There are several lakes and reservoirs in Mexico which cover an area of 2100 km2 and 4810 km2 respectively. The central and eastern part of the country has more than 225 reservoirs (Alcocer et al., 2010). Of these, the reservoir Valle de Bravo (State of Mexico) is particularly important since it provides 38% of the water to the Cutzamala hydraulic system, which in turn, provides water to Mexico City and the City of Toluca. However, as in many other parts of the world (Jeppesen et al., 2007) this reservoir too suffers from an ongoing process of eutrophication principally due to high inflows of nutrients, mainly nitrogen and phosphorus from anthropogenic sources. As a result, there are frequent, sometimes toxic, cyanobacterial blooms, particularly in the summer months, with the common genera being Microcystis sp., Oscillatoria sp., Anabaena sp., Cylindrospermopsis raciborskii and Nostoc sp. (Berry and Lind, 2010; Gaytan-Herrera et al., 2011). High densities of cyanobacteria reduce the potability of water through enhanced levels of cyanotoxins above the permissible limits of 1 µg L–1 of microcystin-LR (Carmichael, 2001; Codd et al., 2005; Alillo-Sánchez et al., 2014). These blooms are also known to have an adverse impact on the aquatic fauna, particularly zooplankton and fish (Kinnear, 2010; Kosten et al., 2012).
The zooplankton community structure is influenced by several biotic and abiotic factors among which food type and density and predation are particularly important (Whitton and Potts, 2000; Cottenie et al., 2001). Several field studies demonstrate that zooplankton densities decrease considerably during cyanobacterial blooms (Burns, 1968; Iglesias et al., 2007). Predation, particularly by vertebrates, is also an important force in structuring zooplankton communities, often in favour of a dominance of small-sized taxa (<1000 µm) (Vanni, 1987; Kagami et al., 2002). The corollary is the presence of high densities of small (<1000 µm) (rotifers of the genus Polyarthra) and/or evasive zooplankton such as copepods (Ghan et al., 1998). On the other hand, the success of biomanipulation efforts to control cyanobacterial blooms often lies in the ability to control the input of nutrients and to regulate zooplanktivory, especially by fish in order to allow high densities of large sized (>1000 µm) generalist feeding cladocerans to flourish (Scholten et al., 2005).
The dominant zooplankton groups, in terms of numbers, in most reservoirs including the Valle de Bravo are rotifers, cladocerans and copepods (Contreras et al., 2009) The biovolume and biomass of zooplankton are better indicators of the availability of zooplankton as grazers or as food for fish (Bottrell et al., 1976) but these are rarely reported for many reservoirs. Rotifers and copepods are more selective in their feeding habits than cladocerans (Kirk and Gilbert, 1992). There is a direct relation between the feeding rates and body size of the cladocerans (Brett et al., 1994). In spite of the importance of studies on the zooplankton community size structure, there has been very little work done in this regard in Mexican water bodies. Seasonal variations on the plankton of Valle de Bravo began a decade ago (Ramírez-García et al., 2002) with emphasis on the density and diversity of rotifers, cladocerans and copepods (Nandini et al., 2008; Contreras et al., 2009). Although these studies present a formidable data base of information on the zooplankton of the reservoir they do not discuss their results in terms of zooplankton size. It has been well documented that cyanobacterial densities are often lower in water bodies dominated by large (>3.0 mm) generalist grazers (Gulati, 1990). We hypothesize that one of the reasons for persistent blooms in Valle de Bravo is the dominance of small sized zooplankton throughout the year. Here we present information on the size structure of the dominant zooplankton taxa in Valle de Bravo and discuss its importance in biomanipulation efforts in the reservoir.
Valle de Bravo (19°11’N and 100°09’W) is a high altitude (1780 m above sea level) drinking water reservoir with a storage capacity of 418.25 x106 m3 with an average depth of about 20 m (Merino-Ibarra et al., 2008; Gaytan-Herrera et al., 2011). Zooplankton samples were collected monthly from June 2010 to May 2011 at three sampling sites (Fig. 1) along with selected physicochemical variables measured at site or at the laboratory (following APHA, 1994) temperature, pH and conductivity, dissolved oxygen using YSI-55, Secchi disk transparency and chlorophyll a. Nutrient concentrations (phosphates and nitrates) were analyzed using a spectrophotometer (YSI 9100). Zooplankton samples were obtained by filtering 50 L of water from the surface (20-30 cm) using a mesh of 50 µm and were fixed immediately with 10% formalin. Quantitative analysis of the zooplankton was carried out using Sedgwick Rafter chamber. For identification of zooplankton we used specialized keys (Koste, 1978; Korovochinsky and Smirnov, 1998). Copepods were classified as nauplii, copepodite and adults. Microcystin levels were quantified using the ELISA immunoassay procedure (Envirologix, Portland, ME, USA).
