Bromeliads (Bromeliaceae) are abundant in neotropical rainforests. Some estimates reach densities above 100 thousand individuals/hectare (Sugden and Robins, 1979; Goffredi et al., 2011). In many bromeliads, the arrangement of the leaves creates a tank or phytotelma (pl. phytotelmata), i.e., a reservoir of fresh water and organic debris. Bromeliad phytotelmata play a central role for many fauna species by providing shelter, breeding grounds, food and water (Benzing, 2000; Kitching, 2000). In return, the microorganisms and metazoans associated with the phytotelma provide nutrients to the plant (Leroy et al., 2015). Phytotelmata are also found in tree and bamboo holes and pitcher plants.
Nematodes are increasingly recognized as important participants in the ecology of aquatic ecosystems. Marine nematodes connect primary producers, decomposers, and macroscopic consumers; and in lentic and lotic ecosystems nematodes respond mainly to the input of organic matter, and secondarily to the microorganisms on which they feed (Majdi and Traunspurger, 2015; Majdi et al., 2017). Thus far little is known about nematodes inhabiting phytotelmata, despite the importance of these aquatic ecosystems in neotropical and temperate forests (Kitching, 2001).
Most studies on phytotelma nematodes are taxonomic: around 44 genera and five new species have been reported (Menzel, 1922; Meyl, 1957; Goss et al., 1964; Krügel, 1993; Benick, 1924, Varga, 1928, and Thienemann, 1934 cited by Kitching, 2000; Zullini et al., 2002; Bert et al., 2003; Holovachov et al., 2004; Zullini, 1977, and Jacobs, 1984 cited by Hodda et al., 2006; Quisado, 2013; Kolicka et al., 2016). The ecology of phytotelma nematodes was first examined by Devetter (2004), who found them to be abundant in tree holes in temperate forests. In these phytotelmata, nematode abundance did not vary significantly across seasons and sampling locations.
More detailed studies were conducted in plastic cups mimicking tree holes, also in temperate forests (Ptatscheck and Traunspurger, 2014, 2015; Ptatscheck et al., 2015). In these studies, nematode abundance varied greatly, with mean values as low as 3 and as high as 5,280 individuals/phytotelma. The nematode trophic structure was dominated by bacterial and hyphal feeders, with a greater proportion of the latter in comparison with previous studies in marine, lentic, and lotic ecosystems. The amount of organic matter impounded in the phytotelma, the biomass of algae living in the water, and the average daily air temperature were the only environmental factors that affected nematode abundance and diversity.
The first study in a neotropical region focused on the tank bromeliads
Seasonal variations in nematode abundance were also found in bromeliads living in a rainforest in Panama (Zotz and Traunspurger, 2016). Total abundance and the abundance of epistrate feeders, omnivorous, and predators were higher in the rainy season, while deposit (mainly bacterial) feeders and suction feeders (on plants and fungi) were more abundant in the dry season. The amount of organic matter impounded in the phytotelma correlated positively with total abundance.
Collectively, these studies suggest that bacterial and hyphal feeder nematodes predominate in phytotelmata; and that nematode abundance (total and per trophic group) varies seasonally. What determines this seasonality is less clear. Rainfall and air temperature are major seasonal variables, but only the latter was found to be relevant for nematode abundance (Ptatscheck and Traunspurger, 2015). The few PCPs of the water so far assessed – volume, pH, oxygen content, and electrical conductivity – did not impact the nematodes, while organic matter input in the phytotelma did (Ptatscheck and Traunspurger, 2014; Zotz and Traunspurger, 2016).
To further understand the ecology of phytotelma nematodes, we decided: (i) to focus on ecosystems distinct from tropical and temperate forests, and (ii) to record several environmental variables over a long time and assess their impact on the nematofauna.
To test our hypothesis, we conducted preliminary assessments to define the best nematode sampling method and to select the bromeliad species and phenological stage to be sampled. We then: (i) described the nematode trophic structure in the bromeliad/phenological stage chosen; and (ii) evaluated the effect of rainfall, air temperature, amount of organic debris fallen into the phytotelma, and PCPs of the water – volume; temperature; pH; dissolved organic carbon, nitrogen, oxygen, and solids; and electrical conductivity – on the nematode trophic structure.
