Oribatid Mites in a Humid Mediterranean Environment under Different Soil Uses and Fertilization Management

Abstract

Measuring soil quality and the use of indicators for its evaluation is a worldwide challenge. In Garrotxa Volcanic Zone Natural Park (northeastern Spain), different parameters related to oribatid mites as indicators of soil quality were evaluated under different land uses: forest, pasture, and a biennial double-crop rotation of forage crops. In forage crops, previous fertilization management (one based on mineral fertilizers, three on cattle manure, and one using both types) was also evaluated. Three samplings (April, June, and September) were performed over one season. Fifty-four oribatid species belonging to 28 families were identified. Abundance was the lowest in June for all land uses (average of 1184 individuals m⁻²). In the study period, abundance, diversity (Shannon index, H’), and dominance (Berger–Parker index, d) varied with different land uses, with the highest values of abundance and H’ in forests (9287 individuals m⁻² and 2.19, respectively) and the lowest dominance in forests (d = 0.29) without differences between the other uses. Additionally, in the studied parameters, no differences were associated with previous fertilization management in forage crops. Hypochthoniella minutissima, Xenillus (X.) tegeocranus characterized the forest system, Epilohmannia cylindrica minima the forage crops, and Tectocepheus sarekensis the pasture. In pasture, the dominance of the parthenogenetic species Tectocepheus sarekensis raises concerns about potential management constraints.

1. Introduction

Soil quality, in terms of soil services like nutrient or carbon cycles, has increasingly become a focus of political attention [1]. The use of indicators to prevent further soil degradation is crucial in order to achieve the objective of harmonious development [2]. To address this problem in the agricultural sector, the current Common Agricultural Policy of the European Union strives to implement changes in agriculture management through an eco-scheme concept that deals with various aspects, such as the diversification of crop rotations and the maintenance of soil fertility [3], to lessen the degradation of these habitats.

Agricultural fields are home to a large web of organisms. Acari, along with Collembola, are the two major components of the soil microarthropod community. Amongst the Acari, Oribatida is the most numerous (accounting for 60–90% of all mites at a given site) and probably the most species-rich mite group. There are more than 11,000 species and subspecies of Oribatida worldwide, and over 1000 of them have been found in Spain [4]. Oribatida are small in size, typically ranging from 200 to 800 μm [5]. They are mainly located in the richest organic matter horizons of the soil [6,7]. Furthermore, they are primarily detritivorous and fungivorous [7,8,9,10,11], contributing to the regulation of fungal and bacterial populations, the transformation of organic matter, and nutrient exchange [12]. Moreover, they regulate the aforementioned nutrient cycles of the soil, including organic matter mineralization [13]. Their low metabolic rates, slow development, low fecundity, and low dispersion ability make them potential indicators of soil stress. However, up to 10% of species are obligate thelytokous parthenogenetic [14], enabling them to proliferate in disturbed environments. Oribatid mites have been previously recognized as sensitive to crop and variety selection [15], as well as to different stressful situations, such as over-fertilization [16] or the presence of bioplastics and microplastics [17]. They have also been used as bioindicators of soil fertility [18,19,20]. Nonetheless, most of the studies have been set in forest soils and in cold climates [21,22,23,24]. Only a few research works can be found in agricultural soils of Mediterranean ecosystems [25,26], although Orobitg [27] presents a comprehensive list of different oribatid species in different environments of Spain.

Garrotxa Volcanic Zone Natural Park (GVZNP) is in the center of an administrative area called Garrotxa in the northeastern part of Spain. The park is characterized by a significant stock-breeding activity (bovine, porcine, ovine, and goat). Organic materials (manure and slurry) are readily available and constitute an agronomic opportunity, favoring nutrient recycling, mainly in double-annual forage cropping systems, which are directly related to stock-breeding. Furthermore, 79% of farms engage in extensive cattle breeding [28] (either full-time or part-time), which supports the maintenance of grazing habitats. Therefore, in the GVZNP, forests, pastures, and agricultural activities coexist.

In this context, the aims of this research were as follows: (i) evaluate oribatid mites as soil quality indicators under different land uses (forage crops, pasture, and forest); (ii) in forage crops, evaluate the suitability of oribatid mites as indicators of the best fertilization management (mineral and organic fertilization) in terms of soil quality. The evaluation was conducted through different measured parameters in the oribatid population: general abundance (individuals m⁻²), diversity (Shannon index), and dominance (Berger–Parker index). For forage crops, the absolute abundance of the different species (individuals in the sampled volume) was included. Our hypothesis was that the studied parameters would act as early indicators of potential soil stress or degradation under different land uses and under different fertilization managements in forage crops.

2. Materials and Methods

2.1. Experimental Site

This research was set up in Garrotxa Volcanic Zone Natural Park (Catalonia, NE, Spain).

The climate is Mediterranean humid. The annual average temperature is 12.4 °C, annual rainfall oscillates between 900 and 1000 mm, and crop reference evapotranspiration (ET_o, Penman–Monteith, [29]) is close to 900 mm. In this study, weather data (Figure 1) were obtained from La Vall d’en Bas (Girona, Spain) automatic meteorological station located 5.5 km southwest of the study site [30].

In the zone where the present research was established, three land uses were present: forest, pasture, and forage crops. Forest soil has been classified as Typic Udorthents [31]. It has an upper organic layer (0–1 cm). The superficial layer (1–8 cm) has a loam texture, a pH of 8.0 with an important presence of calcium carbonate, and an organic carbon content between 10 and 20 g kg⁻¹ [32]. Agricultural soils (pasture and forage crops) have been classified as Fluventic Eutrudept [31]. The surface horizon (0–30 cm) has a pH ranging from 8.2 (forage crops) to 8.4 (pasture), 30 g kg⁻¹ of calcium carbonate, organic carbon content from 11 g kg⁻¹ (forage crops) to 15 g kg⁻¹ (pasture), and the soil texture is sandy loam.

2.2. Management History and Fertilization Practices

The studied areas devoted to the three different land uses (forest, pasture, and forage crops) were located within 200 m of each other (Figure 2). Coordinates were 42°08′29″ N, 2°30′02″ E, altitude 561 m a.s.l. for forest; 42°08′33″ N, 2°30′05″ E, altitude 532 m a.s.l. for pasture; and 42°08′32″ N, 2°30′10″ E, altitude 534 m a.s.l for forage crops.

The forest plot vegetation is representative of the dominant forest community in the area, which is the Isopyro-Quercetum roboris, where Quercus robur is the predominant species. Other present tree species include Tilia cordata, Acer campestre, and Fraxinus excelsior. The shrub layer is dominated by Corylus avellana, and it also includes the Euonymus europaeus, Ilex aquifolium, Cornus sanguinea, and Crataegus monogyna. The herbaceous layer features hydrophilic species such as Pulmonaria afinis, Anemone nemorosa, and Hepatica nobilis. Other species, such as Brachypodium sylvaticum, Stellaria holostea, or Fragaria vesca, are also present. The forest structure is an irregular mass with a density of around 800 trees ha⁻¹, with irregular growing trees having been removed. The oldest trees are less than 80 years old.

The pasture plot was established in a neighboring field, which was converted from forest into pasture four seasons before this study started. One month after tree-filling (in April), daily cattle manure was applied at a rate of 27.4 Mg ha⁻¹, which equaled 99.4 kg N ha⁻¹. Immediately after application, cattle manure was buried using a disk harrow, with no other fertilizer added. Additionally, a mix of grass species (Bromus erectus, Lolium perenne, Lolium multiflorum, Dactylis glomerata, Festuca arundinacea, Phleum pratense) and leguminous crops (Medicago sativa, Trifolium repens, Trifolium pratense, Vicia sativa, Lotus corniculacus) was sowed that month. The pasture was resowed three years later in October (prior to the first Oribatida sampling period, which commenced the following April). In October, four species were sown: Festuca arundinacea, Dactylis glomerata, Trifolium repens, and Trifolium pratense. The sowing rate was 10, 15, 5, and 3 kg of seed ha⁻¹, respectively. The percentage of seeds for each species was 31, 45, 14, and 10%, respectively. The pasture was usually harvested three times a year, the first harvest taking place in mid-May, the second one in mid-July, and the third one at the beginning of October.

The forage crops were included in a biennial rotation established six years prior to this study. The biennial rotation management included a fertilization experiment. In the first biennial rotation, oat (Avena sativa L.)–sorghum (Sorghum bicolor L.) was the crop sequence, and in the second one, the sequence ryegrass (Lolium multiflorum L.)–maize (Zea mays L.) was established. Winter crops (oat or ryegrass) were maintained in the field from October to May. Summer crops (sorghum or maize) were maintained from May to October. Until the establishment of the biennial rotation and the associated fertilization experiment, nutrients were applied exclusively through mineral fertilizers. The fertilization experiment followed a randomized complete block design with three replicates. Five fertilization treatments were established using mineral fertilizers and/or cattle manure (

Table 1. Description of the previous fertilization treatments: at sowing with mineral N fertilizer (MN) ¹ or cattle manure (CM) ², and at side-dressing ³ with mineral N; in the double-annual cropping system (oat–sorghum or ryegrass–maize). Average soil organic matter (OM) ⁴ (±standard deviation) after six biennial rotations and at the start of this Oribatida study is included for each of the former fertilization treatments.

Crop Sequence/Treatment	Winter Crop (Ryegrass or Oat) (kg N ha−1)	Summer Crop (Maize or Sorghum) (kg N ha−1)	Annual Fertilization (kg N ha−1)	Soil Organic Matter (g kg−1)
Mineral (250 MN) 1	100 MN	150 MN	250	19.2 (±1.1)
Manure (170 CM)	0	170 CM	170	18.7 (±0.6)
Manure (170 CM) + Mineral (160 MN)	60 MN	170 CM + 100 MN 3	330	20.0 (±0.8)
Manure (250 CM)	100 CM	150 CM	250	22.6 (±1.3)
Manure (500 CM)	250 CM	250 CM	500	26.7 (±0.8)

¹ Number ahead of the MN acronym indicates the N applied (kg N ha⁻¹) as ammonium nitrosulfate (26% N). ² Number ahead of the CM acronym indicates the nitrogen applied (kg N ha⁻¹) from cattle manure. ³ Number ahead of the MN acronym indicates the nitrogen applied (kg N ha⁻¹) as urea (46% N) at topdressing. ⁴ Averages are obtained after six biennial rotations. No fertilization was applied in the following biennial rotation (oat–sorghum).

). Fertilizers were applied at the sowing of the winter crops and/or at the sowing of the summer crops. According to the fertilization treatment, an additional topdressing fertilization was added to summer crops. The fertilization treatments were maintained for six years until the start of this study. During the period of this study, no fertilization was applied to the established forage crops. Therefore, as explained, fertilization treatments refer to the former ones, which were maintained for a period of six years in the established biennial rotations until this study on oribatid mites as soil quality indicators started. Thus, this study includes the consolidated or residual effect on oribatid mites of different previous fertilization managements in forage crops. During the period of this study, the crop sequence was oat–sorghum. Oat was sowed on 27 December, and it was harvested on 23 May. Sorghum was sowed on 12 June, but it was not harvested in October due to the damage caused by a major storm.

2.3. Description of the Experiment

Sampling dates for oribatid mites’ extraction in all soil uses (forest, pasture, and forage crops) were on 1 April, 3 June, and 24 September. Soil samples were taken at 0–5 cm depth with three soil cores (6 cm in diameter with steel bores).

In forage crop land use, the previous fertilization experiment included three blocks (three replicates or three plots of every five treatments). Three samples for each plot were obtained on each sampling date. The above implies that a total of nine samples (141.37 cm³ each) were studied for each fertilization treatment. The surface occupied for each plot was 50 m⁻², and they were oriented northwest to southeast (Figure 2), according to the shape of the experimental field.

Similarly, in forest and pasture, three areas (replicates) of a similar size as those in the forage plots were defined. In the forest, replicates were oriented northeast to southwest. In the pasture, the replicates were located in the middle of the field and orientated from north to south according to the shape of the field (Figure 2).

On each sampling date, additional soil samplings for gravimetric analysis of soil moisture (drying at 105 °C for 24 h) were collected in all plots. In April, a quantification of soil bulk density (0–5 cm depth) was performed using soil cores similar to those used for oribatid samplings. Soil was also dried at 105 °C for 24 h and weighted.

Oribatids were extracted with a modified Berlese–Tullgren funnel over a period of seven days. The biota obtained was stored in 70% ethanol. Oribatid individuals were counted and identified using an Olympus BX51 (Evident, Tokyo, Japan) microscope or the Olympus SZX16 (Olympus España, Barcelona, Spain) stereomicroscope according to the characteristics that needed to be evaluated. If necessary, in order to better identify taxonomic characters, individuals were immersed in lactic acid. Adult oribatids were identified, when possible, at the species level using several taxonomic keys [33,34,35,36,37,38]. The systematic ordination appearing in Subías [4] was also followed. Juvenile stages were neither identified nor included in statistical analyses.

For each month and for each land use, three variables were calculated: the mean abundance of oribatids (individuals per m²), the Shannon index of diversity (H’, [39]), and the Berger–Parker index of dominance (d, [40,41,42]). The Shannon index of diversity (H’) was calculated as: $H’=-\sum p_i\mathit{ln}\left(p_i\right)$, where p_i represents the number of individuals in ith group relative to the total number of individuals found in each sample. The Berger–Parker index of dominance (d) was calculated as the quotient between the number of individuals of the most abundant species and the total number of individuals found in each sample.

Furthermore, for the April sampling and for forage crops, the oribatid mite abundance was calculated for each species as absolute abundance (number of individuals for each species in each sampled volume).

2.4. Data Treatment

Soil moisture at different sampling dates and for different land uses was analyzed using the GLM procedure from the statistical package SAS (v9.4) [43].

The mean abundance of oribatids and the Shannon and Berger–Parker indexes required a Box–Cox transformation of data in order to be normalized. The best λ values were −0.121, 0.127, and 0.742 for abundance, diversity, and dominance, respectively. The transformation was performed using JMP^® Pro 17 statistical software [44]. Once data were normalized, the statistical analyses were performed using the statistical package SAS [43].

The abundance of oribatids and H’ and d data from the three samplings (April, June, and September) were analyzed using linear mixed models accounting for the repeated measures [45]. The MIXED procedure from the statistical package was used. The best model was selected using the Akaike information criterion (AIC; [46]). We fitted linear mixed models with fertilization (or with uses), time, and their interaction as fixed effects. Block and its interactions were considered random effects. Additionally, different types of variance–covariance matrices were tested to model the correlation structure of data from the repeated measures performed. The significance of the fixed effects was tested using F-tests, and the random effects were assessed using likelihood ratio tests. Comparisons of the least squares mean of the main effects were made with the LSMEANS option. We selected a value of 5% (i.e., p < 0.05) as the minimum criterion for significance. The first analysis included the forage crops and the five former fertilization treatments in a randomized block structure. The second analysis included the three land uses: forest, pasture, and forage crops, and the three replicates for each use. For this second analysis and based on the results from the first analysis (see Section 3.1), only one fertilization treatment (250 CM) was considered for the forage crops. The treatment 250 CM was chosen because it also allowed high yields and optimized nutrient recycling within the farming system [47].

A principal component analysis (PCA) with JMP^® Pro 17 was performed to study the Oribatida community structure (linear response model). Matrix distances were computed only for species represented by more than 5 specimens. From the most relevant species from the PCA, a cluster analysis was set up with seven species using the centroid method (unweighted pair group method using centroids), and distance data were squared. The SAS statistical software was used for the cluster analysis.

In April sampling and for forage crop land use, the number of individuals for the three most representative species (absolute abundance) was analyzed using a generalized mixed model. Time was not included as a factor in this analysis, as only data from April were used when the highest number of individuals was quantified for this land use (see Section 3.1). The GLIMMIX procedure from the SAS statistical package was chosen. Thus, the model included former fertilization treatments, oribatid species, and their interaction as fixed factors, with block as a random effect.