The size and biomass of the most abundant taxa, three species of rotifers (Keratella cochlearis, Polyarthra vulgaris and Trichocerca similis), three of cladocerans (Bosmina longirostris, Chydorus cf sphaericus and Daphnia laevis) were determined. For copepods we measured nauplii, copepodites and adult of cyclopoids and calanoids. We isolated 50 individuals at random and measured the maximum length and width of each with a Nikon E600 microscope fitted with calibrated camera lucida. The data were then used to calculate the dry weight following Ruttner-Kolisko (1974), Dumont et al. (1975) and Bottrell et al. (1976).
The physicochemical variables during the study period are shown in the Tab. 1. The temperature ranged between 15 to 25°C. The pH was between 7.0 and 10.2 with peak values in September and least values between November to January. The conductivity ranged from 104 to 186 µsc m–1 and was nearly similar at all the sites. The water depth was significantly different at the selected sites; it ranged between 1 and 30 m with a maximum depth of 28.7 m at site 1 (Tab. 1). The depth at Site 3 ranged from 1.5 and 8.8 m during the study period. The dissolved oxygen (DO) levels ranged between 3.5 to 10.7 mg L–1 with minimum and maximum in November and February, respectively. The transparency ranged between 1.6 (March to May) to 9.5 m (in August) at sites 1 and 2 and between 0.9 to 6.5 m at site 3 where we also observed high levels of sediments. The nitrate concentrations (N-NO3) ranged between 0.26 (February to March) to 3.52 mg L–1 with peak values in the months of July and December. Phosphate concentrations (P-PO4) were higher and ranged between 1 (January to March) to 14 mg L–1 (in October) with an annual average of 3.38 mg L–1 (Tab. 1). The microcystin concentrations ranged between a minimum of 0.03µg L–1in December to a maximum of 0.77 µg L–1 between February and April (Fig. 2) and were highest at site 1 with peak concentrations of 0.69 µg L–1.
The dominant species of phytoplankton from this reservoir were Microcystis, Anabaena, Aphanizomenon, Lyngbya and Coelastrum. Among zooplankton, we recorded 26 species of rotifers, five cladoceran species, cyclopoids and calanoids (Tab. 2). About 35% of the rotifer species belonged to Brachionidae. The most common rotifers were Keratella cochlearis, Polyarthra vulgaris, Trichocerca similis and Anuraeopsis fissa. K. cochlearis was most abundant (532-840 ind L–1) in April but declined to 1-15 ind L–1 June-July and November-December; P. vulgaris was also abundant (500-750 ind L–1) from March to May (Fig. 3), but became scarce from June to August (20 ind L–1). Similar trends were also observed for A. fissa which reached densities up to 450 ind L–1 during April. T. similis with densities of 90 ind. L–1 was abundant throughout the year. Kellicottia bostoniensis was observed only in the winter months from November to January (3 ind L–1). During this study period we recorded four species of brachionids for the first time in this reservoir: Brachionus angularis, B. calicyflorus, B. caudatus B. havanaensis.
The species richness of crustaceans was generally significantly and lower than that of the rotifers. Bosmina longirostris was present throughout the year with maximal densities of 48-100 ind L–1 in April. Chydorus cf sphaericus reached densities of 20-100 ind L–1 and was dominant in Site 3 during May. The largest cladoceran in the reservoir was Daphnia laevis but was present only from January to March with peak densities of 15 ind L–1 in February. The densities of both, adult cyclopoids and calanoids never exceeded 1 ind L–1; these were observed in greatest numbers at site 3. Nauplii on the other hand, reached densities of 150 ind L–1. Copepodites, with densities of 70 ind L–1, were abundant in April (Fig. 3). The species diversity (Shannon-Wiener index) varied from 1.63 to 1.92 during the study period. It was highest (2.47) during August but declined to lowest (0.57) two months later (Fig. 4).
