The semiarid region of Brazil experiences extreme seasonal variations in rainfall. Precipitation is concentrated in a few months of the year and is followed by a long dry season with significant inter-annual variability. The characteristics of semiarid Northeastern Brazil and high environmental temperatures there create water deficits for at least 70 % of the year. Cycles of both drought and extreme rainfall occur at intervals ranging from a few years to decades. Therefore this environment is vulnerable and its climate is unstable (Marengo, Alves, Beserra, & Lacerda, 2011). These conditions are conducive to high evaporation rates and long reservoir water residence times, and significantly influence lacustrine phytoplankton organization (Bouvy, Falcão, Marinho, Pagano, & Moura, 2000), they favor cyanobacterial blooms in dry periods.

Natural or anthropogenic events that alter the hydrodynamic and limnological characteristics of the environment can adversely affect reservoirs (Straskraba, Tundisi, & Duncan, 1993). Different levels of disturbance affect community organization in various ways (Lopes, Ferragut, & Bicudo, 2009) and may interrupt, postpone, or redirect seasonal phytoplankton successions (Znachor, Zapomělová, Řeháková, Nedoma, & Šimek, 2008). Abrupt changes in species composition can occur that interfere with internally driven self-organization and ecological equilibrium processes (Reynolds, Padisák, & Sommer, 1993). There have been numerous studies on the environmental impact of tropical reservoirs on phytoplankton. Nevertheless, they usually focused on anthropogenic disturbances; few studies examined the impacts associated with natural disturbances (Chellappa, Chellappa, Câmara, Rocha, & Chellappa, 2009a; Câmara, Rocha, Pessoa, Chellappa, & Chellappa, 2015).

Reservoirs in the semiarid region of Brazil are usually very hydrologically stable (Bouvy et al., 2000). This factor is critical for sustaining the long-term dominance of certain phytoplankton species (Huszar, Silva, Marinho, Domingos, & Sant’Anna, 2000). Cyanobacterial species forming perennial blooms have been identified in eutrophic reservoirs in the semiarid region of Brazil (Chellappa, Chellappa, & Chellappa, 2008; Dantas, Moura, & Bittencourt-Oliveira, 2011) and in many other places worldwide (Naselli-Flores, Barone, Chorus, & Kurmayer, 2007; Dejenie et al., 2008; Douma et al., 2010). Factors that contribute to the distribution and stability patterns of phytoplankton in these reservoirs include trophic states (Barone & Naselli-Flores, 1994; Naselli-Flores, 2013), multidimensional environmental gradients (Fabbro & Duivenvoorden, 2000), and survival strategies (Reynolds, 1998).

The present study examined the environmental variables having the greatest influence on the seasonal and spatial dynamics of the phytoplankton communities in a eutrophic reservoir in the semiarid region of Brazil, the Argemiro de Figueiredo reservoir. According to Lins, Barbosa, Minillo and Ceballos (2016), perennial blooms of toxic cyanobacteria are common in this reservoir, particularly during the dry months of the season in the region. Thus, we investigated the roles of natural and anthropogenic disturbances that alter phytoplankton biomass stability. Two questions were raised: (i) Is reservoir overflow a natural disturbance that can disrupt or alter phytoplankton biomass? (ii) Do the nutrients derived from pisciculture in the reservoir cause an anthropogenic disturbance that increases the biomass of cyanobacterial phytoplankton functional groups? To answer these questions, we analyzed the reservoir’s phytoplankton population using the functional group. This approach has been widely employed in ecological studies and is an effective tool for explaining community structures and their responses to alterations in environmental conditions (Kruk, Mazzeo, Lacerot, & Reynolds, 2002; Kruk et al, 2011; Brasil & Huszar, 2011; Reynolds, 2014; Török et al, 2016).

MATERIALS AND METHODS

Description of the study area: The present study was undertaken at the Argemiro de Figueiredo reservoir (7° 36’ 51” S - 35 40’ 31” W) in the median portion of the Paraiba River basin, which is the largest in Paraiba State, Brazil. The reservoir has a surface area of 1 725 ha, a potential volume of 2.53 × 108 m3, a maximum depth of 39 m, and a hydraulic residence time of 146 days. The regional climate is hot semiarid (type BSh) with a high evaporation rate (Silva, Sousa, Kayano, & Araujo, 2008). The average annual rainfall varies from 600 - 1 100 mm with very irregular monthly and annual regimes. There are marked rainy and dry seasons (Governo do Estado da Paraiba, 2007). The reservoir was constructed in 2001 and is used, among other purposes, for supplying water to approximately 450 000 regional inhabitants and for the production of Nile tilapia (Oreochromis niloticus Linnaeus, 1758) in net cages. The latter activity began in December 2006. Before the net cages were deployed, however, the reservoir water was eutrophic and experienced intense cyanobacterial blooms (> 240 000 ind. mL-1). Microcystis aeruginosa (Kützing) Kützing and Cylindrospermopsis raciborskii predominated, among other potentially toxic species (Barbosa & Mendes, 2005).

