Free-living nematode assemblages associated with maize residues and their ecological significance

This study was carried out in two separate trials (March–July 2018 and October–February 2019) at Nyangati, Mwea Sub-county, Kenya (Longitude 0°36'45.5“S and Latitude 37°21'18.0“E). The region is classified in agroecosystem zone 3 at 1,202 m in lowland altitude. The soil type in the area is characterized as vertisols (Jaetzold et al., 2009), and receives two rainy seasons per year. The long rains (from March to May) average about 900 mm, whereas short rains (from October to November) average around 800 mm. The minimum temperature and average rainfall recorded in the study area are provided in Table 1, while soil physicochemical properties are given in Table 2.

This work comprised of two experimental groups, consisting of decomposition plots (incorporated with maize residues at the rate of 5 tons/hectare, which is the rate commonly used by many Kenyan farmers) and control plots, where no maize residues were incorporated. The field plots were arranged in a randomized complete block design (RCBD), with each experimental group consisting of four replicates. Each plot measured 36 m² (6 × 6 m) and was separated by a 1 m buffer zone between plots. In addition, each plot had six rows per plot. The maize residues were first chopped into small pieces (approximately 10 cm) to enhance the decomposition process and subsequently incorporated into the soil to a depth of 30 cm.

Soil samples were taken monthly for a period of 4 months after the incorporation of maize residue for two seasons (March–July 2018 and October–February 2019). In each plot, five soil cores (100 g each) were randomly collected from the inner rows at 0–20 cm depth, using a soil auger measuring 3.5 cm in diameter through a cross diagonal sampling pattern. During each sampling date, the five soil sub-samples were mixed gently and composited into a 500 g homogenous sample (Coyne et al., 2014).

Each soil sample (250 g derived from a 500 g composite sample) was placed in filter trays and nematodes were extracted and recovered for 48 h using a modified Baermann technique (Hooper et al., 2005). After extraction, nematodes were enumerated by counting and identified to genus level under a compound microscope using diagnostic keys. Afterwards, the FLN nematode genera were classified into four functional guilds, including bacterivores, predators, fungivores, and omnivores (Yeates et al., 1993). They were also assigned into their respective colonizer-persister group, from cp-1 to cp-5, where index values represent life course strategies that are attributed with either r- or K- characteristics (Ferris et al., 2001). Those nematodes assigned to cp-1 are referred to as colonizers that are usually characterized by high fecundity and small eggs. They also require enriched nutrients for favorable growth, have large population changes, and short generation times. Those assigned to cp-5 are known as persisters, which are characterized by few offspring, low fecundity, large body size, and great sensitivity to disturbances (Bongers and Bongers, 1998).

Data on average rainfall and minimum temperature (Table 1) during the field trials were obtained from Kenya Meteorological Department, Nairobi, Kenya. The soil physicochemical properties (calcium, clay, copper, iron, magnesium, manganese, phosphorus, potassium, sand, silt, sodium, pH, nitrogen, total organic carbon, and zinc) were analyzed at Kenya Agricultural and Livestock Research Organization, Nairobi, Kenya (Table 2).

Prior to the analysis, log transformation (Log (x + 1)) of nematode count data was done where necessary to meet the normality criteria for statistical data analyses. The influence of maize residues on nematode abundance, ecological, functional, and metabolic footprint indicators was assessed using ANOVA with the R package vegan (Oksanen et al., 2016). The treatment means separation was implemented with Tukey HSD test using the R package agricolae (De Mendiburu, 2015). We evaluated the ecological significance of FLN using ecological, functional, and metabolic footprint indicators. In particular, the ecological indices (Maturity indices; MI and MI2-5) were computed to infer soil food web ecosystem perturbations (Ferris and Bongers, 2009). Then, the functional soil food web indices, including channel index (CI), structure index (SI), basal index (BI), and enrichment index (EI) were evaluated to infer the status and complexity of the soil food web conditions as indicated by nematode assemblage composition (Ferris et al., 2001). Nematode metabolic footprints were explored to indicate various ecosystem functions, services, and structure by partitioning the amount of carbon utilized by nematodes in metabolic activity (respiration) and growth and egg production (biomass) (Ferris, 2010). All ecological, functional, and metabolic footprint indicators were computed using the online program, Nematode Indicator Joint Analysis (Sieriebriennikov et al., 2014). All statistical analysis was performed using R software, version 3.6.2 (R Core Development Team, 2019).

Thirty free-living nematode genera assigned to four trophic levels were identified. The most predominant trophic guild in all decomposition plots in both seasons was bacterivores, followed by fungivores. We observed significant treatment effects (P < 0.05) on a total of 8 nematode genera (first season) and 13 genera (second season) (Tables 3 and  4, respectively). In the first season, there were 15 bacterivores, 4 fungivores, 3 omnivores, and 3 predators. Among the different taxa, Acrobeles and Plectus (bacterivores), Aporcelaimellus (omnivore), and Nygolaimus and Seinura (predators) were the most abundant genera in the decomposition plots (Table 3). On the other hand, in the second season, there were 15, 3, 6, and 3 bacterivores, fungivores, omnivores, and predators, respectively. Among the various taxa, Cephalobus, Eucephalobus, Acrobeles, Heterocephalobus, Aphelenchus,  and Aphelenchoides were significantly higher in decomposition treatments, while Mesodorylaimus (omnivore), Nygolaimus and Mononchus (predators) were more prevalent in the control plots (Table 4). Most of the bacterivores were classified as cp-1 (Ba1) and cp-2 (Ba2) except Alaimus and Prismatolaimus, while all fungivores were assigned to cp-2 (Fu2). Both omnivores and predators were categorized as cp-4 and cp-5 except for Seinura. We further identified that the abundance of cp-1 bacterivores was lower in the decomposition plots than in the control plots, but this was not the case for generalist opportunist (cp-2) bacterivores and fungivores.

There were no differences in ecological indicators (MI, MI2-5) and functional indicators (CI, BI, EI, and SI) in season one at P < 0.05 (Table 5). However, during the second season, all the values of MI, MI2-5, CI, BI, EI, and SI were significantly affected at P < 0.001 (Table 5). In particular, maturity indices (MI and MI2-5), CI, and SI were reduced in the decomposition plots. Conversely, BI and EI were markedly increased at P < 0.001 in the decomposition plots relative to the control. Moreover, the metabolic footprints (composite, structure, predator, fungivore, enrichment, and bacterivore) were statistically different (P < 0.05) during the second season (Table 6). A combination of fungivore, enrichment, and bacterivore footprints, an indicator for mineralization services in the soil food web, was significantly larger in the decomposition treatments than in the control plots in the same season. We then observed that structure and predator footprints developed a lower activity in the decomposition plots in the second season, while the composite footprint was progressively higher in the decomposition plots compared with the control in both seasons (Table 6). Notably, in season one, only predator footprint was statistically different between decomposition and control plots (P < 0.001) (Table 6). Finally, plotting nematode functional metabolic footprints using EI and SI revealed that both control and decomposition plots were characterized as structured (quadrat “C”) in season one (Figure 1). During the second season, the control remained structured (quadrat “C”), while the decomposition plots shifted to quadrat “D” categorized as degraded (Figure 2).

Figure 1:

Figure 2: