers and fruits).

The modified vegetation (MV) area is located at the Parnamirim Irrigated Agriculture Station, belonging to the Federal Rural University of Pernambuco (UFRPE), near the Fomento dam, at coordinates 8°4′53.87″ S and 39°34′36.28″ W. In this area, the terrain has a steep slope, and the soil is classified as sandy clay loam. The native vegetation has been completely suppressed over the years to allow the implementation of activities such as vegetable production and extensive goat farming. After the abandonment of these activities, arboreal vegetation developed in the area, dominated by the exotic and invasive species Prosopis juliflora (Sw) DC, known as mesquite. In addition to mesquite, only two individuals of Mimosa tenuiflora (Willd.) Poir were found.

Data collection was carried out at 35 points marked on the ground downstream of the Fomento and Alexandria dams, forming a 60 × 40 m regular grid, with 10 m spacing between points. In total, 70 disturbed soil samples were collected. Each one was in the 0.0–0.3 m layer. After collection, the materials were properly stored and sent for physical, chemical, and biological analyses.

2.2. Analysis of Environmental Variables

The collected soil was air-dried, pounded to break up clods, and processed in a 2 mm mesh sieve for analysis. Soil moisture was determined by the gravimetric method, while electrical conductivity (EC) and pH were measured using a soil saturation extract in saturated paste, with readings performed with a conductivity meter and pH meter, respectively [33].

The particle-size fractions of sand, silt and clay were established by the hydrometer method, using a mechanical shaker as a physical dispersant and sodium hexametaphosphate as a chemical dispersant, following the methodologies proposed by [34].

The amount of soil organic matter (OM) was determined by multiplying the total organic carbon (TOC) content of the soil by the van Bemmelen factor of 1.724 (100/58). This is based on the consideration that, on average, organic matter accounts for 58% of total organic carbon [35]. TOC content, in turn, was obtained by wet digestion, using a mixture of potassium dichromate and sulfuric acid, with external heating, and titration with ammoniacal ferrous sulfate in the presence of the Ferroin indicator, following a methodology adapted from [36].

2.3. Nematode Analysis

For extracting nematodes from the soil, samples of 300 cm³; of soil were collected. In the laboratory, the samples were homogenized and processed with 60- and 400-mesh sieves to extract the nematodes, using the centrifugal flotation method [37]. The suspensions obtained were placed in jars and stored in the refrigerator for a maximum of three days until counting and identification.

Population estimates were obtained by counting in 1 mL of the suspension on the Peters’ slide with the aid of a 20× optical microscope in two replicates. The results were computed as number of specimens per 300 cm³ of soil. Nematodes were identified at the genus or family level under the optical microscope with 40 and 100× objectives.

2.3.1. Analysis of Nematode Community Structure

For the study of trophic structure, all nematodes were classified according to feeding habits into five trophic groups (plant-parasitic nematodes, bacterivores, fungivores, predators, and omnivores), based on the morphology of the stoma and esophagus [38]. Plant-parasitic nematodes were identified at the genus level [39] and free-living nematodes at the genus and family level according to the identification key in [40].

Nematodes were also classified into functional guilds based on feeding habits and c-p (colonizer–persister) series, which represent life history characteristics and sensitivity to environmental disturbance and, therefore, the conditions of the surrounding environment, ranging from one (1) to five (5) for extreme r- to extreme K-strategists [19,41].

2.3.2. Ecological Indices of Nematodes

To indicate the environmental quality based on the reproductive and life history strategies of the nematode community, the following nematode-based indices were calculated: maturity index (MI) for free-living nematodes using the weighted average of individual cp values (MI = Σvi × fi, where vi is the cp value of the i-taxon, and fi is the frequency of the i-taxon); maturity index 2-5 (MI 2-5), which is similar to the MI except excluding nematodes with a cp1 value; and plant-parasitic index (PPI), calculated based on the plant-parasitic nematodes, but also based on the weighted average of the individual cp values [19,42].

To describe the soil trophic chain using the functional groups of nematodes as indicators, the enrichment (EI) and structure (SI) indices were worked out. EI was calculated using Equations (1) and (2).

$$\mathrm{E}\mathrm{I}=100\times \left({\displaystyle \frac{\mathrm{e}}{\mathrm{e}+\mathrm{b}}}\right)$$

(1)

where

$$\mathrm{e}=\mathrm{K}\mathrm{e}\times \mathrm{N}\mathrm{e}$$

(2)

in which Ke = weight assigned to the group [Ba1(Rhabditidae and Rhabditis) and Fu2 (Aphelenchus, Aphelenchoides, and Nothotylenchus)] and Ne = abundance of group e. b = Kb × Nb [Ba2 (Cephalobidae, Acrobeles and Wilsonema) and Fu2 (Aphelenchus, Aphelenchoides and Nothotylenchus)], where Kb = weight assigned to group e and Nb = abundance of these groups. SI was calculated using Equation (3).

$$SI=100\times \left({\displaystyle \frac{s}{s+b}}\right)$$

(3)

for the groups Ba3-5 (Prismatolaimus), Fu3-5 (Dorylaimoides), Om3-5 (Dorylaimidae, Dorylaimus and Mesodorylaimus), and Pr2-5 (Mononchus and Mononchulus). The channel index (CI), an indicator of the predominant decomposition pathway in the soil, was calculated using Equation (4).

$$CI=100\times {\displaystyle \frac{\left(0.8\times F{u}_2\right)}{\left(3.2\times B{a}_1+0.8\times F{u}_2\right)}}$$

(4)

The basal index (BI), which evaluates a food chain subjected to conditions of stress and resource limitation, was calculated using Equation (5) [20].

$$BI=100\times {\displaystyle \frac{b}{e+s+b}}$$

(5)

Metabolic activity, a parameter that estimates the various services and functions of the nematode and its contributions to the ecosystem, was computed according to [22], as follows: The metabolic activity of enrichment (EF) is related to nematodes that respond more quickly to the enrichment of food resources. The metabolic activity of structure (SF) may have a regulatory function in the soil food web and provides indications of the abundance of organisms with similar functions. Metabolic activities of plant-parasitic nematodes, bacterivores, microphages, and predators (PPF, BaF, FuF, and PrF) are based on energy (carbon) indicators that enter the soil food chain through their respective channels and the composite metabolic activity (CF), which considers a complete set of nematodes, regardless of trophic role or ecosystem function. It is calculated based on the number of nematodes in each taxon ($N_t$), on the estimated weight of the nematodes in μg ($W_t$), and on the cp classification of the taxon t using Equation (6).

$$F=\sum \left(N_t\left(0.1\times \left({\displaystyle \frac{W_t}{M_t}}\right)+0.273\times W_t^{0.75}\right)\right)$$

(6)

Both the ecophysiological attributes of the nematodes, grouped at the genus and family level, and the calculations of all the indices used were obtained from the NINJA—Nematode Indicator Joint Analysis program [43]—on the Nemaplex website (http://plpnemweb.ucdavis.edu/nemaplex/ accessed on 10 January 2025).