Data on the zooplankton body size and biomass are presented in Fig. 5. Most parts of the year, small-sized zooplankton <200 µm dominated. From February to May, densities higher than 800 ind L–1 were contributed by rotifers and nauplii, while larger sized zooplankton (201-600 µm) was represented by cladocerans as B. longirostris and C. cf spaericus. The zooplankton of size >600 µm was observed from November 2010 to May 2011, where of Daphnia laevis was also present for a short period (from January to April). The contribution of large species (>600 µm) to the total zooplankton biomass was highest (410 µg L–1) in April.
Although Valle de Bravo is an important source of drinking water to Mexico City, it has also occasionally high densities of potentially toxic cyanobacteria (Vasconcelos et al., 2010). Among them are Microcystis and Anabaena which frequently dominate the phytoplankton community (Ramírez-García et al., 2002; Gaytan-Herrera et al., 2011). Snowella septentrionalis and Aphanizomenon yezoense were also present in this reservoir but these are often non-toxic (Boutte et al., 2008). Although we did not quantify phytoplankton in this study, a recent article published on the phytoplankton of Valle de Bravo Reservoir (Gaytan-Herrera et al., 2011) clearly shows that cyanobacteria are the dominant primary producers in the reservoir. The four most common taxa of cyanobacteria (densities from Gaytan-Herrera et al. 2011), were Microcystis (1×102 to 1×105 cells mL–1) spp., Anabaena spp. (1×102 to 1×104 cells mL–1), Aphanizomenon spp. (1 to 1×104 cells mL–1) and Lyngbya sp. (1×102 to 1×104 cells mL–1). The common diatom was Fragilaria sp. (1×102 to 1×104 cells mL–1) while the green-algae was Coelastrum sp. (1 to 999 cells mL–1). The aforementioned taxa have been recorded in concentrations ranging from 1 to 1×106 cells mL–1. Cyanobacterial blooms are a common phenomenon in several water bodies in Mexico primarily due to the prevailing high nutrient and temperature regimes found in these habitats (De la Lanza and García, 2002; Berry et al., 2011; Kosten et al., 2012). One of the major health problems associated with cyanobacterial blooms is the presence of toxic microcystins (Carmichel, 1994; Christoffersen, 1996) Although we found that the concentrations of these toxicants were highest in the months of January and February, they were within the stipulated limits of WHO of <1 mg L–1 (Carmichel, 2001). In our study, in spite of the high levels of nitrates and phosphates recorded at the various sites and the low N:P ratio (<4:1), common in tropical lakes (Talling and Lemoalle, 1998), we found very low densities of cyanobacteria and a dominance of chlorophytes and diatoms. Canfield et al. (1989) imply that N:P ratios cannot always explain autotrophy succession in tropical water bodies.
As in a few other waterbodies, one of the important factors influencing the formation of cyanobacterial blooms in this reservoir appears to be related to the fluctuation in the water level in the reservoir (Zohary and Ostrovsky, 2011). These fluctuations result in the instability of the system leading to low cyanobacterial bloom formation (Moustaka-Gouni and Vardaka, 2006). In spite of higher concentrations of nitrogen and phosphorus, we found low levels of phytoplankton; so much so that the volume that was routinely filtered based on previous studies (Nandini et al., 2008) was inadequate in order to read the Chl a concentrations on a visible spectrophotometer. The other important physicochemical variables such as temperature, pH and dissolved oxygen were in the same range as has been reported in previous works (Ramírez-García et al., 2002; Contreras et al., 2009). The Secchi depth, in this study, was significantly greater than that reported by Nandini et al. (2008), which corroborates our observations on the low availability of phytoplankton. Our study also reports higher densities of cladocerans than previous works (Nandini et al., 2008); in other water bodies too it has been reported that low densities of phytoplankton associated with high flushing rates are associated with high densities of zooplankton (Moustaka-Gouni and Vardaka, 2006).
Valle de Bravo has been characterized as a eutrophic reservoir (Merino-Ibarra et al., 2008; Nandini et al., 2008; Contreras et al., 2009). However, during this period, we found that the reservoir was mesotrophic based on the Carlson index (Sheela et al., 2011). This improvement in water quality is corroborated by the presence of high densities of chlorophytes and diatoms (Scenedesmus sp. and Fragilaria sp.) and the increase in Secchi transparency from a maximum of <3 m (Ramírez-García et al., 2002; Contreras et al., 2009) to more than 9m recorded in this study. Sladecek (1983) showed that the ratio of the number of Brachionus (B) to Trichocerca (T) indicates the trophic status of water bodies. Unlike the past in which we found few Brachionus, in this study we recorded four brachionid rotifers. Hence we were able to derive a B/T ratio of 1.3, characteristic of mesotrophic water bodies (Sladecek, 1983).
