INTRODUCTION

The role of estuaries regarding nutrient inputs to the coastal zone has been extensively studied because of its importance to energy transfer and fish production (^{Alpine and Cloern, 1992}; ^{Dutto et al., 2014}; ^{Nelson et al., 2015}). This nutrient load is generated mainly by the increase in human populations and their activities in the coastal zones, which, when excessive, can lead to eutrophication and affect the quality and use of the resources of these ecosystems (^{Nixon, 1995}; Cloern and Jassby, 2010; ^{Paerl et al., 2014}). To determine the trophic status of aquatic systems, indices have been established that reflect the level of deterioration of water quality, and these indices include factors such as nutrient concentration, water transparency, dissolved oxygen concentration, among others, and the effects of this deterioration on life such as phytoplankton, marine grasses, and macroalgae (^{Vollenweider, 1992}; ^{Birk et al., 2012}). All indices are applied according to the feasibility of measuring various factors, including the ^{Karydis index (1992)}, which is based mainly on the concentration of nutrients and has specificity for each one of them. In addition, this index is applied to various types of water, it is highly sensitive to the effects of eutrophication, and it is easy to apply. Initially, this index was established as a model for European waters, but it has been used successfully in the Americas, such as in the Gulf of Mexico (^{Herrera-Silveira et al., 2004}; ^{Tapia-González et al., 2008};^{Moreno et al., 2010}).

The measurement of the trophic level is important for establishing the type of control and monitoring used in the management of ecosystems in short- and medium- term projects. In addition, the current trend suggests that investigations of the trophic status of the coastal zone should be accompanied by knowledge of the community structure of microphytoplankton because it is the community that best responds as an indicator of changes in coastal hydrography (^{Hays et al., 2005}; ^{Reynolds, 2006}; ^{Vinagre and Costa, 2014}). Therefore, Tsirstis and Karydis (1998) indicated that phytoplankton diversity reacts to changes in the trophic level if polluted environments favor the growth of a few opportunistic species.

In Ecuador, 70% of the inland water bodies flow into the coastal zone (^{Rendón et al., 1983}), with most of the country’s fishing and aquaculture activity concentrated in the Gulf of Guayaquil (^{San Martín, 2009}). However, the nutrient load that comes from untreated wastewater of agricultural, livestock, domestic, industrial, and aquacultural origins gives water bodies their trophic characteristics, which affect the water quality and the diversity of the resources from which it develops. In this sense, studies on the hydrographic and biological characteristics of the coastal zone of Ecuador show spatial and temporal changes in the community structure of microphytoplankton in the Gulf of Guayaquil, where the dominance of halotolerant diatoms such as Paralia sulcata and Skeletonema costatum, and oceanic species, such as Rhizosolenia imbricata, Eucampia cornuta, Pseudonitzschia seriata and Guinardia striata, have been found in the coastal zone (^{Prado et al., 2015}).

Regarding community structure and its relation to hydrographic conditions, studies have focused mainly on the influence of the water masses of the currents that dominate the Ecuadorian coast, such as the currents of Panama during the rainy season and Humboldt in the dry period, using temperature as an indicator and determining the absolute dominance of diatoms in both periods (^{Jiménez, 1983}; ^{Gualancañay et al., 2003}, ^{Coello et al., 2010}; ^{Prado and Cajas, 2010}a, 2010b). Based on the studies on plankton and hydrographic conditions, carried out 1.8 km off the coast of the province of Guayas, the goal of this study is the determination of the trophic condition in the interior estuary of the Gulf of Guayaquil and adjacent coastal zone relative to the structure of the microphytoplankton.

STUDY AREA

Ecuador is a country with only two seasons: a rainy season from January to May and a dry season from July to November, whereas December and June are considered transitional months. The province of Guayas is in Southwest Ecuador and is influenced by several marine currents, especially the cold Humboldt current in the dry season and the warm El Niño current in the rainy season, producing a tropical savanna and tropical monsoon climate with average temperatures of 25°C and annual precipitation between 500 and 1000 mm (^{Twilley et al., 2001}).