Sampling and analysis methods: Samples were collected between August 2007 and July 2009, bimonthly for the first year, and at monthly intervals thereafter. Three sampling points were considered: the confluence of two feeder rivers (PC), near the net cages (PNC), and in the dam zone (PD); while samples were taken in the euphotic (Zeup) and aphotic zones (Zaph). Water transparency was measured using a Secchi disk, and the euphotic zone (Zeup) was calculated as 3.0 times the Secchi disk depth (Cole, 1994). The vertical light attenuation coefficient (Ko) was calculated according to Poole and Atkins (1929). Rainfall and monthly reservoir volume data were obtained from the Executive Agency of Water Control of Paraiba State. Water temperatures, pH, and electrical conductivity were measured in situ using an INCOTERM model 2309 thermometer, a Tecnal digital pH meter, and a Lutron model 4303 conductivity meter, respectively. Alkalinity was determined as described by Mackereth, Heron and Talling (1978).

Samples for the analyses of nutrients, dissolved oxygen, alkalinity, and phytoplankton community composition were taken at two depths using a Van Dorn-type collector (5 liters). The samples used for the identification and quantification of phytoplankton were fixed in formaldehyde (4 %) and Lugol’s solution, respectively. The dissolved oxygen content was determined following the method of Golterman, Clymo and Ohnstad (1978). Soluble reactive phosphorus (SRP) was determined using the ammonium molybdate methodology (APHA, 2005). Dissolved inorganic nitrogen (DIN) was calculated from the sums of the ammonia, nitrite, and nitrate concentrations obtained using the phenol, diazotization of sulfanilamide-NED, and cadmium reduction techniques respectively (APHA, 2005). The DIN:SRP molar ratio was used to evaluate the possibility of phytoplankton growth restrictions due to nitrogen or phosphorus limitations, where DIN:SRP < 13 indicates limiting nitrogen, DIN:SRP > 50 indicates limiting phosphorus, and 13 < DIN:SRP < 50 indicates that neither of these nutrients is limiting (Morris & Lewis, 1988; Kosten et al., 2009). The Carlson index of Tropic States, adapted by Toledo, Talarico, Chinez and Agudo (1983) for tropical regions, was used for trophic characterization. The quantitative analyses of the phytoplankton followed the method of Utermöhl (1958). At least 100 individuals from the most frequently encountered species were enumerated (error < 20 %, p < 0.05) and this number increased during bloom periods, when were counted in each sample, at least 400 individuals of the dominant species (error < 10 %, p < 0.05) (Lund, Kipling, & Le Cren, 1958).

The phytoplankton biovolume (mm³.L-¹) was estimated on the basis of a geometrical formula (Hillebrand, Dürselen, Kirschtel, Pollingher, & Zohary, 1999; Sun & Liu, 2003) using an average of 20-30 individuals and was expressed in fresh weight units, where 1 mm³. L-¹ = 1 mg. L-¹ (Wetzel & Likens, 2000). All phytoplankton species were assembled into functional groups following the criteria established in the study by Reynolds, Huszar, Kruk, Naselli-Flores and Melo (2002) and reviewed by Padisák, Crossetti and Naselli-Flores (2009).

A Canonical Correspondence Analysis (CCA) was performed using Canoco 4.5 forward selection, based on a biotic matrix (functional phytoplankton groups), and an abiotic matrix (Ter Braak & Smilauer, 2002). The significance of the environmental variables (p < 0.05) was determined using the Monte Carlo test with 999 unrestricted permutations. Individual analyses were made to test the temporal and spatial variability of the data and the factors that interfered most with the biotic and abiotic variables. These factors were analyzed using a generalized linear model (the covariance analysis module of Statistica 8.0) which incorporated the following components: collection points, water depth, rainfall, and presence or absence of overflow events.

RESULTS

Regional climatic conditions influenced reservoir volume and hydrodynamics, and, consequently, functional groups replacement and biomass. Two dry seasons with sparse rainfall as well as two rainy seasons were identified during the study period. During some of the rainy months, the accumulated water exceeded the reservoir capacity, resulting in overflow for 59 days during the first event (March-May 2008) and 114 days during the second (May-September 2009) (Fig. 1).

The reservoir water was warm (> 24.0 °C; N = 132), alkaline (> 36.0 mg CaCO3.L-1; N = 132), basic (minimum = 7.0; maximum = 10.0; N = 132), and high in electrical conductivity (> 357.3 µS.cm-1; N = 132) (Table 1). Eutrophic conditions predominated with no significant seasonal variations. Reductions in pH values, electrical conductivity, alkalinity, and dissolved oxygen (DO) were only observed during periods of overflow. Significant spatial differences were noted for pH and DO. The highest values for both were measured near the net cages. pH and DO in the vertical profiles were highest in the euphotic zone, while alkalinity, SRP, and DIN were highest in the aphotic zone (Table 2).

Transparency values for both the dry and rainy seasons indicated that waters were turbid at all collection points and the euphotic zone was reduced (Table 1). A single period of relatively clear water occurred at the second overflow event (Transparency = 1.5 m and Zeup = 4.4 m).