3. Results

In forage crops, during the studied oat–sorghum biennial rotation period (from October to September), maximum rainfall was recorded in November (173.4 mm) and August (175.8 mm), and the highest monthly mean temperature was recorded in August (19.9 °C, Figure 1). During the sampling period from April to September, the maximum ET_o was recorded in June (123.5 mm, Figure 2).

During the sampling period, soil moisture was higher in the forest than in the other land uses (Figure 3). Pasture and forage crops only differed in September (9.3% and 14.5%, respectively.

Summer rainfall (Figure 1) allowed for four harvests in pastures: 16 May, 3 July, 8 August, and 19 September.

Average soil bulk density (±standard deviation) was 1046 (±76) kg m⁻³, 1593 (±108) kg m⁻³, and 1325 (±81) kg m⁻³ for forest, pasture, and forage crops, respectively.

3.1. Oribatid Mite Diversity Indices at Study Sites

In forage crops, the previous fertilization treatments did not result in significant differences in abundance, in H’ or in d values (

Table A1. Analysis of variance for mean abundance, the Shannon index of diversity, the Berger–Parker index of dominance over time in the forage crops, and for the former five different fertilization treatments. The Box–Cox transformation was applied to the data.

Variable		Abundance		Diversity		Dominance
Source	df	Den df	p	Den df	p	Den df	p
Time	2	18	0.202	16	0.659	17	0.209
Block	2	18	0.489	16	0.089	17	0.119
Treatment	4	18	0.097	16	0.970	17	0.876
Time × Treatment	8	18	0.087	16	0.455	17	0.077

). However, different land uses exhibited significant differences in all three studied parameters. Additionally, abundance changed significantly with time (

Table A2. Analysis of variance for mean abundance, the Shannon index of diversity, and the Berger–Parker index of dominance over time for the three uses: forest, pasture, and forage crops (under the former 250 CM fertilization management). The Box–Cox transformation was applied to the data.

Variable		Abundance		Diversity		Dominance
Source	df	Den df	p	Den df	p	Den df	p
Time	2	12	0.005	10	0.187	11	0.178
Uses	2	12	0.014	10	0.001	11	0.001
Rep (Use)	6	12	0.469	10	0.435	11	0.067
Time × Use	4	12	0.261	10	0.614	11	0.479

). Forest attained the highest abundance (Figure 4a) and diversity (Figure 4b) and lowest dominance (Figure 4c). In June, the lowest abundance was recorded for all land uses, with an average of 1310 individuals m⁻² (Figure 4a).

3.2. Patterns among Oribatid Species

A total of fifty-four oribatid mite species belonging to 28 families were identified. The number of species present in April, June, and September was 36, 21, and 46, respectively (

Table A3. Oribatid families and species (April samplings) for different uses (forage crops, pasture, and forest) and former fertilization treatments ¹ in forage crops.

Family	Species	250 MN	170 CM	170 CM + 160 MN	250 CM	500 CM	Pasture	Forest
Hypochthoniidae Berlese, 1910.	Hypochthonius luteus (Oudemans, 1917)	-	-	-	-	-	-	1
Eniochthoniidae Grandjean, 1947.	Hypochthoniella minutissima (Berlese, 1904)	-	-	-	-	-	-	2
Epilohmanniidae Oudemans, 1923.	Epilohmannia (E.) cylindrica cylindrica (Berlese, 1904)	-	-	-	-	-	-	4
	Epilohmannia (E.) cylindrica minima (Schuster, 1960)	-	-	-	-	-	-	1
Euphthiracaridae Jacot, 1930.	Acrotritia ardua (Koch, 1841)	-	-	-	-	-	-	2
	Acrotritia ardua americana (Ewing, 1909)	1	1	-	-	2	4	18
Oribotritiidae Balogh, 1943.	Paratritia baloghi (Moritz, 1966)	-	-	-	-	-	-	1
Phthiracaridae Perty, 1841.	Phthiracarus sp.	-	-	-	-	-	-	1
	Steganacarus sp.	-	-	-	-	-	-	2
Astegistidae Balogh, 1961.	Cultroribula bicultrata (Berlese, 1905)	-	-	-	-	-	-	1
Xenillidae Woolley and Higgins, 1966.	Xenillus (X.) tegeocranus (Hermann, 1804)	-	-	-	-	-	-	6
Damaeolidae Grandjean, 1965	Fosseremus laciniatus (Berlese, 1905)	2	3	-	5	1	-	4
	Fosseremus sp.	-	1	-	1	-	-	-
Autognetidae Grandjean, 1960.	Cosmogneta impedita (Grandjean, 1960)	-	-	-	-	-	-	6
Oppiidae Sellnick, 1937.	Ramusella (Insculptoppia) insculpta (Paoli, 1908)	7	-	1	8	3	2	7
	Tainsculptoppia subiasi (Pérez-Íñigo jr., 1990)	-	-	-	-	-	-	5
	Rhinoppia (R.) subpectinata (Oudemans, 1900)	-	-	-	-	-	-	35
	Berniniella (B.) bicarinata (Paoli, 1908)	-	-	-	-	-	-	16
	Berniniella (B.) latidens (Subías, Rodríguez and Mínguez, 1987)	-	-	-	-	-	-	1
	Lauroppia similifallax (Subías and Mínguez, 1986)	-	-	-	-	-	-	1
	Oppiella (O.) nova (Oudemans, 1902)	14	11	1	85	4	-	23
	Ramusella (Insculptoppia) sp.	-	-	-	1	3	-	-
	Rhinoppia sp.1/Rhinoppia sp.2	-	-	-	-	-	-	3/8
Quadroppiidae Balogh, 1983.	Quadroppia (Q.) obsoleta (Mínguez, Ruiz, and Subías, 1985)	-	-	-	-	-	-	1
Suctobelbidae Jacot, 1938.	Suctobelbella (Flagrosuctobelba) nasalis (Forsslund, 1941)	-	-	-	-	-	-	2
	Suctobelbella (S.) sp.	-	-	-	-	-	-	1
Carabodidae Koch, 1837.	Carabodes coriaceus (Koch, 1835)	-	-	-	-	-	-	1
Tectocepheidae Grandjean, 1954.	Tectocepheus sarekensis (Trägårdh, 1910)	2	10	4	6	1	92	5
	Tectocepheus minor (Berlese, 1903)	-	-	-	2	-	15	16
Achipteriidae Thor, 1929.	Achipteria (A.) coleoptrata (Linnaeus, 1758)	-	-	-	-	-	-	4
Ceratozetidae Jacot, 1925.	Zetomimus (Protozetomimus) acutirostris (Mihelčič, 1957)	17	15	18	27	14	-	-
Oribatulidae Thor, 1929.	Oribatula (O.) tibialis (Nicolet, 1855)	-	-	-	-	-	-	12
	Oribatula (Zygoribatula) undulata (Berlese, 1916)	-	-	-	-	-	41	-
Scheloribatidae Grandjean, 1933.	Scheloribates (S.) barbatulus (Mihelčič, 1956)	-	-	-	-	-	2	-
Protoribatidae J. and P. Balogh, 1984.	Protoribates (P.) capucinus (Berlese, 1908)	-	-	-	-	-	-	3