2.4. Statistical Analysis

The data were subjected to descriptive statistical analysis, including calculation of the mean, standard deviation, and coefficients of skewness, kurtosis, and variation. Next, the Kolmogorov–Smirnov test, at 5% significance level, was performed to assess the normality of the data. Data that did not follow a normal distribution were transformed to log (x + 1).

A one-way analysis of variance (ANOVA) was applied to all variables to assess the existence of significant differences in the areas studied (p < 0.05). Before ANOVA, the Bartlett test was performed to verify the homogeneity of variances. In addition, radial analysis was used to allow a quantitative visualization of metabolic activities, simplifying complexity and highlighting the relative changes in these indices in each area studied.

Redundancy analysis (RDA) was used to assess the relationships between nematodes and soil properties, being a multivariate analysis technique that uses principal component analysis and multiple regression to study the relationships between the Y data matrix (response variables) and the X data matrix (explanatory variables). In this specific case, the analysis sought to verify which soil attributes explain most of the variations in the nematode community [44]. A permutation test with 999 permutations (α = 0.05) was used to evaluate the model, and the selection of soil properties was made after redundancy analysis.

To determine the differences in the taxonomic composition of the nematode community between soil areas, a non-metric multidimensional scaling analysis (nMDS) per cluster was performed, along with an analysis of similarity (ANOSIM), using the Sørensen–Dice distance measurement. The Sørensen–Dice distance was calculated to access the presence or absence of different nematode taxa in the studied areas [44,45]. Similarity percentage analysis (SIMPER) with a 50% cut-off was performed to complement the ANOSIM, identifying which nematode taxa explain the differences between the groups. The Bray–Curtis similarity measure (multiplied by 100) was used with SIMPER [45].

ANOSIM and SIMPER analyses were conducted using PAST software version 4.04 [45], while the other analyses were performed with the aid of the R program, version 4.0.3 [46], and the ggplot2 [47], corrplot [48], and vegan [49] packages.

Finally, an analysis of spatial variability was performed using a geostatistical technique, fitting the classical semivariogram [50]. The semivariance γ(h) was calculated based on the assumption of stationarity, following Equation (7) according to [51].

$$\gamma \left(h\right)={\displaystyle \frac{1}{2N\left(h\right)}}{\sum }_{i=1}^{N\left(h\right)}{\left(Z\left(x+h\right)-Z\left(x\right)\right)}^2$$

(7)

where

$h$—spacing between samples.

$N\left(h\right)$—number of pairs.

$Z\left(x+h\right)$ and $Z\left(x\right)$—values of the property of interest at the locations $\left(x+h\right)$ and $\left(x\right)$.

The spherical, exponential, and Gaussian models were tested, according to [52]. By fitting the mathematical model to the calculated values, the coefficients of the theoretical model for the semivariogram were estimated: nugget effect (C0); sill (C0 + C1); and range (a).

The degree of spatial dependence (DSD) was classified according to [53], which is given by Equation (8).

$$DSD={\displaystyle \frac{C}{C+C0}}$$

(8)

In this model, strong dependence < 25%; moderate dependence is between 25 and 75%; and weak dependence > 75%.

The spherical, exponential, and Gaussian models fitted to the semivariograms were subjected to the jack-knifing cross-validation process [54], considering the mean values close to zero and the standard deviation close to one. R² values were also evaluated. Then, maps were constructed from the estimation of the data at unsampled sites by the kriging method. The maps were made using SURFER^® software, version 13.0.

3. Results

3.1. Environmental Variables

As shown in

Table 1. F values for environmental variables and community and nematode indices in modified and native vegetation areas in the Caatinga, within riparian forest fragments of the Brígida River Watershed, northeastern Brazil.

Variables	F	Mean Values
		MV	NV
Environmental variables
Moisture (%)	0.033 *	17.02	14.96
OM (%)	2.0 × 10−16 **	1.61	3.85
Total sand	0.865 ns	63.82	64.28
Fine sand	0.542 ns	45.16	44.02
Coarse sand	0.345 ns	18.68	20.26
Clay	2.34 × 10−10 **	27.04	13.08
Silt	1.67 × 10−9 **	9.13	22.64
pH	4.54 × 10−4 **	6.80	6.27
E.C.	0.42 ns	1.63	1.80
Nematodes
Rhabditidae	0.742 ns	101.9	144.5
Acrobeles	0.754 ns	322.94	341.7
Dorylaimus	0.069 ns	57.97	50.2
Tylenchorhynchus	0.268 ns	212.9	276.8
Trophic groups
Bacterivores	0.099 ns	450.07	591.64
Fungivores	7.64 × 10−6 **	36.09	98.01
Omnivores	0.213 ns	125.83	94.54
Plant-parasitic	0.001 **	218.39	432.89
Functional guilds
Ba1	0.181 ns	111.24	144.51
Ba2	0.353 ns	337.54	397.70
Ba3	1.04 × 10−8 **	1.29	49.43
Fu2	1.3 × 10−6 **	31.01	98.01
Om4	0.090 ns	125.83	83.91
PP3	0.008 **	214.24	380.84
PP5	2.49 × 10−8 **	2.17	51.94
Ecological indices
MI	0.614 ns	2.20	2.17
MI 2-5	0.051 ns	2.52	2.41
PPI	2.75 × 10−5 **	3.03	3.19
CI	0.062 ns	10.26	18.59
BI	0.365 ns	29.29	32.29
EI	0.91 ns	53.60	55.13
SI	0.157 ns	55.05	52.17
Metabolic footprints
CF	0.25 ns	512.1	814.7
EF	0.184 ns	115.45	154.10
SF	0.944 ns	308.03	394.92
HF	1.41 × 10−11 **	14.63	114.85
FF	6.86 × 10−4 **	2.46	9.20
BF	0.205 ns	172.51	294.71
OF	0.942 ns	308.03	445.75
Totals
TNB	0.322 ns	4.01	4.43
TND	0.010 *	830.4	1236.9
DIVER	7.13 × 10−7 **	7.17	10.4

E.C.: electrical conductivity of soil saturation extract; OM: soil organic matter; *, **, and ns: significant at 5%, significant at 1%, and not significant, respectively. Nematode-based indices are abbreviated as follows: maturity index (MI), maturity index 2–5 (MI2-5), plant-parasitic index (PPI), sigma maturity index (∑MI), enrichment index (EI), structure index (SI), channel index (CI), basal index (BI), and metabolic footprints (MFs), including the enrichment footprint (EF), structure footprint (SF), herbivore footprint (HF), bacterivore footprint (BF), fungivore footprint (FF), and omnivore footprint (OF). TNB: total nematode biomass; TND: total nematode density (total number of nematodes); DIVER: nematode diversity.