The total phytoplankton biomass varied between 0.01 mm3. L-1 and 28.42 mm3. L-1 (Table 1). The lowest values were recorded during overflow events in the rainy season. The dominance of cyanobacterial functional groups (> 90 % of the total biomass) occurred during the dry season and some rainy months (Fig. 2). The point the net cages (PNC) had the highest total biomass in both the euphotic (6.00 ± 5.3 mm3. L-1) and aphotic (3.69 ± 3.6 mm3. L-1) zones. This portion of the reservoir was also the shallowest (Zmax = 7.7 meters) (Fig. 2).

The thirteen dominant cyanobacterial species were arranged in functional groups S1, K, M, SN, H1, F, MP, P, Lo, and X2, which corresponded to the groups described by Reynolds et al. (2002). Groups S1, SN, and K consisted mainly of Planktothrix agardhii, C. raciborskii, and Aphanocapsa incerta, respectively. They predominated during the dry months in the warmest waters with high turbidity and alkalinity. Group P was represented by Aulacoseira granulata (Ehrenberg) Simonsen and Closterium sp. It predominated during the rainy months when the waters were coldest but there were no overflow events. Functional groups F, M, MP, Lo, and X2 occurred principally during rainy months marked by overflow events. They included the colonial species Botryococcus braunii Kützing, M. aeruginosa, and Microcystis protocystis Crow, the filamentous species Oscillatoria sp., and the unicellular flagellates Peridinium umbonatum Stein and Chlamydomonas sp., respectively. Group H1, with Dolichospermum circinalis (Ralfs ex Bornet & Flahault) as its principal component, was observed during both dry and rainy months (Fig. 2).

Between August 2007 and August 2008, the confluence point of the tributary rivers (PC) and the dam zone (PD) had low total biomass in the euphotic (1.30 ± 2.0 mm3. L-1 for PC and 1.51 ± 2.0 mm3. L-1 for PD) and aphotic (0.53 ± 0.7 mm3. L-1 for PC and 0.29 ± 0.5 mm3. L-1 for PD) zones. The highest biomass values there occurred only in isolated cases in October 2007 (7.1 mm3. L-1 for PC) and February 2008 (6.7 mm3. L-1 for PD). Both of these were dry months. The phytoplankton biomass was highest near the net cages (PNC) until the beginning of the rainy season (10.81 ± 6.4 mm3.L-1 for Zeup; 7.49 ± 4.9 mm3.L-1 for Zaph). It decreased considerably (Fig. 2) during the rainy months and the overflow events (2008).

Filamentous heterocyte and non-heterocyte cyanobacteria in the functional groups S1, SN, H1, and MP dominated and accounted for 57.5 % and 97.0 % of the total biomass, respectively, during the dry period (2007 and 2008) (Fig. 2).

Planktothrix agardhii, the principal representative of group S1, contributed the most to the total biomass. It accounted for 95 % at point PC (August 2007), 59 % at PNC (March 2008), and 83 % at PD (February 2008). The 36 % reduction in P. agardhii biomass in the aphotic zone at PC (October 2007) favored increases in coccoid cyanobacteria biomass in the functional groups M and K. The latter contributed with more than 90 % of the total biomass. In PNC and PD, reductions in the biomass of this species occurred when the reservoir overflowed.

During the overflow period at the start of the 2008 (rainy season), the cyanobacteria contributed less, and these conditions favored the dominance of other functional groups. Group Lo dominated in the euphotic zone (64.7 % of the total biomass) and X2 dominated in the aphotic zone (84.2 % of the total biomass) at collection point PC. Group X2 (65.6 % of the total biomass) and F (99.0 % of the total biomass) dominated at point PD in the euphotic zone. The contributions of cyanobacterial functional groups remained high at collection point PNC. Reductions in cyanobacteria populations (> 40 %) occurred only when the rainy period ended, which the dominance of group P (maximum 0.02 mm3.L-1) mainly in the euphotic zone (Fig. 2).

At the start of the 2008 dry period, as rainfall decreased, phytoplankton biomass increased (4.48 mm3.L-1 for PC; 6.92 mm3. L-1 for PNC; 4.96 mm3. L-1 for PD). Cyanobacterial dominance was frequently observed (> 90 % of the total biomass) and consisted mainly of Groups S1 and K. During the rainy period, phytoplankton biomass did not change and remained high (13.22 ± 13.2 mm3. L-1 for PC; 8.42 ± 3.1 mm3.L-1 for PNC; 9.48 ± 10.6 mm3. L-1 for PD). The beginning of a new rainfall period in February 2009 caused an overflow event that lasted six months. It reduced phytoplankton biomass (0.73 mm3. L-1 for PC; 0.29 mm3. L-1 for PNC; 0.90 mm3. L-1 for PD) until the end of the period. Closterium sp. predominated in Group P and Group M contributed to a biomass increase during this period (4.38 mm3. L-1 for PC; 8.76 mm3. L-1 for PNC; 4.02 mm3. L-1 for PD) (Fig. 2).