Samples were collected in a preserved area of the Restinga de Jurubatiba National Park (RJNP) (coordinates 22°11′07.0″S; 41°25′45.4″W). The sampling area is roughly rectangular, about 600 × 300 m (18 hectares). This region’s climate is AW (tropical savanna) according to the Köppen classification, with mean monthly temperature ranging from 20°C in July to 26°C in February. Rainfall is seasonal, with monthly mean of about 40 mm from June through August, and 130 to 175 mm from November through February.
Sampling methods were compared for two bromeliad species with wide distribution in the RJNP:
Samplings were carried out on a single day, in February and September 2013 (summer and winter, respectively). On each sampling date, for both bromeliad species, eight plants with inflorescence and eight plants without it were randomly chosen along a random 500 m path.
Since each bromeliad has an unknown abundance of nematodes, methods M1 through M5 were assessed successively in the same plants, and their relative efficiencies to recover additional nematodes were compared.
Method M1 consisted of suction of the phytotelma water with an automatic pipette connected to a thin rubber hose. The water collected was passed through precision sieves (60 and 500 mesh), and the resulting suspension was placed in a plastic flask. For M2, the same phytotelma was washed with 200 ml of tap water applied under pressure by a back-mounted sprayer, to suspend the organic matter and nematodes retained in the small space between leaf axils. The water was collected by pipetting and placed in a flask. This procedure was repeated twice, and the three volumes obtained were pooled and passed through precision sieves. The resulting suspension was stored in a flask. Method M3 consisted of the same three cycles of water jetting-pipetting, pooling of suspensions, and sieving employed for M2. Method M4 consisted of three additional cycles. Method M5 involved removing the same bromeliad from the soil and taking it to the laboratory in a plastic bag. The plant was entirely defoliated and the material was washed in tap water, collecting all the water in a five liter bucket. The suspension was passed through sieves and stored in a flask.
The 160 samples were individually submitted to extraction of nematodes by the method of Coolen and D’Herde (1972), with modification – without previous grinding in a blender – with centrifugation at 760.24 G for 3 min and 190.06 G for 2 min, and sieving as described before. The resulting nematode suspensions were entirely counted on Peter’s slides under an inverted microscope. The total abundance of nematodes and the abundance per trophic group were computed. Nematodes were assigned as plant, hyphal or bacterial feeders, unicellular eukaryote feeders or carnivores (Moens et al., 2006) by examination in a Nikon Eclipse® microscope with Nomarski interference contrast.
This study was conducted in mature
Samplings were carried out every three months from June 2014 through March 2016. On each sampling date, the tank water from eight bromeliads was suctioned and subjected to the following measurements: volume (in ml); temperature (in °C); pH; dissolved oxygen (DO2, in mg.L−1); dissolved solids (DS, in mg.L−1); and electrical conductivity (EC, in mS.cm−1). These were measured with an Icel Manaus® meter (model PH–1500) with appropriate sensors. A 10 ml aliquot was collected in an unused penicillin flask and used for measurement of dissolved organic carbon (DOC, in mg.L−1) and dissolved nitrogen (N, in mg.L−1) using a Shimadzu meter (model TOC–Vcph). Macroscopic organic debris (OD) fallen into the phytotelma was collected in a paper bag, dried for 24 hr in an oven at 80°C and expressed in grams.
Each bromeliad was removed from the soil and placed in a plastic bag, with care to add the water initially sucked from the phytotelma for the PCP measurements. In the laboratory, the M5 sampling method, sample processing, and nematode counting were performed as described before.