, soil moisture (p < 0.05), organic matter (OM) (p < 0.01), clay content (p < 0.01), silt (p < 0.01), and soil pH (p < 0.01) showed statistically significant differences between areas under modified vegetation (MV) and those under native vegetation (NV) (Table 1). Specifically, the lowest mean values for soil moisture (14.96%), clay content (13.08%), and soil pH (6.27) were observed in the native vegetation areas, indicating drier and less alkaline conditions in these regions. Conversely, organic matter (3.85%) and silt content (22.64%) exhibited higher average values in native vegetation areas compared to modified vegetation, suggesting greater soil fertility and finer soil texture in these locations. These differences highlight the influence of vegetation cover on key soil properties in riparian zones of the Caatinga biome (Table 1).

Table 2. Descriptive statistics of soil properties, trophic groups, and functional guilds of nematodes in modified and native vegetation areas in the Caatinga, within riparian forest fragments of the Brígida River Watershed, northeastern Brazil.

	Modified Vegetation							Native Vegetation
	Mean	SD	CV (%)	Kurtosis	Skewness	Largest Error	K-S	Mean	SD	CV (%)	Kurtosis	Skewness	Largest Error	K-S
Environmental variables
Moisture (%)	17.02	3.96	23.27	2.99	1.63	0.16	0.23	14.96	4.00	26.74	−1.11	−0.10	0.11	0.23
OM (%)	1.61	0.59	36.84	−1.22	0.15	0.13		3.85	0.61	15.77	1.18	−0.49	0.09
Total sand	63.82	15.30	23.98	1.94	−1.40	0.15		64.28	4.17	6.48	−0.49	−0.24	0.10
Fine sand	45.16	9.48	20.99	2.38	−1.36	0.18		44.02	5.59	12.69	−1.23	0.01	0.13
Coarse sand	18.68	9.41	50.40	−1.01	−0.27	0.11		20.26	2.94	14.51	−0.97	0.15	0.10
Clay	27.04	10.89	40.26	1.47	0.99	0.14		13.08	2.22	16.96	−0.14	−0.11	0.11
Silt	9.13	11.19	122.51	3.06	1.85	0.21		22.64	2.56	11.29	0.34	0.64	0.12
pH	6.80	0.49	7.14	−0.47	−0.27	0.13		6.27	0.69	11.00	−1.07	−0.16	0.14
E.C.	1.63	0.94	57.51	0.54	1.15	0.15		1.80	0.79	43.94	−1.10	0.29	0.10
Trophic groups
Bacterivores	450.07	410.35	91.17	3.75	1.86	0.20	0.23	591.64	289.46	48.92	−0.47	0.42	0.07	0.23
Fungivores	36.09	37.19	103.06	−0.14	1.04	0.17		98.01	65.83	67.17	−1.30	0.32	0.14
Predators	-	-	-	-	-	-		19.83	31.50	158.85	1.72	1.59	0.31
Omnivores	125.83	135.28	107.51	5.39	2.14	0.22		94.54	57.77	61.11	−0.90	0.27	0.10
Plant-parasitic	218.39	294.12	134.68	7.70	2.69	0.27		432.89	240.56	55.57	−0.25	0.34	0.07
Functional guilds
Ba1	111.24	103.57	93.10	0.53	1.21	0.14	0.23	144.51	102.45	70.89	−0.93	0.55	0.15	0.23
Ba2	337.54	329.92	97.74	4.13	1.89	0.15		397.70	190.27	47.84	−0.34	0.61	0.15
Ba3	1.29	4.66	362.64	13.98	3.73	0.49		49.43	43.45	87.91	1.27	1.16	0.15
Fu2	31.01	35.17	113.41	−0.04	1.09	0.19		98.01	65.83	67.17	−1.30	0.32	0.14
Fu4	5.07	14.96	295.02	8.30	3.02	0.46		-	-	-	-	-	-
Pr4	-	-	-	-	-	-		19.83	31.50	158.85	1.72	1.59	0.31
Om4	125.83	135.28	107.51	5.39	2.14	0.22		83.91	50.23	59.86	−0.52	0.35	0.08
PP3	214.24	293.22	136.86	7.95	2.74	0.27		10.63	20.09	188.97	1.42	1.63	0.42
PP4	1.97	5.40	274.12	7.66	2.80	0.47		380.94	215.49	56.57	−0.65	0.31	0.06
PP5	2.17	6.87	316.21	14.44	3.72	0.45		51.94	46.20	88.95	−0.03	0.74	0.13

SD: standard deviation; CV: coefficient of variation; K-S: Kolmogorov–Smirnov test; E.C.: electrical conductivity of soil saturation extract; OM: soil organic matter.

presents descriptive statistics comparing soil properties, nematode trophic groups, and functional guilds between modified and native vegetation areas within riparian forest fragments of the Brígida River Watershed in the Caatinga biome of northeastern Brazil.

Soil moisture was slightly higher in modified vegetation areas (17.02%) compared to native areas (14.96%), with a similar coefficient of variation (23.27% vs. 26.74%). However, organic matter content was more than twice as high in native vegetation areas (3.85%) than in modified ones (1.61%), and variability was lower in the native environment. Soil texture showed notable differences, particularly in clay content, which was considerably higher in modified areas (27.04%) compared to native areas (13.08%). In contrast, silt content was greater in native areas (22.64% vs. 9.13%). Soil pH was slightly more acidic in native areas (6.27) than in modified areas (6.80), while electrical conductivity values were relatively similar between the two environments.

Regarding nematode trophic groups, native vegetation areas showed a higher abundance of bacterivores (591.64) than modified areas (450.07), with greater relative variability in the latter. Fungivores were almost three times more abundant in native areas (98.01) than in modified ones (36.09). Predatory nematodes were absent in modified vegetation but present in native areas (19.83), although with high variability. Omnivores occurred in both environments with similar values, while plant-parasitic nematodes were more abundant in native vegetation (432.89 vs. 218.39), again with lower variation in the native environment.