Canonical correspondence analysis indicated eigenvalues of 0.299 and 0.128 for the first two axes, respectively (Fig. 3). Pearson’s correlation coefficient of environment-species factors for these axes indicated significant relationships between environmental variables and biomass (Table 3). The Monte Carlo test was significant (p < 0.01), indicating that ordination did not occur randomly. The canonical coefficients and intra-set correlations demonstrated that the variables related to overflows and reservoir volumes were the most important for axis I ordination (Table 3). These variables contributed to the grouping on the positive side of this axis for overflow months and the negative side during months without overflow. The functional groups F, M, and X2 appeared to be closely correlated to rainy months with overflow. The opposite was true for groups S1, SN, and K, which were associated with the dry months (Fig. 3).

The most important variables on axis 2 were the coefficients of light attenuation, rainfall, and water temperature. On the negative side of the axis were the rainy and dry months with the lowest water temperatures, pH, alkalinity, electrical conductivity, and functional groups P and F. On the positive side were the months with the warmest and most alkaline waters, the greatest coefficients of light extinction, and the functional groups Lo and X2 (Fig. 3).

DISCUSSION

The phytoplankton community in the Argemiro de Figueiredo (Acauã) Reservoir consisted of functional groups found in eutrophic, mixed aquatic environments. Algal biomass increased and was influenced by seasonal changes. The largest biomass was recorded during dry periods. Phytoplankton community structure changed and its biomass declined (> 50 %) during the rainy periods mainly when the reservoir overflowed. Intense rainfall causing hydrological instability and overflow can drastically reduce the biomass of cyanobacterial species (Oliveira et al., 2015).

Increasing water flow in the reservoir disturbed the phytoplankton. It altered light availability and reduced pH, electrical conductivity, dissolved oxygen, alkalinity, and biomass. In tropical regions where rainfall is concentrated into just a few months of every year, reservoir volumes can change drastically, disturbing the water column and aquatic communities. These disturbances have a large influence on phytoplankton composition and biomass (Figueredo & Giani, 2005; Câmara et al., 2015).

The long residence time of reservoir water, high evaporation rates, and synergy between high water temperatures and high nutrient concentrations accelerated eutrophication. These factors were also responsible for the dominance of cyanobacteria, especially during the dry season when the functional groups S1, SN, H1, M, K and MP predominated. Planktothrix agardhii, the principal representative of functional group S1, was dominant for most of the dry period. It has frequently been found in eutrophic Brazilian reservoirs (Chellapa, Câmara, & Rocha, 2009, b; Bonilla et al., 2012) and throughout the world (Karadžić, Subakov-Simić, Krizmanić, & Natić, 2010). Its biomass is often very high regardless of the season (Poulíčková, Hašler, & Kitner, 2004).

In the present study, the only time in which P. agardhii did not dominate was during the rainy season, when reservoir water residence time and temperatures were low and overflow caused significant biomass losses. In these conditions, the species belonging to the functional groups F, P, Lo and X2 prevailed. They are found in meso-eutrophic environments (Reynolds et al., 2002; Padisák et al., 2009). Nevertheless, the proliferation of Groups Lo and X2 phytoflagellates in April 2008 resulted from limiting light conditions (transparency < 0.15 m) as well as nutrient abundance.
When they analyzed phytoplankton functional groups in a Turkish reservoir, Gurbuz, Kivrak, Soyupak and Yerli (2003) reported the presence of Group Lo dinoflagellates during periods when light levels were low.

Functional groups Lo and X2 are sensitive to water mixing, so they dominated only in April 2008 when no movement occurred. By May, the water was clear and mixing, so Group F species prevailed since they are common in these conditions (Reynolds et al., 2002; Padisák et al., 2009). Group P, represented mainly by Aulacoseira granulata, predominated when water temperatures fell and light levels increased (Borges, Train, & Rodrigues, 2008; Chellappa et al., 2009a). This group is commonly found in shallow, turbulent, eutrophicated waters (Reynolds et al., 2002; Pádisak et al., 2009). During most of the study period, nitrogen and inorganic phosphorus levels exceeded minimum phytoplankton requirements (3-5 µg.L-1 soluble reactive phosphorus (SRP) and 80-100 µg.L-1 dissolved inorganic nitrogen [DIN]) (Reynolds, 1997). The elevated phosphorus and nitrogen concentrations in the reservoir demonstrated that DIN: SRP ratios alone do not explain temporal phytoplankton community dynamics.

The present study showed that seasonality is the principal factor driving the temporal dynamics of phytoplankton communities in eutrophic reservoirs of semiarid regions. This finding concurs with Seip and Reynolds (1995) who reported that the physical factors affected by seasonal gradients explain the distributions and abundances of these organisms. In the Argemiro de Figueiredo Reservoir, the physical and chemical characteristics of the water influenced phytoplankton functional groups structure. Physical factors, particularly overflow events, broke the dominance of cyanobacterial functional groups and opened windows of opportunity for phytoflagellate, coccoid chlorophyceae, diatom, and elongated desmid biomass to increase.

Intensive pisciculture caused significant anthropogenic disturbances in the study reservoir. The greatest phytoplankton biomass occurred at pisciculture sites near the net cages (PNC). Pisciculture interfered with the species compositions of local phytoplankton functional groups and caused cyanobacteria from Groups S1, SN, and K to predominate. Fish ration residues near the net cages enriched nutrient levels in the reservoir system and significantly increased phytoplankton biomass. The shallow water at this point (Zmax = 7.7 m) increased interactions with bottom sediments and synergistically enhanced nutrient availability. This area is distant from the reservoir dam and is not, therefore, affected much by overflow events. These conditions favor high cyanobacterial biomass. Pisciculture residues increase nitrogen and phosphorus concentrations in the water column and sediments, which provokes eutrophication (Figueredo & Giani, 2005; Guo, Zhongjie, Xie, & Ni, 2009; Borges, Train, Dias, & Bonecker, 2010). Studies have also shown that Nile tilapia (Oreochromis niloticus Linnaeus, 1758) can cause ichthyoeutrophication due to their high defecation rates (Lazzaro et al., 2003; Panosso et al., 2007). Borges et al. (2010) reported significant increases in phytoplankton (especially cyanobacterial) biomass after the introduction of pisciculture in the Rosana reservoir (Southeastern Brazil).

In summary, the present study demonstrated that (a) reservoir overflows are natural disturbances that reduce phytoplankton biomass and alter local community structures, (b) pisciculture is an anthropogenic disturbance that increases nutrient availability and stimulates increases in phytoplankton functional groups (mainly cyanobacterial) biomass; and (c) phytoplankton functional groups are reliable sentinels of environment conditions in the reservoirs of tropical semiarid regions.

ACKNOWLEDGMENTS

The authors would like to thank the Conselho Nacional de Desenvolvimento Científico e Tecnológico for their financial support through the universal project MCT/CNPq O2/2006 and for the doctoral grant awarded to the first author (Edital MCT/CNPq/CT-Hidro N° 040/2006), as well as the Laboratório de Saneamento da Universidade Federal de Campina Grande and the Laboratório de Microplâncton do Núcleo de Estudos e Pesquisas dos Recursos do Mar (NEPREMAR) at the Universidade Federal da Paraíba.

RESUMEN

Grupos funcionales de fitoplancton en una represa tropical de una región semiárida de Brazil. La estructura de grupos funcionales del fitoplancton y la abundancia de especies varían de acuerdo con las condiciones ambientales. Este estudio investigó los estresores naturales y antropogénicos que afectan la biomasa de grupos funcionales de fitoplancton en una represa (la represa de Argemiro de Figueiredo) en una región semiárida de Brasil. La recolecta de datos fue entre agosto 2007 y julio 2009 de forma bimensual durante el primer año, y de forma mensual durante los últimos dos años. Estos se recolectaron en tres sitios (PC: confluencia del río; PNC: cerca de las jaulas; PD: sitio de la represa). El análisis de las variables abióticas del agua incluyó: temperatura, transparencia, coeficiente de atenuación vertical de la luz, oxígeno disuelto, pH, conductividad eléctrica, alcalinidad, nitrógeno inorgánico disuelto y fósforo reactivo soluble. Las muestras de fitoplancton fueron recolectadas en una botella Van Dorn, y fueron preservadas en lugol acético y cuantificadas utilizando un microscopio invertido para determinar la densidad y la biomasa del fitoplancton, las especies identificadas fueron agrupadas en grupos funcionales. Los datos fueron explorados mediante un análisis de correspondencia canónica. Los análisis individuales fueron hechos para probar la variabilidad espacial y temporal de los datos y los factores que más interfieren con las variables bióticas y abióticas. Los grupos funcionales S1, SN, y K, incluyen las algas filamentosas: Planktothrix agardhii (Gomont) Anagnostidis & Komárek, Cylindrospermopsis raciborskii (Woloszynska) Seenaya & Subba Raju, y algas cocoides: Aphanocapsa incerta (Lemmermann) Cronberg & Komárek, respectivamente dominando los meses cálidos cuando el agua estuvo caliente, turbia y alcalina. El desbordamiento de la reserva funciona como una alteración natural, reduciendo la biomasa del fitoplancton a menos de un 50 % y la dominancia de cianobacterias, promoviendo el dominio de los grupos funcionales F, M, MP, Lo y X2. La llegada de nutrientes debido a la pesca intensiva, asociado con una baja profundidad local (Zmax = 7.7 m), cerca de las jaulas (PNC), resulta en una alteración humana significativa que incrementa la prevalencia de los grupos funcionales S1, SN y K, los cuales están compuestos principalmente por cianobacterias. Concluimos que, en las represas, eventos de desborde son perturbaciones naturales que tienen la habilidad para reducir la biomasa del fitoplancton y alterar la estructura de las comunidades locales, y que la pesca intensa es una alteración antropogénica que incrementa la disponibilidad de nutrientes y estimula el incremento de la biomasa de los grupos funcionales que incluyen las cianobacterias. Además, los grupos funcionales de fitoplancton fueron controles confiables de las condiciones ambientales en las represas de las regiones semiáridas tropicales.

Palabras clave: fitoplancton, grupos funcionales, condiciones climáticas, estabilidad de la biomasa, reservorio, región semiárida, factores estresantes naturales y antropogénicos.

REFERENCES

APHA-American Public Health Association. (2005). Standard methods for examination of water and wastewater (20th ed). Washington, DC: APHA, AWWA & WEF.