¹ Numbers behind the MN and CM acronyms indicate the mineral or the organic N applied (kg N ha⁻¹), respectively, in forage crops.

,

Table A4. Oribatid families and species (June samplings) for different uses (forage crops, pasture, and forest) and former fertilization treatments ¹ in forage crops.

Family	Species	250 MN	170 CM	170 CM + 160 MN	250 CM	500 CM	Pasture	Forest
Hypochthoniidae Berlese, 1910.	Hypochthonius luteus (Oudemans, 1917)	-	-	-	-	-	-	3
Eniochthoniidae Grandjean, 1947.	Hypochthoniella minutissima (Berlese, 1904)	-	-	-	-	-	-	1
Epilohmanniidae Oudemans, 1923.	Epilohmannia (E.) cylindrica minima (Schuster, 1960)	-	-	-	-	-	-	6
Euphthiracaridae Jacot, 1930.	Acrotritia ardua (Koch, 1841)	-	-	-	-	-	4	8
	Acrotritia ardua americana (Ewing, 1909)	-	1	-	-	2	-	-
Phthiracaridae Perty, 1841.	Steganacarus sp.1	-	-	-	-	-	-	2
Oribellidae Kunst, 1971.	Pantelozetes paolii (Oudemans, 1913)	-	-	-	-	-	-	2
Damaeolidae Grandjean, 1965.	Fosseremus laciniatus (Berlese, 1905)	1	1	3	2	4	-	1
Autognetidae Grandjean, 1960.	Cosmogneta impedita (Grandjean, 1960)	-	-	-	-	-	-	2
Oppiidae Sellnick, 1937.	Ramusella (Insculptoppia) insculpta (Paoli, 1908)	9	-	-	1	1	-	2
	Microppia minus longisetosa (Subías and Rodríguez, 1988)	-	-	-	-	-	-	4
	Rhinoppia (R.) subpectinata (Oudemans, 1900)	-	-	-	-	-	-	6
	Berniniella (B.) latidens (Subías, Rodríguez, and Mínguez, 1987)	-	-	-	-	-	-	1
	Oppiella (O.) nova (Oudemans, 1902)	5	1	1	1	-	-	7
Suctobelbidae Jacot, 1938.	Suctobelbella (Flagrosuctobelba) alloenasuta (Moritz, 1971)	-	-	-	-	-	-	1
Tectocepheidae Grandjean, 1954.	Tectocepheus sarekensis (Trägårdh, 1910)	25	9	11	12	17	18	-
	Tectocepheus minor (Berlese, 1903)	-	-	-	4	-	-	4
Achipteriidae Thor, 1929.	Achipteria (A.) coleoptrata (Linnaeus, 1758)	-	-	-	-	-	-	1
Ceratozetidae Jacot, 1925.	Zetomimus (Protozetomimus) acutirostris (Mihelčič, 1957)	3	2	6	-	9	-	-
Chamobatidae Thor, 1937.	Chamobates (C.) confusus (Subías, 2000)	-	-	-	-	-	-	2
Oribatulidae Thor, 1929.	Oribatula (Zygoribatula) undulata (Berlese, 1916)	-	-	-	-	-	2	3

¹ Numbers behind the MN and CM acronyms indicate the mineral or the organic N applied (kg N ha⁻¹), respectively, in forage crops.

and

Table A5. Oribatid families and species (September samplings) for different uses (forage crops, pasture, and forest) and former fertilization treatments ¹ in forage crops.