As for functional guilds, the bacterivore subgroups Ba1 and Ba2 were more abundant in native areas. A particularly notable difference was observed in Ba3, which showed extremely low values in modified vegetation (1.29) compared to native areas (49.43), suggesting a possible decline or absence of certain bacterivore types in degraded environments. The fungivore group Fu2 also showed higher abundance in native areas, while Fu4 was recorded only in modified vegetation. Predatory nematodes (Pr4) were exclusive to native areas, indicating their sensitivity to environmental disturbance. Omnivores (Om4) and most plant-parasitic groups (PP4 and PP5) were more abundant in native vegetation, whereas PP3 was more expressive in modified areas, suggesting that some plant-parasitic nematodes may find favorable conditions even in disturbed environments.

3.2. Structure and Composition of Nematode Communities

In all samples, 24 nematode taxa were identified: 21 at the genus level and 3 at the family level (

Table 3. Abundance and mean dominance of nematode taxa in modified and native vegetation areas in the Caatinga, within riparian forest fragments of the Brígida River Watershed, northeastern Brazil.

Trophic Groups	FG	Mass (µg)	Modified Vegetation			Native Vegetation
			A	Mean ± SD	D (%)	A	Mean ± SD	D (%)
Free-living			21,420	611.9 ± 539.8	73.7	28,141	804.0 ± 379.5	65.01
Bacterivores			15,753	450.077 ± 410.34	54.2	20,708	591.6 ± 289.4	47.85
Rhabditidae ●	Ba1	5.03	3566.5	101.9 ± 93.98	12.27	5058	144.5 ± 102.4	11.7
Rhabditis	Ba1	7.5	327	9.34 ± 20.76	1.12	0	0.00 ± 0.00	0
Acrobeles ●	Ba2	0.6	11303	322.94 ± 312.20	38.89	11963	341.7 ± 165.9	27.63
Cephalobidae ●	Ba2	0.41	435	12.42 ± 22.16	1.49	1636	46.4± 45.9	3.78
Prismatolaimus ●	Ba3	0.41	45	1.28 ± 4.66	0.15	1730	49.4 ± 43.4	4
Wilsonema ●	Ba2	0.05	76	2.17 ± 5.73	0.26	321	9.17± 15.3	0.74
Fungivores			1263	36.08 ± 37.18	4.34	3431	98.01 ± 65.8	7.92
Aphelenchus ●	Fu2	0.23	294.5	8.41 ± 16.90	1.01	270	7.7 ± 18.2	0.62
Aphelenchoides ●	Fu2	0.14	781	22.31 ± 31.23	2.69	1453	41.5 ± 44.9	3.35
Dorylaimodes	Fu4	1.16	177.5	5.07 ± 14.96	0.61	0	0.00 ± 0.00	0
Nothotylenchus ●	Fu2	0.26	10	0.28 ± 1.69	0.03	1707.5	48.8 ± 43.8	3.95
Predators			0	0.00 ± 0.00	0	694	31.5 ± 19.8	1.6
Mononchus	Pr4	3.87	0	0.00 ± 0.00	0	130	3.7 ± 13	0.3
Mononchulus	Pr4	0.98	0	0.00 ± 0.00	0	564	16.1 ±28.2	1.3
Omnivores			4404	125.8 ± 135.28	15.15	3309	94.5 ± 57.7	7.64
Dorylaimidae	Om4	12.84	2375	67.85 ± 78.31	8.17	0	0.00 ± 0.00	0
Dorylaimus ●	Om4	39.28	2029	57.97 ± 66.18	6.98	2937	50.2 ± 83.9	6.78
Mesodorylaimus	Om5	1.31	0	0.00 ± 0.00	0	372	10.6 ± 20.1	0.86
Plant-parasitic			7643.5	218.4 ± 294.1	26.3	15151	432.8 ± 240.5	34.99
Tylenchorhynchus ●	PP3	0.23	7452	212.9 ± 292.4	25.64	9689	276.8 ± 170.7	22.38
Helicotylenchus	PP3	0.29	0	0.00 ± 0.00	0	1083	30.9 ± 37.8	2.5
Rotylenchulus	PP3	1.91	0	0.00 ± 0.00	0	1779	50.8 ± 37.9	4.11
Scutellonema	PP3	0.51	11.5	0.33 ± 1.35	0.03	0	0.00 ± 0.00	0
Xiphodorus	PP5	2.59	6	0.17 ± 1.01	0.02	0	0.00 ± 0.00	0
Xiphinema ●	PP5	5.67	70	2 ± 6.84	0.24	1818	51.9 ± 46.2	4.19
Trichodorus	PP4	1.03	69	1.97 ± 5.40	0.23	0	0.00 ± 0.00	0
Paratylenchus	PP3	0.05	0	0.00 ± 0.00	0	782	22.34 ± 28.7	1.81
Pratylenchus	PP3	0.13	35	1 ± 3.10	0.12	0	0.00 ± 0.00	0

A: abundance of nematodes in 300 cm³; of soil; Mean ± SD: mean number and standard deviation; D (%): dominance of each trophic group and taxon expressed as a percentage; FG: functional guild (combination of trophic groups and ratio of colonizers–persisters (c-p)); ●: common taxa for both conditions.

). In the area of modified vegetation (MV), 18 taxa were found (15 at the genus level and 3 families), while in the area of native vegetation (NV), 17 taxa were found (15 at the genus level and 2 families) (Table 3). Although both areas have the same number of nematode genera, six genera (Rhabditis, Dorylaimoides, Scutellonema, Xiphodorus, Trichodorus, and Pratylenchus) and one family (Dorylaimidae) were identified only in the MV area, while six genera (Mononchus, Mononchulus, Mesodorylaimus, Helicotylenchus, Rotylenchulus, and Paratylenchus) were identified exclusively in the NV area. Eleven taxa were common to both areas, including Tylenchorhynchus and Xiphinema, which were the only plant-parasitic nematodes present in both conditions evaluated (Table 3).

The dominant trophic groups in the areas were bacterivores, followed by plant-parasitic nematodes (Table 3). In the MV area, bacterivores (six taxa), fungivores (four genera), omnivores (two taxa), and plant-parasitic nematodes (six genera) were identified, whereas predators were not found. However, in the NV area, representatives of all trophic groups found in the study were identified: bacterivores (five taxa), fungivores (three genera), predators (two genera), and omnivores (two genera), in addition to plant-parasitic nematodes (five genera) (Table 3). Only bacterivores and omnivores did not show significant differences between the study conditions (p > 0.05), while the total number (p < 0.01) and diversity (p < 0.01) of nematodes obtained the highest means in the NV area (Table 1).