Barbosa, J. E. L., & Mendes, J. S. (2005). Estrutura da comunidade fitoplanctônica e aspetos físicos e químicos das águas do reservatório de Acauã-semiárido paraibano. In Anais da X Reunião da Sociedade Brasileira de Ficologia (Eds.), (pp. 339-360). Rio de Janeiro: Museu Nacional

Barone, R., & Flores, L. N. (1994). Phytoplankton dynamics in a shallow, hypertrophic reservoir (Lake Arancio, Sicily). Hydrobiologia, 289, 199-214.

Bonilla, S., Aubriot, L., Soares, M. C. S., González-Piana, M., Fabre, A., Huszar, V. L. M., & Kruk, C. (2012). What drives the distribution of the bloom-forming cyanobacteria Planktothrix agardhii and Cylindrospermopsis raciborskii? FEMS Microbiology Ecology, 79, 594-607.

Bouvy, M., Falcão, D., Marinho, M., Pagano, M., & Moura, A. (2000). Occurrence of Cylindrospermopsis (Cyanobacteria) in 39 Brazilian tropical reservoirs during the 1998 drought. Aquatic Microbial Ecology, 23, 13-27.

Borges, P. A. F., Train, S., Dias, J. D., & Bonecker, C. C. (2010). Effects of fish farming on plankton structure in a Brazilian tropical reservoir. Hydrobiologia, 649, 279-291.

Borges, P. A. F., Train, S., & Rodrigues, L. C. (2008). Spatial and temporal variation of phytoplankton in two subtropical Brazilian reservoirs. Hydrobiologia, 607, 63-74.

Brasil, J., & Huszar, V. L. M. (2011). O papel dos traços funcionais na ecologia do fitoplâncton continental. Oecologia Australis, 15, 799-834.

Câmara, F. R. A., Rocha, O., Pessoa, E. K. R., Chellappa, S., & Chellappa, N. T. (2015). Morphofunctional changes of phytoplankton community during pluvial anomaly in a tropical reservoir. Brazilian Journal of Biology, 75, 628-637.

Chellappa, N. T., Chellappa, T., Câmara, F. R. A., Rocha, O., & Chellappa, S. (2009a). Impact of stress and disturbance factors on the phytoplankton communities in Northeastern Brazil reservoir. Limnologica, 39, 273-282.

Chellappa, N. T., Câmara, F. R., & Rocha, O. (2009). Phytoplankton community: indicator of water quality in the Armando Ribeiro Gonçalves reservoir and Pataxó channel, Rio Grande do Norte, Brazil. Brazilian Journal of Biology, 69, 241-251.

Chellappa, N. T., Chellappa, S. L., & Chellappa, S. (2008). Harmful phytoplankton blooms and fish mortality in a eutrophicated reservoir of Northeast Brazil. Brazilian Archives of Biology and Technology, 51, 833-841.

Cole, G. (1994). Textbook of limnology. Illinois: Waveland Press.

Dantas, Ê. W., Moura, A. N., & Bittencourt-Oliveira, M. do C. (2011). Cyanobacterial blooms in stratified and destratified eutrophic reservoirs in semi-arid region of Brazil. Anais Da Academia Brasileira de Ciências, 83, 1327-1338.

Dejenie, T., Asmelash, T., De Meester, L., Mulugeta, A., Gebrekidan, A., Risch, S., … Declerck, S. (2008). Limnological and ecological characteristics of tropical highland reservoirs in Tigray, Northern Ethiopia. Hydrobiologia, 610, 193-209.

Douma, M., Ouahid, Y., Campo, F. F. Del, Loudiki, M., Mouhri, K., & Oudra, B. (2010). Identification and quantification of cyanobacterial toxins (microcystins) in two Moroccan drinking-water reservoirs (Mansour Eddahbi, Almassira). Environmental Monitoring and Assessment, 160, 439-450.

Fabbro, L. D., & Duivenvoorden, L. J. (2000). A two-part model linking multidimensional environmental gradients and seasonal succession of phytoplankton assemblages. Hydrobiologia, 438, 13-24.

Figueredo, C. C., & Giani, A. (2005). Ecological interactions between nile tilapia (Oreochromis niloticus, L.) and the phytoplanktonic community of the furnas reservoir (Brazil). Freshwater Biology, 50, 1391-1403.

Golterman, H. L., Clymo, R. S., & Ohnstad, M. A. M. (1978). Methods for Physical and Chemical Analysis of Freshwaters. Oxford, UK: IBP Handbook. Blackwell Science Publication.

Governo do Estado da Paraiba. (2007). Plano Estadual de Recursos Hídricos: Resumo Executivo e Atlas. João Pessoa: Secretaria de Estado da Ciência e Tecnologia e do Meio Ambiente - SECTA, Agência Executiva de Gestão das Águas do Estado da Paraíba - AESA.

Guo, L., Li, Z., Xie, P., & Ni, L. (2009). Assessment effects of cage culture on nitrogen and phosphorus dynamics in relation to fallowing in a shallow lake in China. Aquaculture International, 17, 229-241.

Gurbuz, H., Kivrak, E., Soyupak, S., & Yerli, S. V. (2003). Predicting dominant phytoplankton quantities in a reservoir by using neural networks. Hydrobiologia, 504, 133-141.