Family	Species	250 MN	170 CM	170 CM + 160 MN	250 CM	500 CM	Pasture	Forest
Hypochthoniidae Berlese, 1910.	Hypochthonius luteus (Oudemans, 1917)	-	-	-	-	-	-	3
Eniochthoniidae Grandjean, 1947.	Hypochthoniella minutissima (Berlese, 1904)	-	-	-	-	-	-	5
Brachychthoniidae Thor, 1934.	Poecilochthonius spiciger (Berlese, 1910)		-	-	-	-	-	3
Epilohmanniidae Oudemans, 1923.	Epilohmannia (E.) cylindrica cylindrica (Berlese, 1904)	-	-	-	-	-	-	13
Euphthiracaridae Jacot, 1930.	Acrotritia ardua (Koch, 1841)	-	-	-	-	-	-	6
	Acrotritia ardua americana (Ewing, 1909)	-	1	4	-	3	3	13
Phthiracaridae Perty, 1841.	Atropacarus phyllophorus (Berlese, 1904)	-	-	-	-	-	-	2
	Phthiracarus sp.	-	-	-	-	-	-	4
	Steganacarus sp.1/Steganacarus sp.2	-	-	-	-	-	-	2/1
Malaconothridae Berlese, 1916.	Malaconothrus (M.) monodactylus (Michael, 1888)	-	-	-	-	-	-	1
Astegistidae Balogh, 1961.	Cultroribula bicultrata (Berlese, 1905)	-	-	-	-	-	-	4
Xenillidae Woolley and Higgins, 1966.	Xenillus clypeator (Robineau-Desvoidy, 1839)	-	-	-	-	-	-	1
	Xenillus (X.) tegeocranus (Hermann, 1804)	-	-	-	-	-	-	1
Oribellidae Kunst, 1971.	Pantelozetes paolii (Oudemans, 1913)	-	-	-	-	-	-	4
Damaeolidae Grandjean, 1965.	Fosseremus laciniatus (Berlese, 1905)	1	1	-	-	-	-	9
Autognetidae Grandjean, 1960.	Cosmogneta impedita (Grandjean, 1960)	-	-	-	-	-	-	1
Oppiidae Sellnick, 1937.	Ramusella (Insculptoppia) insculpta (Paoli, 1908)	5	-	2	-	1	-	17
	Tainsculptoppia subiasi (Pérez-Íñigo jr., 1990)	-	-	-	-	-	-	12
	Microppia minus longisetosa (Subías and Rodríguez, 1988)	-	-	-	-	-	-	133
	Rhinoppia (R.) subpectinata (Oudemans, 1900)	-	-	-	-	-	-	29
	Berniniella (B.) bicarinata (Paoli, 1908)	-	-	-	-	-	-	5
	Berniniella (B.) latidens (Subías, Rodríguez, and Mínguez, 1987)	-	-	-	-	-	-	13
	Oppiella (O.) nova (Oudemans, 1902)	8	3	8	5	31	-	37
	Rhinoppia sp.1/Rhinoppia sp.2/Rhinoppia sp.3	-	-	-	-	-	-	5/6/2
Quadroppiidae Balogh, 1983.	Quadroppia (Q.) obsoleta (Mínguez, Ruiz, and Subías, 1985)	-	-	-	-	-	-	1
Suctobelbidae Jacot, 1938.	Suctobelbella (S.) acutidens sarekensis (Forsslund, 1941)	-	-	-	-	-	-	11
	Suctobelbella (S.) subcornigera (Forsslund, 1941)	-	-	-	-	-	-	7
	Suctobelbella (Flagrosuctobelba) nasalis (Forsslund, 1941)	-	-	-	-	-	-	12
	Suctobelbella (Flagrosuctobelba) alloenasuta (Moritz, 1971)	-	-	-	-	-	-	2
	Suctobelbella (S.) sp.	-	-	-	-	-	-	11
Carabodidae Koch, 1837.	Carabodes coriaceus (Koch, 1835)	-	-	-	-	-	-	1
Tectocepheidae Grandjean, 1954.	Tectocepheus sarekensis (Trägårdh, 1910)	7	3	9	9	13	59	8
	Tectocepheus minor (Berlese, 1903)	2	-	-	11	2	1	37
Phenopelopidae Petrunkevitch, 1955.	Eupelops occultus (Koch, 1835)	-	-	-	-	-	1	-
Microzetidae Grandjean, 1936.	Microzetes (M.) pyrenaicus (Travé, 1956)	-	-	-	-	-	-	3
Achipteriidae Thor, 1929.	Achipteria (A.) coleoptrata (Linnaeus, 1758)	-	-	-	-	-	-	14
Oribatellidae Jacot, 1925.	Joelia fiorii (Coggi, 1898)	-	-	-	-	-	-	1
Ceratozetidae Jacot, 1925.	Hispanozetes aragonensis (Pérez-Íñigo jr., Herrero, and Pérez-Íñigo, 1988)	-	-	-	-	-	-	3
	Zetomimus (Protozetomimus) acutirostris (Mihelčič, 1957)	1	1	2	-	-	-	-
Punctoribatidae Thor, 1937.	Mycobates (M.) parmeliae (Michael, 1884)	-	-	-	-	-	-	2
Oribatulidae Thor, 1929.	Oribatula (O.) tibialis (Nicolet, 1855)	-	-	-	-	-	-	14
	Oribatula (Zygoribatula) connexa connexa (Berlese, 1904)	-	-	-	-	-	4	1
Protoribatidae J. and P. Balogh, 1984.	Protoribates (P.) capucinus (Berlese, 1908)	-	-	-	-	-	-	1

¹ Numbers behind the MN and CM acronyms indicate the mineral, or the organic N applied (kg N ha⁻¹), respectively, in forage crops.