As for the functional guilds of free-living nematodes, Ba2 and PP3 were dominant in the two areas studied, with values of 40.6% and 25.8%, respectively, for the MV area and 32.1% and 30.8%, respectively, for the NV area. Next, the guilds Om4 (15.1%) and Ba1 (13.3%) were observed in MV, and Ba1 (11.7%) and Fu2 (7.9%) were observed in NV (Table 3 and Figure 2). However, only the guilds Ba3, Fu2, PP3, and PP5 showed significant differences (p < 0.01) between the conditions studied, with higher means in NV (Table 1).

The taxonomic composition of the nematode community differed among the vegetation types of the Caatinga, with an overall mean differentiation of 56.8% (PERMANOVA: all with p = 0.001) (Figure 3). The genera Acrobeles (22.12%) and Tylenchorhynchus (20.2%) were the ones that most contributed to the differences in the taxonomic composition of the nematode community (responsible for 42.3% of the dissimilarity) between the two study areas (

Table 4. SIMPER analysis results showing the top ten nematode taxa that contributed the most to the dissimilarity of nematode communities between modified and native vegetation areas in the Caatinga within riparian forest fragments of the Brígida River Watershed, northeastern Brazil.

Nematode	Contribution (%)	Mean Abundance
		Modified Vegetation	Native Vegetation
Vegetation: modified vs. native (overall mean dissimilarity: 56.86%)
Acrobeles	22.12	323	342
Tylenchorhynchus	20.2	213	277
Rhabditidae	9.839	102	145
Dorylaimus	6.534	58	83.9
Dorylaimidae	5.153	67.9	0
Rotylenchulus	4.798	0	50.8
Xiphinema	4.377	2	51.9
Nothotylenchus	4.329	0.286	48.8
Cephalobidae	4.164	12.4	46.7
Prismatolaimus	4.128	1.29	49.4

). These nematodes were dominant in both the MV and the NV areas, although at different levels of abundance (Table 4).

3.3. Nematoid Indices

The plant-parasitic index (PPI) was the only ecological index that showed a significant difference between the MV and NV areas (Table 1), with higher mean in the NV area. Although the enrichment index (EI) and the structure index (SI) did not show significant differences between the areas (Table 1), faunal analysis revealed that both the MV and NV soils are characterized by a low C/N ratio, being in the process of maturation of the food chain, and are nitrogen-enriched soils, with the decomposition of organic matter occurring mainly through bacteria (Figure 4).

Regarding the metabolic activities of nematodes, the indices PPF (metabolic activities of plant-parasitic nematodes), BaF (metabolic activities of bacterivores), FuF (metabolic activities of microphages), and PrF (metabolic activities of predators) were analyzed between the MV and NV areas. Only PPF and FuF showed significant differences between the areas, with mean values of 14.63 and 2.46, respectively, for MV, and 114.8 and 9.2, respectively, for NV (Table 1 and

Table 5. Descriptive statistics of ecological indices, total biomass, and metabolic activity indices of nematodes in modified and native vegetation areas in the Caatinga, within riparian forest fragments of the Brígida River Watershed, northeastern Brazil.

	Modified Vegetation							Native Vegetation
	Mean	SD	CV (%)	Kurtosis	Skewness	Largest Error	K-S	Mean	SD	CV (%)	Kurtosis	Skewness	Largest Error	K-S
Ecological indices
MI	2.20	0.22	9.91	−0.56	0.04	0.08	0.23	2.17	0.18	8.35	−0.44	−0.40	0.08	0.23
MI 2-5	2.52	0.27	10.82	5.98	1.80	0.14		2.41	0.19	7.93	−0.59	−0.25	0.11
PPI	3.03	0.06	2.09	2.51	1.95	0.37		3.19	0.20	6.41	−0.64	0.42	0.09
CI	10.26	19.04	185.62	12.07	3.36	0.30		18.59	17.69	95.17	10.36	2.94	0.21
BI	29.29	12.41	42.37	1.57	0.75	0.12		32.29	15.00	46.46	9.50	2.80	0.19
EI	53.60	18.31	34.17	−0.23	0.13	0.08		55.13	15.57	28.25	3.80	−1.81	0.17
SI	55.05	16.30	29.61	2.51	−0.82	0.12		52.17	17.80	34.12	1.07	−1.17	0.12
Metabolic footprints
CF	512.11	669.48	130.73	2.92	1.69	0.16		814.69	434.72	53.36	−0.29	0.29	0.05
EF	115.45	154.90	134.17	0.74	1.26	0.15		154.10	149.08	96.74	−0.92	0.55	0.14
SF	308.03	489.84	159.02	4.54	1.99	0.20		394.92	270.28	68.44	−0.55	0.35	0.10
PPF	14.63	30.89	211.17	5.37	2.28	0.28		114.85	67.94	59.16	−0.18	0.46	0.05
FuF	2.46	5.49	223.30	5.90	2.30	0.22		9.20	6.28	68.24	−1.06	0.40	0.10
BaF	172.51	209.58	121.49	2.12	1.56	0.15		294.71	176.54	59.91	−0.83	0.52	0.11
PPF	-	-	-	-	-	-		7.89	14.07	178.25	3.80	2.08	0.29
OmF	308.03	489.37	158.87	4.59	1.99	0.19		445.75	266.09	59.70	−0.55	0.34	0.09
TNB	4.01	3.97	99.04	3.50	1.79	0.17		4.43	2.52	56.83	−0.37	0.31	0.08
TND	830.4	694.2	83.9	0.7	1.25	0.20		1236.9	591.3	47.8	−0.05	0.44	0.058
DIVER	7.17	2.34	34.64	−0.92	0.34	0.12		10.4	2.59	24.9	0.10	−0.31	0.10

SD: standard deviation; CV: coefficient of variation; K-S: Kolmogorov–Smirnov test. Nematode-based indices are abbreviated as follows: maturity index (MI), maturity index 2-5 (MI 2-5), plant-parasitic index (PPI), sigma maturity index (∑MI), enrichment index (EI), structure index (SI), channel index (CI), basal index (BI), and metabolic footprints (MFs), including the enrichment footprint (EF), structure footprint (SF), herbivore footprint (HF), bacterivore footprint (BF), fungivore footprint (FF), and omnivore footprint (OF). TNB: total nematode biomass; TND: total nematode density (total number of nematodes); DIVER: nematode diversity.