Hillebrand, H., Dürselen, C. D., Kirschtel, D., Pollingher, U., & Zohary, T. (1999). Biovolume calculation for pelagic and benthic microalgae. Journal of Phycology, 35, 403-424.

Huszar, V. L. M., Silva, L. H. S., Marinho, M., Domingos, P., & Sant’Anna, C. L. (2000). Cyanoprokaryote assemblages in eight productive tropical Brazilian waters. Hydrobiologia, 424, 67-77.

Karadžić, V., Subakov-Simić, G., Krizmanić, J., & Natić, D. (2010). Phytoplankton and eutrophication development in the water supply reservoirs Garaši and Bukulja (Serbia). Desalination, 255, 91-96.

Kosten, S., Huszar, V. L. M., Mazzeo, N., Scheffer, M., Sternberg, L. D. S. L., & Jeppesen, E. (2009). Lake and watershed characteristics rather than climate in shallow lakes influence nutrient limitation. Ecological Applications, 19, 1791-1804.

Kruk, C., Mazzeo, N., Lacerot, G., & Reynolds, C. S. (2002). Classification schemes for phytoplankton: a local validation of a functional approach to the analysis of species temporal replacement. Journal of Plankton Research, 24, 901-912.

Kruk, C., Peeters, E. T. H. M., Van Nes, E. H., Huszar, V. L. M., Costa, L. S., & Scheffer, M. (2011). Phytoplankton community composition can be predicted best in terms of morphological groups. Limnology and Oceanography, 56, 110-118.

Lazzaro, X., Bouvy, M., Ribeiro-Filho, R. A., Oliviera, V. S., Sales, L. T., Vasconcelos, A. R. M., & Mata, M. R. (2003). Do fish regulate phytoplankton in shallow eutrophic Northeast Brazilian reservoirs? Freshwater Biology, 48, 649-668.

Lins, R. P. M., Barbosa, L. G., Minillo, A., & De Ceballos, B. S. O. (2016). Cyanobacteria in a eutrophicated reservoir in a semi-arid region in Brazil: dominance and microcystin events of blooms. Revista Brasileira de Botânica, 39, 583-591.

Lopes, M. R. M., Ferragut, C., & Bicudo, C. E. M. (2009). Phytoplankton diversity and strategies in regard to physical disturbances in a shallow, oligotrophic, tropical reservoir in Southeast Brazil. Limnetica, 28, 159-174.

Lund, J. W. G., Kipling, G., & Le Creen, E. D. (1958). The inverted microscope method of estimating algae numbers and the statistical basis of estimation by counting. Hydrobiologia, 11, 143-170.

Mackereth, F. J. H., Heron, J., & Talling, J. F. (1978). Water analysis: some revised methods for limnologists. London: Freshwater Biological Association.

Marengo, J. A., Alves, L. M., Beserra, E. A., & Lacerda, F. F. (2011). Variabilidade e mudanças climáticas no semiárido brasileiro. In S. S. Medeiros, H. R. Gheyi, C. O. Galvão, & V. P. S. Paz (Eds.), Recursos hídricos em regiões áridas e semiáridas (pp. 384-422). Campina Grande: IJUSA-Instituto Nacional do Semiárido.

Morris, D. P., & Lewis, W. M. (1988). Phytoplankton nutrient limitation in Colorado mountain lakes. Freshwater Biology, 20, 315-327.

Naselli-Flores, L., Barone, R., Chorus, I., & Kurmayer, R. (2007). Toxic Cyanobacterial Blooms in Reservoirs Under a Semiarid Mediterranean Climate: The Magnification of a Problem. Environmental Toxicology, 22, 399-404.

Naselli-Flores, L. (2013). Morphological analysis of phytoplankton as a tool to assess ecological state of aquatic ecosystems: The case of Lake Arancio, Sicily, Italy. Inland Waters, 4, 15-26.

Oliveira, F. H. P. C., da Silva, J. D. B., Costa, A. N. S. F., Ramalho, W. P., Moreira, C. H. P., & Calazans, T. L. S. (2015). Comunidade de cianobactérias em dois reservatórios eutróficos e tropicais no nordeste do Brasil. Acta Scientiarum - Biological Sciences, 37, 169-176.

Padisák, J., Crossetti, L. O., & Naselli-Flores, L. (2009). Use and misuse in the application of the phytoplankton functional classification: A critical review with updates. Hydrobiologia, 621, 1-19.

Panosso, R., Costa, I. A. S., Souza, N. R., de Attayde, J. L., Cunha, S. R. de S., & Gomes, F. C. F. (2007). Cianobactérias e cianotoxinas em reservatórios do estado do Rio Grande do Norte e o potencial controle das florações pela tilápia do nilo (Oreochromis niloticus). Oecologia Brasiliensis, 11, 433-449.

Poole, H. H., & Atkins, W. R. G. (1929). Photo-electric measurements of submarine illumination throughout the year. Journal of Marine Biological Association of the United Kingdom, 16, 297-324.

Poulíčková, A., Hašler, P., & Kitner, M. (2004). Annual cycle of Planktothrix agardhii (Gom.) Anagn. & Kom. nature population. International Review of Hydrobiology, 89, 278-288.