In the coastal zone of this province is the Gulf of Guayaquil (3°S, 80°W), in which an outer and an inner estuary have been established; the former has its limit in the platform along the meridian 81°W to the interior of the island Puná (80°15’W) in the mouths of the El Morro and Jambelí channels. The depth of the Gulf of Guayaquil on the continental shelf ranges from 180 m on the western edge to less than 20 m on the interior (^{Montaño-Armijos and Sanfeliu-Montolio, 2008}).

The inland estuary is known as the Guayas River Estuary; to the northwest is the Salado Estuary, which receives sewage from the city of Guayaquil; and to the southeast lies the Churute Estuary, which receives the Taura and Churute rivers. A total of 24 hydrographic basins discharge into the Gulf of Guayaquil, with the Guayas River, which is formed by the confluence of Daule and Babahoyo (^{Twilley et al., 2001}; ^{Montaño-Armijos and Sanfeliu- Montolio, 2008}), being the most important.

The study area corresponds to the area of fishing activity along the coast of Guayas Province and northwest of Puná Island, 1.8 km away from the continental margin and where 16 stations were located, of which eight were in the coastal area (stations 1-7 and 16) and eight were in the interior of the Gulf of Guayaquil (stations 8 to 15) (Figure 1).

Figure 1 Geographic location of the sampling stations in Guayas Province. 

MATERIALS AND METHODS

Sampling and analysis

Surface surveys were carried out between 8:00 and 11:00 AM on board an outboard motorboat, 1.8 km from the coast from June to November 2012. In each sample, temperature and salinity were measured with a Seabird CTD, model Seacat S19. Simultaneously, a sample of water was collected at each station using a 5-L Niskin bottle, from which 500 mL was taken for the determination of nitrite, nitrate, phosphate, and silicate, according to the methodology of ^{Parsons et al. (1984)}, and 240 mL was taken for the microphytoplankton analysis and preserved with Lugol solution for later analysis in the laboratory.

The identification of the microphytoplankton species was carried out following the codes of ^{Cupp (1943)}, ^{Cleve (1951)}, ^{Schiller (1971)}, ^{Jiménez (1983)}, ^{Tomas (1997)}, and ^{Bérnard-Therriault et al. (1999)}, whereas species abundance was estimated using the Utermöhl method referred to in ^{Edler and Elbrächter (2010)}. In this method, a motorized inverted Leica DMI8 microscope with Leica Application Suite (LAS) software and 10 mL sedimentation chambers were used, with the samples being allowed to stand in the chambers for 24 h. Next, the quantitative data were used to calculate, using the logarithm base 2, the total abundance, relative abundance of functional groups, indices of equitability, richness, and the Shannon and Wiener index (Tsirstsis and Karydis,1998; ^{Krebs, 1999}).

To characterize the trophic status by zones, based on each nutrient, the following formula was used to calculate the ^{Karydis multivariate index (1992)}:

where TI is the trophic index for a specific nutrient, C is the sum of the nutrient concentration per season, Xi is the average monthly nutrient concentration per season, and A is the number of seasons.

The numerical scale that determines the trophic level is the following: <3, oligotrophic; 3-5, mesotrophic; and > 5, eutrophic.

RESULTS

Environmental variables

The average temperature was 25.2°C, with the minimum value (24.3°C) in August and the maximum (26.5°C) in June. Monthly differences and differences between significant zones (p <0.05) were detected, with the lowest records in the coastal zone (Figure 2a and b). The average salinity was 31.3, with the minimum (30.6) in June and the maximum (32.2) in December. Significant monthly and interzone differences (p <0.05) were recorded with the lowest records in the internal zone (Figure 2c and d).

Figure 2 Monthly and zone variation of a and b) temperature, c and d) salinity, e and f) dissolved oxygen, g and h) transparency in the coastal zone (Coastal) and in the inland estuary of the Gulf of Guayaquil (estuary) along the coast of Guayas Province between June and December 2012. 

The mean dissolved oxygen concentration was 7.4 mg/L, with a minimum (6.2 mg/L) in June and a maximum (8.5 mg/L) in November, with significant differences (p <0.05) recorded in the estuary (Figure 2e and f). The average transparency was 1.9 m, with the lowest (1.3 m) in July and the maximum (2.6 m) in June, with significant monthly differences (p <0.05) and differences between zones, with the lowest records in the inland estuary (Figure 2g and h).

Nitrite varied significantly monthly and between zones (p <0.05), with the lowest values being recorded in the coastal zone. The mean concentration of nitrite was 0.25 μmol/L, with the lowest concentration (0.16 μmol/L) in November and the maximum (0.36 μmol/L) in August (Figure 3a and b).

Figure 3 Monthly and zone variation of a and b) nitrite, c and d) nitrate, e and f) phosphate, and g and h) silicate in the coastal zone (Coastal) and inland estuary of the Gulf of Guayaquil (Estuary) along the coast of Guayas Province between June and December 2012. 

In the case of nitrate, the lowest concentration occurred in September (4.9 μmol/L) and the highest in November (8.3 μmol/L), with an average of 5.8 μmol/L. No significant monthly differences (p> 0.05) were observed, but differences were found between zones (p <0.05), with the lowest concentrations being recorded in the coastal zone (Figure 3c and d).

Phosphate had concentrations between 1.1 μmol/L in December and 1.8 μmol/L in August; the mean value was 1.4 μmol/L. No significant differences (p> 0.05) were observed, but differences were observed between zones (p <0.05), with the lowest concentrations being recording in the coastal zone (Figure 3e and f).

The average concentration of silicate was 45.6 μmol/L, with the lowest concentration (13.6 μmol/L) in December and the maximum (84.6 μmol/L) in June. In this case, the spatial (by zone) and temporal (monthly) differences were significant (p <0.05), with the lowest values being recorded in the coastal zone (p <0.05) (Figure 3g and h).

In general, in the coastal zone, lower temperatures and nutrient concentrations were observed, but higher salinity, oxygen, and transparency were observed in the coastal zone compared to the internal area of the Gulf of Guayaquil (Table 1).

Table 1 Averages of physical-chemical variables of the coastal zone and inland estuary of the Gulf of Guayaquil along the coast of Guayas Province between June and December of 2012. 

The analysis of the hydrographic variables through the ACP showed 73.0% of accumulated variance in the first three components in the coastal zone; in the first axis, an inverse relation to salinity and transparency was observed between nutrients, while in the second component, the relation was inversely related to temperature and dissolved oxygen (Figure 4a). In the inner estuary, the variance accumulated in the first three axes was 71.0%; in component 1, the silicate, phosphate, and nitrite correlated inversely with the transparency, salinity, and dissolved oxygen, whereas in component 2, an inverse correlation was detected between temperature and nitrate (Figure 4b).

Figure 4 Orthogonal projection of the first two components of the principal components analysis in a) coastal zone and b) inland estuary of the Gulf of Guayaquil off Guayas Province between June and December 2012. (TEMP: temperature; salinity, OD: dissolved oxygen, NO2: nitrite, NO3: nitrate, PO4: phosphate, SiO4: silicate, Trans: transparency). 

Trophic levels

The range of variation in the trophic index values was between 3 and 7, with significant differences in the trophic level between the coastal zone and the inner estuary, the latter being the one with the highest values for all nutrients. Thus, for nitrite, the lowest value was observed at station 6, which corresponds to the sampling site farthest from the coast, while the highest values were observed at stations located in the Channel of Cascajal, Balao, and West of the island of Puná (stations 11, 15, and 16, respectively) (Figure 5a and b). For nitrate, the trophic index values were low throughout the coastal zone, while in the inner estuary, all values were greater than 4 Karydis units, especially at station 13, which corresponds to the sector of El Guabo (Figure 5c and d). A similar tendency occurred with phosphate (Figure 5e and f) and silicate (Figure 5g and h).

Figure 5 Seasonal variation of the Karydis trophic index (TI) for a) nitrite, b) nitrate, c) phosphate, and d) silicate in the coastal zone (Coastal) and inland estuary of the Gulf of Guayaquil (Estuary) along the coast of Guayas Province between June and December 2012. 

Microphytoplankton

Seventeen species of microphytoplankton were identified, of which approximately 78% were diatoms, 18% dinoflagellates, 2% cyanobacteria, and 2% silicoflagellates (Annex 1). The average abundance of microphytoplankton was 7.0 x 105 cell/L with a minimum of 4.1 x 105 cell/L in August and a maximum of 10.6 x 105 cell/L in July, showing significant differences between months (KW: 28.73; p <0.05) and between zones (MW-U: 2087; p <0.05), with the greatest abundance in the interior estuary (Figure 6a and b). The average richness was 19 species, and monthly differences (KW: 21.28; p <0.05) were observed, showing a similar tendency to abundance because the lowest value was observed in August (12 species) and the highest in July (23 species). No significant differences in richness (MW- U: 1428; p> 0.05) were observed between zones (Figure 6c and d). The minimum equitability (0.76) was recorded in August and the highest (0.86) in December, with significant differences between months (KW: 12.71, p <0.05) and between zones (MW-U: 12.71, p <0.05), with the highest equitability in the coastal zone (Figure 6e and f).

Figure 6 Monthly variation of a and b) abundance, c and d) richness, e and f) equitability, and g and h) diversity of microphytoplankton in the coastal zone (Coastal) and inland estuary of the Gulf of Guayaquil (Estuary) along the coast of Guayas Province between June and December 2012. 

The average diversity was 3.4 bits/Ind, with the highest value in October (3.8 bits/Ind) and the lowest in August (2.7 bits/Ind), months in which the abundance was lower. Significant differences between months (KW: 19.6; p <0.05) and between zones (MW-U: 19.6; p <0.05) were detected, with the greatest diversity occurring in the coastal zone (Figure 6g and h).

When exploring the similarity relations (MDS- Anosim) between the microphytoplankton communities, significant differences (R: 0.44, p <0.05) were established between the coastal and inland estuary areas of the Gulf of Guayaquil; however, some species are found in both zones, with greater heterogeneity being observed in the coastal zone, with a stress of 0.18 (Figure 7).

Figure 7 Analysis diagram of multidimensional scaling (MDS) of the community structure of microphytoplankton in the coastal zone and inland estuary of the Gulf of Guayaquil along the coast of Guayas Province between June and December 2012. 

When similarities percentages (SIMPER) were analyzed to explore aspects of the internal structure of the inner and coastal estuary areas, differences were noted between the two zones. Paralia sulcata, Nitzschia longissima, Skeletonema costatum, and Guinardia striata were the species that contributed the most to separating the zones (Table 2).

Table 2 Summary of the SIMPER analysis of the microphytoplankton species in the coastal zone and inland estuary of the Gulf of Guayaquil along the coast of Guayas Province between June and December 2012. 

The redundancy analysis (RDA) allowed the relationship between the microphytoplankton structure and the environmental variables of the interior estuary and coastal zone to be studied, showing a correlation of 0.54 and 0.58 in the first two components with a significant Monte Carlo analysis (p <0.05). In the first component, a high positive correlation was observed for diatoms Paralia sulcata, T. nitzschioides, T. frauenfeldii, Navicula transitrans, and Skeletonema costatum, with nutrients and negative correlation with salinity. In the second component, Guinardia striata, Cerataulina pelagica, Nitzschia longissima, Chaetoceros decipiens, and Navicula transitrans var. derasa were significantly correlated with temperature, and in turn, they were inversely related to dissolved oxygen (Figure 8a).

Figure 8 Orthogonal projection of the redundancy analysis in the a) coastal zone and b) interior estuary of the Gulf of Guayaquil along the coast of Guayas Province between June and December of 2012. TEMP: temperature; SAL: salinity; DO: dissolved oxygen; NO2: nitrite; NO3: nitrate; PO4: phosphate; SiO4: silicate; Theccen: Thalassionema eccentrica; Csnrad: Coscinodiscus radiatus; Melomon: Melosira moniliformis; Parsul: Paralia sulcata; Thlin: Thalassiosira limnetica; Csnsp: Coscinodiscus sp.; Chdec: Chaetoceros decipiens; Ntlong: Nitzschia longissima; Odsin: Odontella sinensis; Guinstr. Guinardia striata; Cerpel: Pelagic Cerataulina; Rhimb: Rhizosolenia imbricata; Chcurv: Chaetoceros curvisetus; Navtransde: Navicula transitrans var. derasa; Navtrans: Navicula transitrans; Lepdan: Leptocylindrus danicus; Sklcos: Skeletonema costatum; Thnitzs: Thalassiosira nitzschoides; Odmob: Odontella mobiliensis; Thfrau: Thalassionema frauenfeldii. 

In the interior estuary, a correlation of 0.64 was found between the species and environment for the first two components, and the Monte Carlo test was significant (p <0.05). In the first axis, N. longissima, C. decipiens, and S. costatum were associated with temperature and, to a lesser extent, nitrite; Coscinodiscus radiatus and Melosira moniliformis were negatively related to phosphate and nitrate. In the second component, the silicate was inversely correlated with the diatoms T. frauenfeldii, Thalassiosira limnetica, P. sulcata, and N. transitrans, andthedinoflagellate Gyrodinium sp., which, in turn, were significantly related to salinity and dissolved oxygen (Figure 8b).

DISCUSSION AND CONCLUSIONS

The eutrophication processes undergone by estuaries in recent decades have been widely discussed by many authors; these processes occur not only because of the solid and liquid contributions from the natural processes of the rivers that discharge their waters into them but also from the high nutrient content that originates in anthropic activities. This interaction leads to physical processes such as turbidity and biogeochemical processes that generate an effect on the composition and abundance of microphytoplankton, with consequent effects on fisheries, aquaculture, and tourism (^{Alpine and Cloern, 1992}; ^{Paerl et al., 2014}).

In the present study, the differences in the hydrographic conditions of the coastal zone and estuary are clear, indicated by the relations among the nutrients, salinity, and transparency that are different for each zone; however, an interaction exists between the zones due to the contributions of the Bay of Guayaquil Estuary, which have a wide reach that affects much of the coastal area (^{Chiriguaya and Burgos, 1990}; ^{Torres et al., 2003}), as well as the exchange with the oceanic zone due to the extensive tidal regime on the Ecuadorian coast (^{Fiedler and Talley, 2006}). When the hydrographic variables in the inner estuary were analyzed, the inverse relation between the salinity and dissolved oxygen versus the silicate concentrations is evidence of continental influences across rivers that carry large amounts of sediment and turbid water, which show activity in the biogeochemical processes. Similar results have been reported by ^{O’Boyle and Silke (20109} and ^{Cloern et al. (2014)}, who emphasized the low concentrations of dissolved oxygen in estuaries.

As for the trophic condition of the study area, the contribution of nutrients of continental origin in the study area was significant; however, the levels reached did not indicate eutrophication, which could result from the location of the sampling stations, as they were located 1 mile from the continental margin, where exchange is high due to the dynamic tidal flow (^{Twilley et al., 2001}; ^{Fiedler and Talley, 2006}). However, recorded values greater than 3 Karydis units in nitrogen and phosphorus at the stations located in the interior estuary of the Gulf of Guayaquil indicate that this zone is at the limit of passing from the mesotrophic to the eutrophic state, an important aspect to consider for the management of this area. In this regard, ^{Herrera-Silveira et al. (2002)} and Herrera-Silveira and Morales-Ojeda (2009) recommend the use of intermediate categories (oligomesotrophic and mesoeutrophic) because, with only three levels, the Karydis index does not allow detection of trends from one level to another as all allowed with more advanced detection indexes of trophic status, a methodology that could be considered in future research. In this sense, studies showed that Cienfuegos Bay in Cuba was considered mesotrophic (^{Seisdedo et al., 2010}) even though all nutrients presented a trophic index of less than 3. In a coastal lagoon in Mexico, ^{Tapia-González et al. (2008)})obtained levels higher than 3, which is why they described the lagoon as mesoeutrophic in nitrate and silicate; these results are similar to the averages found in the interior estuary of Guayas. The levels are not alarming, but they indicate of an initial state of eutrophication produced by river inputs that are exacerbated by anthropogenic activities, requiring medium-term evaluation to exercise control plans (^{Cloern et al., 2016}).