). These indices were the ones that had the lowest values among the metabolic activities of the nematodes in the two areas; however, PPF showed a greater amplitude of values between the two areas. The mean values for BaF and PrF were close, with a lower amplitude of variation for the two areas (Figure 5).

PPI (p < 0.01) was the only ecological index that showed a significant difference between the MV and NV areas (Table 1), with a higher mean in the NV area. Even though EI and SI did not show any difference between the areas (Table 1), it was possible to see through faunal analysis that MV and NV areas are characterized by a low C/N ratio, a food chain in the process of maturation, nitrogen-enriched soils, and organic matter decomposition by bacteria (Figure 4).

Among the metabolic activity indices, only PPF and FuF differed significantly between the areas, with mean values of 14.63 and 2.46, respectively, for MV, and 114.8 and 9.2, respectively, for NV (Table 1 and Table 5). Among the metabolic activity indices, these had the lowest values in both areas, but PPF showed greater amplitude than FuF. The mean values for EF, SF, OmF, and BaF were similar (lower range of values for the two areas) (Figure 5).

3.4. Relationship Between Environmental Variables and Nematode Community Under Different Conditions in the Caatinga

In the RDA that analyzes the relationship between functional nematode guilds, environmental variables, and vegetation types, environmental variables accounted for 56.74% of the total variation in nematode guilds in different vegetation types. The first two axes explained 41.5% of the total variance.

Nematode guilds in the MV area were found in greater quantity in places with higher soil moisture, soil pH, and contents of clay and sand in all their fractions, as well as electrical conductivity (EC). On the other hand, in the NV area, nematode guilds were more associated with values of soil organic matter (OM) and silt (Figure 6A). The environmental conditions of the NV area favored the functional guilds Ba3, PP5, Fu2, Pr4, and On5, while in the MV area, the conditions were more favorable for the guilds PP4, Fu4, Ba1, and Om4 (Figure 6A). The Ba2 and PP3 guilds showed relationships with environmental conditions favored by both areas (MV and NV).

The guilds Ba3, PP5, Pr4, and On5 showed positive relationships with soil organic matter (OM). The guilds Ba1, Ba2, and Om4 showed positive relationships with soil moisture and electrical conductivity (EC). The PP4 and Mi4 guilds were exclusively found in the MV area (Table 3), which was also evident in the RDA graph. These guilds showed positive relationships with attributes related to lower porosity and higher soil density, reflected by higher particle-size fractions of clay and silt, as well as correlations with soil pH. The guilds Ba2 and PP3 showed positive relationships with moisture, coarse sand electrical conductivity (EC), organic matter (OM), and soil moisture.

In the RDA, which relates the indices of metabolic activity of soil nematodes, environmental variables, and vegetation types, the first two axes explained 42.586% of the total variance. The metabolic activity indices FuF, PrF, and PPF showed greater relationships with organic matter (OM) and with the NV area, whereas the metabolic activity indices OmF, CF, BaF, SF, and FuF were related to both areas and were more correlated with the particle-size fractions of total sand and coarse sand, as well as soil moisture (Figure 6B).

3.5. Spatial Distribution of Environmental Variables and Functional Guilds of Nematodes

The spherical, Gaussian, and exponential models were fitted and validated according to the jack-knifing criterion. The values of range (A0) for the variables that showed spatial dependence ranged from 12 to 76 m, with spatial dependence ranging from weak to moderate. These dependency range values were lower in NV. In the MV area, according to the semivariogram parameters of the environmental variables, only silt and EC showed pure nugget effect (PNE), i.e., there was no spatial dependence for these variables at the adopted distance (

Table 6. Semivariogram model and degree of spatial dependence for environmental variables and functional nematode guilds in modified and native vegetation areas in the Caatinga, within riparian forest fragments of the Brígida River Watershed, northeastern Brazil.

Variable	Fitted Model	C0	C + C0	C	A0	r2	C/[C0 + C]	DSD	Jack-Knifing
									Mean	SD
Modified Vegetation
Environmental Variables
Moisture (%)	Gaussian	5.75	17.88	12.13	51.15	1	0.678	Moderate	−0.01	0.986
OM (%)	Spherical	0.0201	0.3151	0.295	19.38	0.826	0.936	Weak	0.006	0.937
Total sand	Exponential	30.3	259.7	229.4	24.09	0.994	0.883	Weak	−0.039	0.91
Fine sand	Spherical	1.9	91.76	89.86	15.42	0.992	0.979	Weak	−0.041	1.09
Coarse sand	Spherical	15.6	90.86	75.26	20.49	0.974	0.828	Weak	−0.022	0.939
Clay	Exponential	70.01	128.9	58.89	43.05	0.786	0.457	Moderate	−0.005	0.823
Silt	NUG
pH	Gaussian	0.132	0.295	0.163	37.01	0.98	0.553	Moderate	0.001	0.976
E.C.	NUG
Functional Guilds (five most abundant and common in both studied areas)
Ba1	Spherical	2790	15,010	12,220	42.21	0.986	0.814	Weak	−0.005	0.829
Ba2	Gaussian	29,500	147,800	118,300	33.93	1	0.800	Weak	−0.104	1.411
Fu2	Exponential	113	1295	1182	21.12	0.952	0.913	Weak	−0.034	0.962
Om4	Gaussian	8710	26,360	17,650	47.70	0.881	0.670	Moderate	−0.115	1.216
PP3	Gaussian	35,700	122,500	86,800	76.92	0.962	0.709	Moderate	−0.142	1.292
Native Vegetation
Environmental Variables
Moisture (%)	Exponential	0.01	14.96	14.95	17.85	0.81	0.999	Weak	0.001	1.066
OM (%)	Exponential	0.001	0.341	0.34	21.6	0.874	0.997	Weak	0.025	1.04
Total sand	NUG
Fine sand	Exponential	20.22	41.6	21.38	46.0	0.821	0.514	Moderate	−0.006	0.879
Coarse sand	Gaussian	1.33	9.69	8.36	13.6	0.776	0.863	Weak	0.01	0.942
Clay	NUG
Silt	NUG
pH	NUG
E.C.	Exponential	0.001	0.585	0.584	18.2	0.652	0.998	Weak	−0.047	1.074
Functional Guilds (five most abundant and common in both studied areas)
Ba1	Exponential	500	10,780	10,280	24.72	0.841	0.954	Weak	0.022	1.02
Ba2	NUG
Fu2	Gaussian	10	4006	3996	12.08	0.988	0.998	Weak	0.039	0.95
Om4	NUG
PP3	Exponential	7200	52,200	45,000	28.34	0.77	0.862	Weak	0.026	1.017

C0: nugget effect; C: sill; A0: range (m); r²: coefficient of determination; DSD: degree of spatial dependence; SD: standard deviation; NUG: nugget effect; E.C.: electrical conductivity of soil saturation extract; OM: soil organic matter.