).

For the PCA analysis, 30 species were included, excluding those represented by fewer than five specimens. The PCA revealed that 75.21% of data variability was explained by the first two principal components and 98.06% by the first seven components (

Table 2. Ordination of oribatid mite species according to the principal component analysis. Species with the largest absolute eigenvector values are noted.

Axes	1	2	3	4	5	6	7
Eigenvalue	17.89	3.92	2.46	2.28	1.24	0.38	0.27
Accumulated percentage of fitted variance	61.69	75.21	83.69	91.54	95.83	97.15	98.06
Species	H. minutissima	X. tegeocranus	E. minima	T. sarekensis	O. nova	Z. acutirostris	R. insculpta
Eigenvector	0.24	0.43	0.48	0.55	0.52	0.63	−0.86

, Figure 5). The species with the highest eigenvector value for each of the seven principal components were, in order: Hypochthoniella minutissima, Xenillus (X.) tegeocranus, Epilohmannia cylindrica minima, Tectocepheus sarekensis, Oppiella (O.) nova, Zetomimus (Protozetomimus) acutirostris, and Ramusella (Insculptoppia) insculpta.

The hierarchical clustering analysis of the seven species from the PCA established two main groups. The first group consisted of T. sarekensis, while the second one included the remaining species (Figure 6). Furthermore, H. minutissima and X. tegeocranus were found only in the forest. Z. acutirostris was present only in forage crops. O. nova was found in both forage crops and the forest but not in the pasture. T. sarekensis was present everywhere, but it predominated in pasture.

In April, within the forage crops, the absolute abundance of the three most representative species was significantly different (

Table 3. Differences in absolute abundance (associated probability) between species in forage crops during the April sampling.

Species 1	T. sarekensis (1.1)	O. nova (2.9)	Z. acutirostris (5.9)
Tectocepheus velatus sarekensis	–	0.0527	0.0004
Oppiella nova		–	0.0777
Zetomimus acutirostris			–

¹ Numbers in parenthesis next to the species correspond to the mean of the sum of individuals across all replicates in the sampled volume (1272 cm⁻³ sampled at 5 cm depth).

). The presence of Z. acutirostris was higher than T. sarekensis, but they did not differ significantly from O. nova.

4. Discussion

In forage crops, the absence of differences in oribatid abundance, H’ and d between the five former fertilization treatments (Table A1) aligns with results from [48] for a wheat-barley rotation using cattle manure. Additionally, the former fertilization treatments allowed similar yields at the end of the previous fertilization period [47]; yields were only slightly lower in the 170 CM treatment due to N availability constraints linked to a biennial crop rotation. It can be assumed that similar yields indicate a similar input of plant materials that can feed oribatids [8,9,11]. Thus, it could also be assumed that the soil quality was adequate, which may explain the absence of differences in the studied oribatid indicators. Furthermore, the predominance of Z. acutirostris over T. sarekensis in forage crops (Table A2, Table A3, Table A4 and Table A5) serves as another indicator of acceptable soil quality, as Z. acutirostris has been characterized as sensitive to environmental impact [49], whereas T. sarekensis is commonly found when disturbances are present, i.e., contaminated soils [50].

Regardless of the type of land use, the maximum oribatid abundance in various samplings (Figure 4a) was lower than the average values of 23,903 individuals m⁻² found under similar Mediterranean climate conditions in a cereal cropping system [48] and much lower than numbers from areas under a continental Mediterranean climate at higher altitudes (>860 m a.s.l.) [51]. During the studied period, the maximum average abundance (9287 individuals m⁻²) was recorded in the forest land use. However, for all land uses, oribatid presence significantly diminished in June compared to the April sampling by c. 5000 in m⁻² (80% reduction) (Figure 4a) or even when compared with September sampling. Mesofauna is sensitive to changes in climatic factors [52]. Among those factors, soil moisture is considered to be one of the most decisive influences affecting oribatid communities [11]. However, in this study, the minimum soil moisture of 140 g kg⁻¹ in June (Figure 3), close to water-holding capacity values for a sandy loam texture (around 150 g kg⁻¹), was not a constraint. Temperature was not a limiting factor either, as negative effects of temperature on oribatid mites have only been found above 35 °C [11], and mean temperatures in June are around 19 °C (Figure 1). However, it has been noted that seasonal environmental oscillations can control the life cycles of many oribatid species, outweighing the role of other intrinsic factors [53,54]. Mediterranean climates are characterized by predictable year-to-year fluctuations; in this case, June is associated with peak ET_o demand (123.5 mm, Figure 3), which typically exceeds average rainfall (84.9 mm, Figure 1). In fact, rainfall amounts in the sampling period were 103.0, 71.8, and 137.5 mm for April, June, and September, respectively. June was the driest month when rainfall values were compared with ET_o figures (Figure 1 and Figure 3) despite the humid Mediterranean climate. Additionally, and without considering that life cycles between oribatid species present some asymmetries as iteroparous, species with annual life cycles exist, and most of the species showed the highest abundances in winter [54,55]. Some species may hatch in late spring, with juvenile life forms the most abundant in summer [54], and as explained previously, juvenile forms were not classified in this study.