A WatchDog® weather station was used from May 2014 through April 2016 to monitor the following parameters: rainfall (incidence and accumulated monthly volume, in mm); mean monthly air temperature (in °C); and mean monthly temperature (in °C) of the water retained in a specimen of
To assess the efficacy of the sampling methods M1 through M5 in
To identify the phenological stage at which nematodes were more abundant, only the nematode counts obtained using the best sampling method (M5, see Results) were considered. Data from the two sampling dates were pooled and tested for homogeneity of variance and normality of errors. Since the data did not satisfy the requirements, they were transformed to √
To determine how nematode trophic structure relates to rainfall, air temperature and PCPs of the water, regression and redundancy analyses (RDA) of nematode counts, climate and PCP variables were conducted using the R language v.3.6.1. (R Core Team, 2019). The RDA was performed with the R package vegan. All the raw data are available at
Among the five sampling methods assessed, M5 stood out with higher efficiency to recover nematodes from the phytotelma of
In the assessment to select the bromeliad species and phenological stage for the long-term sampling,
Nematode abundance (total and per trophic groups) recovered from the phytotelma of mature
Abundance | |||||
---|---|---|---|---|---|
Species/phenological stage | Total | Bacterial feeders | Hyphal feeders | Carnivores | |
|
194.6aa | 42.1a | 48.8a | 103.7a | |
|
139.6b | 15.2b | 22.3ab | 102.1a | |
|
40.7c | 1.2b | 9.7b | 29.8b | |
|
49.9c | 3.7b | 10.2ab | 36b |
In
Descriptive statistics of nematode abundance (total and per trophic group) recovered from the phytotelma of
Abundance | ||||
---|---|---|---|---|
Statistics | Total | Bacterial feeders | Hyphal feeders | Carnivores |
Mean (min–max) | 62.8a (1-534) | 30 (1-504) | 13.3 (0-201) | 14.2 (0-87) |
SD | 14 | 15.6 | 47.7 | 11 |
CV% | 22.3 | 52 | 358.6 | 77.5 |
The rainfall was atypical during the 24 months, including a dry spell near the end of the study (Fig. 3). The air temperature and the temperature of the water retained in the phytotelma varied in parallel, except for July and December 2014 and January through March 2015. The regression analysis showed no significant association between nematode abundance (total and per trophic group), air temperature, and rainfall (Table S2). From the results of the permutation test, the RDA also provided no significant association (
Statistical parameters for the correlation analysis between nematode abundance and climate variables in the phytotelma
Statistical parameters | |||
---|---|---|---|
Abundance |
|
|
|
|
|||
Total | 0.39 | 0.87 | 0.13 |
Bacterial feeders | 0.78 | 0.08 | 0.01 |
Hyphal feeders | 0.09 | 4.17 | 0.4 |
Carnivores | 0.82 | 0.06 | 0.01 |
|
|||
Total | 0.97 | 0.001 | 0.0002 |
Bacterial feeders | 0.77 | 0.09 | 0.02 |
Hyphal feeders | 0.85 | 0.04 | 0.01 |
Carnivores | 0.35 | 1.04 | 0.15 |
In the phytotelma of
Descriptive statistics of impounded organic debris and physico-chemical variables of the water in the phytotelma of
Variables | |||||||||
---|---|---|---|---|---|---|---|---|---|
Statistics | ODa(g) | Volume (ml) | DOC (mg.L−1) | N (mg.L−1) | Temp (°C) | pH | DO2 (mg.L−1) | DS (mg.L−1) | EC (mS.cm−1) |
Mean (min–max) | 13.7b (1–51) | 222.5 (37–910) | 97.4 (15.9–574) | 4.2 (0.1–12.8) | 22.1 (20–29) | 5.6 (3.5–8.1) | 5.7 (3.2–8) | 80 (8–247) | 0.1 (0.01–0.4) |
SD | 10.4 | 157.8 | 108.5 | 2.9 | 2.6 | 0.7 | 1 | 56.6 | 0.1 |
CV % | 75.8 | 71 | 111.4 | 68.2 | 11.7 | 12.5 | 17.6 | 70.7 | 72.7 |
In
Hence, it appears that the relative abundance of hyphal feeders is typical of phytotelmata. This contrasts with their low numbers in other aquatic ecosystems, such as rivers and lakes (Beier and Traunspurger, 2003; Majdi and Traunspurger, 2015). In phytotelmata, hyphal feeder nematodes certainly thrive on the large fungal biomass. Fungi are the main decomposers of impounded plant litter, which is the main carbon source in bromeliad phytotelmata (Benzing, 2000; Osono, 2007).
In
To understand the temporal pattern of nematode abundance in a particular phytotelma, one should look at rainfall and air temperature because these are climate variables known to impact nematodes. In this study, rainfall was atypical during the seasons, and included a 4-month drought. Nonetheless, water was always available in the phytotelma, provided by rain and/or moisture carried by the onshore sea breeze that condensed at night as dew. The air temperature fluctuated little over the months, and it did so in a range suitable for nematode development. Hence, rainfall and air temperature did not impact nematode abundance (total and per trophic group).
The minor roles of rainfall and air temperature in determining the abundance of nematodes in
In phytotelmata, the amount of impounded organic debris is a determining factor on the fauna present, as well as on the nutrient and energy cycles (Kitching, 2001). In this study, nematode abundance (total and per trophic group) correlated positively with the amount of organic debris fallen into the phytotelma. This agrees with previous reports of the impact of organic input on nematode abundance in bromeliads (Zotz and Traunspurger, 2016), and on nematode abundance and diversity in plastic cups (Ptatscheck and Traunspurger, 2014). The primary production by algae inhabiting the phytotelma also affects positively nematode abundance, as reported by Ptatscheck and Traunspurger (2015).
Since phytotelmata are aquatic ecosystems, it seems plausible that the PCPs of the water impact their biota. Indeed, communities of algae, archaea, bacteria, micro- and macroinvertebrates have been shown to respond to the water PCPs in different phytotelmata (Goffredi et al., 2011; Marino et al., 2011, 2013; Carrias et al., 2012, 2014; Gossner et al., 2016; Louca et al., 2016, 2017). Interestingly, this seems not to be the case for nematodes.
In the present study, among the eight water parameters monitored, the only significant correlation was between the abundance of hyphal feeders and DO2. Even in a more controlled setting – plastic cups mimicking tree holes – water volume, DO2, pH and EC had no impact on nematodes (Ptatscheck and Traunspurger, 2014, 2015). Nematodes have an assortment of physiological adaptations – such as the selective permeability of the cuticle; the ability to exchange water and ions with the environment to avoid osmotic stress; and flexible energy metabolism – that enable them to face severe environmental conditions (Perry and Wright, 1998; Eyualem-Abebe et al., 2006). These adaptations certainly contribute to their survival in phytotelmata, apparently without major impacts from the water PCPs.
In this study we found additional patterns of nematode abundance that require further studies to explain.
Furthermore,
This study also brings valuable information on sampling methods for studying the micro- and meiofauna in bromeliad phytotelma. Pipetting the phytotelma water has been the method of choice in many taxonomic and ecological studies. Method M5 (collecting the entire plant and defoliating it in the laboratory) was more efficient than several cycles of pipetting-water jetting (M1-M4), both for
Our study contributes to understanding of the ecology of phytotelma nematodes by confirming some trends and raising new questions. Bacterial and hyphal feeders predominate over other trophic groups. Generally, bacterial feeders are more abundant, but hyphal feeders may outnumber them at times. Factors that likely impact the rate between bacterial and hyphal feeders include the amount and type of plant debris fallen into the phytotelma; the stochastic death and decomposition of visiting macrofauna; and the occasional or seasonal release of plant substances in the phytotelma water, such as resins in tree holes, nectar from inflorescences that remain immersed in the water, and prey-digesting secretions in pitcher plants.
Apparently, nematode abundance (total and per trophic group) may or may not fluctuate seasonally. Seasonal variations that have been reported in bromeliads and plastic cups cannot be easily explained yet, since major climate variables such as air temperature and rainfall have minor effects on phytotelma nematodes, if any.
The amount of organic matter impounded in the phytotelma is a key factor acting on the nematodes. Intriguingly, the PCPs of the water retained in the phytotelma are not. This underscores the adaptability of nematodes to most aquatic ecosystems, often as the most abundant and/or diverse metazoan group.