). Moisture (Gaussian model, r² = 0.999), clay (exponential model, r² = 0.786) and pH (Gaussian model, r² = 0.980) showed a moderate degree of spatial dependence (DSD), while OM (spherical model, r² = 0.826), total sand (exponential model, r² = 0.994), fine sand (spherical model, r² = 0.992), and coarse sand (spherical model, r² = 0.974) showed a weak spatial dependence. The variables soil moisture, clay, and pH showed high values of range (Table 6).

In the NV area, only moisture (exponential model, r² = 0.810), OM (exponential model, r² = 0.874), fine sand (exponential model, r² = 0.821), coarse sand (Gaussian model, r² = 0.776), and EC (exponential model, r² = 0.652) showed spatial dependence. Of these, only fine sand had moderate DSD, while the others had weak DSD (Table 6).

The five functional guilds most present in MV showed spatial dependence. The guilds Om4 (Gaussian model, r² = 0.881) and PP3 (Gaussian model, r² = 0.962) showed a moderate spatial dependence, while the guilds Ba1 (spherical model, r² = 0.986), Ba2 (Gaussian model, r² = 1), and Fu2 (exponential model, r² = 0.952) showed weak spatial dependence. In the NV area, only Ba1 (exponential model, r² = 0.841), Fu2 (Gaussian model, r² = 0.988), and PP3 (exponential model, r² = 0.77) showed weak DSD; Ba2 and Om4 showed pure nugget effect (Table 6).

The kriging maps for soil moisture in MV showed higher values in the upper part of the map on the left, in the area close to the banks of the Brígida River (Figure 7A). Soil OM showed lower values in regions farther away from the river (lower part of the map) and in the central region of the map (Figure 7B). Total sand and the sand fractions (fine sand and coarse sand) showed higher values on the right side and central-lower region of the map (Figure 7C–E), while clay showed higher values in the left region of the map (Figure 7F). The upper-central region of the map showed the highest values for soil pH (Figure 7G).

The functional nematode guilds of MV showed higher values as the region approaches the riverbanks (Figure 8), except for the Fu2 guild (Figure 8C), which obtained higher values in the central and side regions of the kriging map.

In NV, the environmental variables showed a more random distribution (Figure 9). Soil moisture did not follow the same distribution pattern observed in MV. Moisture, OM, and EC showed lower values in a strip area on the left side of the map (Figure 9A,B,E). Fine sand showed lower values in the central region of the map (Figure 9C), where coarse sand values were higher (Figure 9D).

The kriging maps for the functional guilds in the NV area showed that the behavior of these variables in relation to the spatial distribution of the maximum and minimum values was similar, especially when comparing the guilds Ba1 (Figure 10A) and PP3 (Figure 10B). The Fu2 guild (Figure 10C) showed a greater variation, with maximum values reaching a larger region on the map when compared to the Ba1 and PP3 guilds. Guilds also did not follow a gradual increase towards the regions near the riverbank (which, in this case, is located at the bottom of the map), as observed in the MV area.

4. Discussion

The history of use and the different types of vegetation in the studied areas showed different edaphic characteristics, such as moisture and availability of organic material, percentage of clay, and pH in the soils. This can consequently influence the soil fauna, decreasing or increasing soil density and diversity.

When losing native vegetation and imposed on intense agricultural cultivation systems, soil tends to manifest a new state of equilibrium, mainly due to changes in physicochemical and biological attributes [55]. This new condition may reflect an unfavorable state for conservation of the soil production capacity.

As found by [31], the history of agricultural use, responsible for causing modifications in the native vegetation of the Caatinga, negatively affected the nematode community, reducing the total abundance and diversity of these organisms. The absence of predatory nematodes, as well as the significantly lower numbers of fungivores and plant-parasitic nematodes in the area with modified vegetation, may be related, along with other factors, to the lower diversity of plant species in this location.

The higher the diversity, the more diversified are the inputs that transfer from plants to the soil system, resulting in a greater and more diverse presence of decomposing organic matter in the topsoil, as well as increases in the heterogeneity of microhabitats in complex plant assemblages [56,57,58]. Conservation of plant species and plant diversity is necessary to maintain the structure of the subterranean community and the function of the ecosystem. The greater diversity of plant species sustains higher trophic levels of nematode communities, such as predators, and indicates soils with more structured food chains [59,60]. Lower trophic levels of soil biota, such as plant-parasitic nematodes, are shown to be more responsive to changes in plant species diversity and composition than organisms of higher trophic levels in the soil food web [26,61].

In addition to greater plant diversity, the area with native vegetation presented a higher number of plantlets and a community of herbaceous and shrubby species compared to the area with modified vegetation, favoring a greater abundance of plant-parasitic nematodes. Mesquite (P. juliflora) is an invasive and dominant plant species in the modified vegetation area, adversely impacting the growth of herbaceous species and plantlets native to the Caatinga [12]. These detrimental effects are attributed to the accumulation of invasive litter, which inhibits the germination and development of native species and promotes the emergence of diseases [62]. The inhibitory effects are amplified by the production and release of allelopathic compounds from this accumulated litter layer also [63]. The increase in fine root biomass (more common in herbaceous and shrubby species and in plantlets) favors the appearance of plant-parasitic nematodes [64].

Fungivore nematodes play a crucial role in the structure of the soil food chain, being directly associated with the fungi present in this environment. These fungi form a broad symbiosis with plants in nature; thus, tree diversity exerts a significant influence on the taxonomic richness of fungi in the soil [65,66,67]. Conversely, a reduction in fungal root infection has been reported [68] due to the action of juliflorine, an antimicrobial active compound extracted from mesquite, present in the modified vegetation. This reduction may be responsible for the lower observed values of fungivore nematodes in the modified area, compared to the native vegetation (NV) area.

It is interesting to note that, even with the history of intense agricultural use, the effect of soil degradation in the modified area may have been mitigated by the establishment of invasive mesquite vegetation. This area had maturation conditions like those found in the native vegetation area, as indicated by the analysis of soil food chain conditions. Similarly, the overall structure of the soil nematode community responded differently to the expansion of native and non-native woody plants. In a succession of native angiosperms, the nematode community reflected the conditions of increasing enrichment, while under the expansion of invasive species, the nematode community became less mature, with a more basal and simplified food web [69].