This reservoir has rotifer dominance (Ramírez-García et al., 2002) with more microphagous than raptorial (Asplanchna, Synchaeta and Trichocerca) taxa. Peak rotifer densities often reached more than 800 ind L–1 although this was lower than those registered a few years ago (Nandini et al., 2008). Keratella and Polyarthra are commonly dominant in reservoirs (Devetter, 1998); this is also the case in this study. However, unlike previous works (Ramírez-García et al., 2002; Nandini et al., 2008; Contreras et al., 2009), we found four species of the genus Brachionus and a few other rotifer species not reported previously in the Valle de Bravo Reservoir (marked on Tab. 1). This suggests changes have taken place since the first extensive sampling in this reservoir a decade ago, both in terms of the number of species and their densities. Ramírez-García et al. (2002) recorded 245-345 ind L–1 of K. cochlearis whereas in this study we found 500-800 ind L–1 of the same species. Among cladocerans the density of Bosmina longirostris also increased significantly, from 100 ind L–1 in 2002 (Ramírez-García et al., 2002) to >500 ind L–1 in the present study. The State of Mexico has close to 200 species of rotifers (Sarma and Elias-Gutiérrez, 1997; Sarma et al., 2009) of which around 20% have been recorded from this reservoir.
Among the cladocerans, there was no significant change in the species richness, with the species Bosmina longirostis, Chydorus sphaericus and Daphnia laevis being the common taxa as in previous reports. However, the density of the cladocerans ranged between 35 to 286 ind L–1 with a maximum in May and though higher than in previous studies, the contribution of this group to the total biomass was very low. As reported in Obertegger and Manca (2011), we also found greater numbers of microphagous rotifer taxa which are known to be associated with low densities of cladocerans. While the species richness was low in Site 1 (7 species), the species diversity ranged between 0.57 to 2.47 at all the sites with peak values in spring and summer.
The size structure of zooplankton communities is influenced by food availability (Brett et al., 1994; Torres-Orozco and Zanatta, 1998) and predation (Dodson and Frey, 1991; Iglesias et al., 2011). In water bodies containing cyanobacteria rotifers often dominate (Gliwicz and Lampert, 1990). While there are several studies in Mexico on the species richness and diversity of freshwater zooplankton (Enriquez et al., 2009), the present work examined the size structure of zooplankton communities. This information is important while considering the possible use of zooplankton for lake management (Vanni, 1987). In terms of density, we found a dominance of taxa in the less than 200 mm range for most of the year 2010-2011. However, in terms of biomass, dominance due to zooplankton was low almost throughout the year except from February to April when the contribution of large species to the total zooplankton biomass was highest. Gulati (1990) mentions that a minimum of 0.2-0.5 mg C L–1 of zooplankton is needed to achieve a significant daily clearance of phytoplankton in meso- to eutrophic lakes. In our study, using a factor of 0.48 to convert biomass to carbon (Andersen and Hessen, 1991), the zooplankton biomass ranged from 0.048 to 0.506 mg C L–1. Except for the months of April and May, the total zooplankton biomass was less than 0.25 mg C L–1. Thus, in Valle de Bravo, high biomass of zooplankton is found only in the spring; for the greater part of the year zooplankton alone is not enough to control phytoplankton blooms. Chirostoma spp., Lepomis macrochirus and Micropterus salmoides are among the dominant fish species in Valle de Bravo (Renteria et al., 2006). A recent study (Gallardo-Torres et al., 2013) indicates that these taxa have a significantly higher preference for microcrustaceans than rotifers exerting a high predation pressure on crustacean zooplankton in the reservoir.
Comparing our findings to previous studies, we observe that Valle de Bravo has begun to show signs of recovery. The phytoplankton densities were lower compared to the past and the communities were dominated by diatoms and chlorophytes instead of cyanobacteria. However, the high fish predation pressure still appears to control the size structure and biomass of zooplankton in this reservoir; Gallardo-Torres et al. (2013) show that cladocerans, copepods, amphipods and insect larvae are the most preferred prey for three of the previously mentioned fish species in this reservoir. We suggest that the several studies on the diversity of zooplankton in Mexico include data on the size structure and biomass in order to reach generalizations on the potential use of zooplankton in lake management.