Reynolds, C. S. (1997). Vegetation Processes in the Pelagic: A model for ecosystem theory. Germany: Ecology Institute.

Reynolds, C. S. (1998). What factors influence the species composition of phytoplankton in lakes of different trophic status? Hydrobiologia, 369370, 11-26.

Reynolds, C. S., Elliott, A. J., & Frassl, M. A. (2014). Predictive utility of trait-separated phytoplankton groups: A robust approach to modeling population dynamics. Journal of Great Lakes Research, 40, 143-150.

Reynolds, C. S., Huszar, V., Kruk, C., Naselli, L., & Melo, S. (2002). Towards a functional classification of the freshwater phytoplankton. Journal Plankton Research, 24, 417-428.

Reynolds, C. S., Padisak, J., & Sommer, U. (1993). Intermediate disturbance in the ecology of phytoplankton and themaintenance of species diversity: a synthesis. Hydrobiologia, 249, 183-188.

Seip, K. L., & Reynolds, C. S. (1995). Phytoplankton functional attributes along trophic gradient and season. Limnology and Oceanography, 40, 589-597.

Silva, D. F., Sousa, F. A. S., Kayano, M. T., & Araujo, L. E. (2008). Climatic accompaniment of watersheds from Mundaú River, State of Alagoas and Pernambuco, and from Paraiba River, State of Paraiba, Brazil (in Portuguese). Engenharia Ambiental, 5, 79-93.

Straskraba, M., Tunidisi, J. G., & Duncan, A. (1993). State-of-art of reservoir limnology and water quality management. In M. Straskraba, J. G. Tundisi, A. Duncan (Eds.), Comparative Reservoir Limnology and Water Quality Management (pp. 213-288). Netherlands: Kluwer Academic Publishers.

Sun, J., & Liu, D. (2003). Geometric models for calculating cell biovolume and surface area for phytoplankton. Journal of Plankton Research, 25, 1331-1346.

Ter Braak, C. J. F., & Smilauer, P. (2002). CANOCO Reference Manual and CanoDraw for Windows User’s guide: Software for Canonical Community Ordination (version 4.5). Ithaca, New York: Microcomputer Power.

Toledo Jr, A. P., Talarico, M., Chinez S. J., & Agudo, E. G. (1983). The application of simple models for evaluating eutrophication processes in tropical lakes and reservoirs. In Annals of the 12th Brazilian Congress of Sanitary and Environmental Engineering (Eds.), (pp. 1-34). Camboriú: Brazilian Association of Sanitary and Environmental Engineering-Abes.

Török, P., T-Krasznai, E., B-Béres, V., Bácsi, I., Borics, G., & Tóthmérész, B. (2016). Functional diversity supports the biomass-diversity humped-back relationship in phytoplankton assemblages. Functional Ecology, 30, 1593-1602.

Utermöhl, H. (1958). Zur Vervollkommnung der quantitativen Phytoplankton-Methodik.. Verhandlungen der Internationalen Vereinigung für Theoretische und Angewandte Limnologie, 9, 1-38.

Wetzel, R. G., & Likens, G. E. (2000). Limnological Analysis. New York: Springer-Verlag.

Znachor, P., Zapomělová, E., Řeháková, K., Nedoma, J., & Šimek, K. (2008). The effect of extreme rainfall on summer succession and vertical distribution of phytoplankton in a lacustrine part of a eutrophic reservoir. Aquatic Sciences, 70, 77-86.

TABLE 2

Summary of the covariance analyses performed on the abiotic variables, phytoplankton species and functional groups related found

in the Argemiro de Figueiredo Reservoir (Brazil) from August 2007 to July 2009

Secchi

pH

EC

DO

Alkalinity

DIN

SRP

S1

M

P

SN

F

K

MP

H1

X2

Lo

Plag

Pslin

Maer

Mcpr

Aulg

Closp

Crac

Botb

Apin

Ossp

Dcir

Chlsp

Pumb

Collection points

-

*

-

**

-

-

-

-

-

-

*

-

-

-

*

-

-

-

-

*

Depth

-

***

-

***

*

**

*

-

-

-

-

-

**

-

*

-

-

-

**

-

Precipitation

-

-

-

-

-

-

-

-

*

-

-

-

-

-

-

-

-

*

-

-

Overflow

***

**

***

***

***

-

-

*

-

-

**

-

*

-

-

-

-

-

**

-

Significant (* = p < 0.05, ** = p < 0.01, *** = p < 0.001); - = not significant (p > 0.10); EC = electrical conductivity, DO = dissolved oxygen, DIN = dissolved inorganic nitrogen, SRP = soluble reactive phosphorus, Plag = Planktothrix agardhii, Pslin = Pseudanabaena limnetica, Maer = Microcystis aeruginosa, Mcpr = Microcystis protocystis, Aulg = Aulacoseira granulata, Closp = Closterium sp., Crac = Cylindrospermopsis raciborskii, Botb = Botryococcus braunii, Apin = Aphanocapsa incerta, Ossp = Oscillatoria sp., Dcir = Dolichospermum circinalis, Chlsp = Chlamydomonas sp., Pumb = Peridinium umbonatum.