On the other hand, the greatest diversity was recorded in October, with the lowest in August; Likewise, in the coastal zone, the diversity was higher than in the interior estuary, which is attributable to the great turbidity that exists in this latter zone and originates from the contribution of alien materials from contributing rivers because tropical aquatic ecosystems, especially those at low altitudes, such as rivers and estuaries, are usually very muddy due to the high material leaching that occurs in these regions. This turbidity prevents the passage of light, a situation that directly affects the productivity and energy flow of the ecosystem, as it affects phytoplankton development and diversity (^{Roldán and Ramírez, 2008}).

When establishing correlations between phytoplankton species and the environment, evidence of independent processes is important to consider. In other words, in the coastal zone, Paralia sulcata, T. frauenfeldii, Skeletonema costatum, and T. nitzschioides were tolerant to wide areas of salinity as well as being associated to areas with important nutrient inputs (^{Pednekar et al., 2014}; ^{Rakshit et al., 2015}), as is the normal tendency of the ecosystem of the Gulf of Guayaquil (^{Chiriguaya and Burgos, 1990}; ^{Gualancañay et al., 2003}; ^{Jimenez, 2008}). Guinardia striata, Cerataulina pelagica, and C. decipiens, in contrast, would be associated with waters of low temperature and high concentrations of dissolved oxygen, which are representative of the flow of the Humboldt current (^{Okuda et al., 1983}; ^{Macias, 1999}; ^{Pennington et al., 2006}).

In the estuarine zone, for the first component, the RDA showed C. decipiens and Nitzschia longissima to be associated to the temperature and nitrite, while showing an inverse correlation to nitrate and phosphate; these relations would indicate activity in biogeochemical processes that are typical of estuaries (^{Huisman et al., 2004}), and species such as Coscinodiscus radiatus and Melosira moniliformis are indicative of areas with high nutrient concentrations (^{De et al., 2015}). In component 2, species associated with low salinity and dissolved oxygen, as well as to high silicate, are present, such as P. sulcata, T. limnetica, N. transitrans, and Gyrodinium sp., which indicate that fluvial contributions are the main source of eutrophication (^{Day et al., 1989}; ^{Ninčević-Gladan et al., 2015}).

Importantly, although the coastal and estuarine areas of the study area are in constant interaction, the characteristic species of each area can be distinguished, with the estuarine community being more homogeneous than the coastal zone. This is because the estuary is subject to high turbulence; thus, species are limited in terms of light intensity and are adapted to low oxygen concentrations. Therefore, Paralia sulcata, Skeletonema costatum, and T. nitzschioides, which have been registered as indicators of eutrophic waters (^{Huisman et al., 2004}; ^{Garmendia et al., 2013}), should be considered for environmental baseline studies and environmental monitoring of the interaction between estuaries and coastal zone.

In summary, when compared to the Guayas Province, the hydrographic conditions in the coastal zone and inland estuary of the Gulf of Guayaquil were different in the structure, abundance, and diversity of microphytoplankton. In the coastal zone, the dominant species were Nitzschia longissima, Guinardia striata, and Leptocylindrus danicus, and in the interior estuary, S. costatum, Thalassionema frauenfeldii, and T. nitzschioides, which are more tolerant to turbulence, low oxygen, and low light intensity (^{Huisman et al., 2004}).

In addition, the entire study area was determined to be mesotrophic; however, the values in the inner estuary were higher, probably because the estuary receives contributions from anthropic activities more directly, and although the trophic levels 1.8 km from the coastal margin in front of Guayas Province did not indicate eutrophication, studies are recommended at a distance of approximately 500 m to reflect the situation of the ecosystem in an area closer to the coast; in addition, management plans should be developed to avoid situations where contributions from drainage basins and urban settlements increase coastal trophic conditions.