In our study, the invasive and exotic species P. juliflora has been shown to be able to contribute, to a certain extent, to the regeneration of a degraded ecosystem. In addition, this plant species can survive in areas with low rainfall and dry periods, as the roots extend deep into the soil, reducing leaching, which is common in riparian areas. Even under drought conditions, P. juliflora maintain an extensive and dense canopy, promoting the accumulation of organic carbon and nutrients in the soil due to the concentration of organic matter from litter, fruits, gum, seeds, and root exudates [70].

A significantly higher abundance of nematodes was observed in a secondary forest in regeneration after the abandonment of slash-and-burn agriculture compared to a native Caatinga forest [31], supporting nematodes as good indicators of soil recovery in the Caatinga. However, contrary to our original hypothesis (native vegetation is in a structured condition), no significant differences were found between native and modified vegetation areas, according to the BI, EI, SI, and CI [20,59], suggesting no considerable changes in the soil food web functioning between the studied areas. The higher EI and lower CI values indicate greater bacterial activity in the soil organic matter decomposition in both areas, although the high EI values were driven by the great dominance of the genus Acrobeles (Ba2 guild), which characterizes basal food webs and is present under all soil environmental conditions [20].

Nevertheless, nematode families can be represented by different species in different locations. Nematodes of the family Cephalobidae can be found, in general, in various ecosystems, especially in habitats with adverse conditions and limited opportunities for colonization [71]. Thus, this family is not particularly indicative of the level of bacterial decomposition. In this study, there was a significant difference in the fungivore metabolic activity index (FuF), reflected by the greater presence of fungivore nematodes in the native vegetation area, indicating greater flow of resources through the fungal channels, when compared to the area of modified vegetation [22].

The PPI values in native vegetation do not necessarily indicate greater disturbance in this environment [19]. This result was mainly driven by the dominance of the genus Tylenchorhynchus, a group of plant-parasitic nematodes common in various environments, either causing damage to several crops [72,73] or living in environments without agricultural intervention [31,74,75].

Considering the high dominance of Tylenchorhynchus in Caatinga areas, it is important to understand its role in the soils of this biome. In studies on the potential threat of plant-parasitic nematode infestation to Indian forests [18], the consequences one may face by neglecting research on nematode infestation in the forest environment were emphasized. The author addresses the need for studies that translate the real vulnerability of these natural environments to the infestation of plant-parasitic nematodes.

As for the PPI, the higher abundance of plant-parasitic nematodes in the native vegetation influenced the difference between the areas for the PPF index, expressing greater flow of resources to the food web through herbivorous channels in areas with greater plant diversity.

Moisture and organic matter were the environmental variables that most influenced the distribution of nematode guilds in riparian soils of the Caatinga, as reported in several studies [26,31,76]. Soil organic matter in semi-arid regions, as found in the native vegetation area, is related to the death of fine roots, especially of the herbaceous stratum that does not withstand water deficit, and to the loss of foliage of shrub and tree plant species adapted to drought conditions (seasonal behavior in Caatinga areas) [31]. These common characteristics of arid and semi-arid ecosystems have a significant impact on soil nematodes. According to [77], organic matter content was the only measured variable correlated with nematode abundance; as the climate continues to change, impacts on plant distribution in these extreme environments will also affect soil nematodes.

The native vegetation, a native Caatinga forest system with no history of agricultural use, tends to accumulate a greater amount of litter in the soil, since it has higher values of organic matter compared to the modified vegetation, which has a history of agricultural use and shrubby vegetation with great dominance of the invasive exotic species (mesquite) [78].

The increase in organic matter improves soil fertility and aggregation, favoring the activities and foraging conditions of microbial communities, such as fungi and bacteria, consequently increasing the food base of free-living nematodes, bacterivores, fungivores, and, finally, omnivore and predator nematodes [76,79,80,81], as shown by the positive correlations with Ba3, Fu2, On5, Pr4, and PP5 guilds.

The guilds Ba1, Ba2, and Om4 were related to coarse sand and total sand because the size of the particles influenced the mobility of the nematodes, which move best in pores with diameters much larger than the width of their bodies [82]. The omnivores found in the areas, represented by the family Dorylaimidae, are large nematodes that require more space among soil particles to move around [21].

The maps presented indicated that, in the modified vegetation area, the gradual slope of the terrain, where the altitude is decreases toward the riverbanks, favored the relationship of moisture with most nematodes and, consequently, with the guilds. It is important to highlight that the topographic shape of the terrain in the area may have favored the spatial relationships found. In high- or medium-precipitation events, there is a marked effect on soil moisture, with infiltration occurring at greater depths throughout the soil profile and the possibility of leaching, thus carrying nutrients and soil organisms [7]. In the modified vegetation area, the slope of the terrain, and consequently the moisture gradient formed, may have favored not only moisture but also organic matter and the guilds Ba1, Ba2, Ba3, Om4, and H3, with increasing values and concentrations toward the riverbanks.

On the other hand, in the native vegetation area, having a flatter terrain (with floodplain characteristics), soil moisture remained more uniformly distributed, which was proven by the lower values of range and lower degree of dependence when compared to modified vegetation. However, it is possible to observe in the spatial distribution maps that, as found in modified vegetation, the guilds Ba1, Fu2, and PP3 had their spatial distribution influenced by soil moisture and organic matter, since these variables have higher values in coincident regions on the map.

5. Conclusions

The nematode community of riparian soils is highly affected by suppression of the native vegetation, being dominated by exotic and invasive species. Bacterivores and plant-parasitic nematodes were the dominant groups in the two areas studied, with the species Acrobeles and Tylenchorhynchus standing out. The disturbance in the modified area lowered nematode diversity and depleted predatory nematodes. The food chain of the two areas studied is in a maturation phase, despite the greater diversity of plant species in the native vegetation area.

Soil variables, such as moisture, sand, electrical conductivity, organic matter, and silt, exerted influence on most guilds in the native vegetation area, while clay, pH, and fine sand influenced only the guilds Fu4 and PP4 in the modified vegetation area.

These results demonstrate that the conversion of native vegetation to vegetation dominated by exotic species can lead to significant changes in the soil nematode community, affecting the diversity and structure of the food chain. Soil variables play an important role in determining the composition of nematode guilds, highlighting the importance of conserving native vegetation and maintaining healthy soil for preservation of soil fauna and its ecology. This information is valuable to support strategies for the sustainable management of riparian ecosystems and to understand the impacts of environmental changes on soil biodiversity.