In the studied period, maximum average diversity (2.19) and minimum average dominance (0.29, Figure 4) were found in the forest. These results could be due to a wider range of microhabitats for oribatids to inhabit in this ecosystem [56,57]. Higher diversity of organic matter linked to litter heterogeneity increases oribatid diversity, as noted [58]. For instance, in Japanese coniferous forests, it was shown that the abundance and species richness of Oribatida was greater in a mixed litter (litter of several tree species) than in litter consisting of only one tree species [59] because it produces microhabitat heterogeneity [58]. In contrast, pasture and forage crops are more simplified ecosystems. The average values of dominance (Figure 4c) in forage crops (0.56) and in the forest (0.29) are close to the values found rainfed cereal crops under no tillage (0.52) and in the forest (0.26) by [48]. However, H’ values (Figure 4b) were higher than [48]. In fact, the diversity values found were close to Arroyo and Iturrondobeitia [51], who found H’ values from 2.4 in the forest and from 0 to 2.4 in crops, although their values for pastures (from 3.0) were higher than our records, and probably linked to their potential number of different plant species present.

Under different land uses and from the seven most representative species extracted from the PCA analysis and included in the cluster (Table 1, Figure 4 and Figure 5), we can assert that H. minutissima and X. tegeocranus were only present in the forest system (Table A3, Table A4 and Table A5). In fact, X. tegeocranus has been described in forest habitats where oak (Quercus robur) was present or dominant [60]. Moreover, H. minutissima has been described in several wild habitats of Tenerife island (Spain), especially in mosses, lichens, and laurisilva [61], which is a humid and warm environment close to that studied here.

The three most prevalent species in agricultural land uses (T. sarekensis, O. nova, and Z. acutirostris) were used to understand patterns between pasture and forage crops. Tectocepheidae have been classified as the most insensitive group to environmental destruction [49] or strong disturbances [62]. T. sarekensis is an opportunistic herbofungivorous species. It is well known for having a relatively rapid metabolism and a thelytokous parthenogenetic reproduction, a life strategy that enables organisms to occupy a wide range of ecological niches, including highly disturbed habitats. In contrast, most species of Oribatida have slower metabolism and development and have much lower dispersal capacity [6,8], which makes it difficult for them to adapt to new conditions, to easily escape from the stress of disturbance, or quickly recolonize disturbed habitats [50]. The species T. sarekensis is present in all the studied land uses, but it was the predominant species in pasture (Table A3, Table A4 and Table A5). This result disagrees with [63], who observed that the diversity of oribatid communities decreased from natural forests to pastures and finally to cultivated lands.

In our conditions, the predominance of T. sarekensis in pasture can raise concerns about soil vulnerability under this land use. Soil bulk density has been previously considered a reliable estimator of the soil’s physical condition and porosity [64]: soil pores are available places for oribatids to live in [65], and Oribatida partially distribute themselves in soil layers according to pore diameter [66]. Macroporosity is highly reduced by compaction [67], and this reduction might affect mite dispersal and possibly sexual contact. Indeed, a negative correlation between oribatid mite abundance and soil bulk density has been described [68]. In this case, the predominance of T. sarekensis in pasture can be related to the highest bulk density among the studied soils. This was 1593 kg m⁻³, indicating higher compaction than in the forage crop system, with a bulk density of 1325 kg m⁻³, although both bulk density values were below the maximum accepted bulk density (1600 kg m⁻³) for sandy loam soils [69]. However, the increase in bulk density in pasture (compared with forage crops) can also explain the absence of Z. acutirostris in this land use. In fact, Z. acutirostris is an indicator of stressful human intervention [50] related to soil requirements for Oribatida. The sexual reproduction of Z. acutirostris makes this species less competitive against compaction disturbance than O. nova and T. sarekensis with a parthenogenetic strategy. However, it should be noted that the presence of the fungivore grazer O. nova in pasture should not be affected, as it can survive in compacted soils [70]. However, O. nova disappears in June samplings (Table A4). This disappearance could be related to the sensitivity of this species to drought [71], as in May and June, ET_o was higher than rainfall (Figure 1 and Figure 2). A slight reduction in soil moisture in the upper soil layer under pasture might control the presence of O. nova. Pasture has a high evaporative demand throughout the year because of its permanent soil vegetative cover, as can be observed in the highest soil water depletion (lowest soil moisture content) in September (Figure 3), even though the rainfall in the sampling period allowed four harvests. In the studied area, oribatid species raised concerns about undesirable compaction increase in the pasture, which should be considered in its management in order to maintain, and even enhance, the general advantages of pastures in terms of organic matter increase and related benefits [72].

5. Conclusions

Oribatid adult abundance, regardless of land use, was the lowest in summer (June), with no differences between spring (April) and autumn (September). Agricultural uses (pasture and forage crops) did not differ in oribatid diversity and dominance, which were lower and higher, respectively, than those in the forest. In spring samplings and in forage crops, the number of individuals of the species Z. acutirostris (sensitive to disturbances and which reproduces sexually) predominated compared to T. sarekensis, and both did not differ from O. nova numbers. In the pasture, T. sarekensis (with a reproductive parthenogenetic strategy) was the most abundant species, indicating some management constraints (probably linked to the highest bulk density) but which deserves further research. H. minutissima and X. tegeocranus were only present in the forest, which also exhibited the largest diversity of species. Among the studied parameters, the presence of specific oribatid species showed high potential as soil quality indicators.