Modeling a hypersaline lagoon to evaluate salinity changes due to morphology of an ocean connection

Modelagem de uma laguna hipersalina para avaliar mudanas de salinidade devidas à morfologia de sua conecção com o mar

ABSTRACT

Araruama is a perennial hypersaline lagoon where seawater flows in through a narrow connection and spreads, where it is intensively evaporated. Although the inflow of seawater increases the salts load, a high inflow may reduce flushing time, decreasing salinity. Although reducing flushing time would improve water quality, the accompanying reduction in the salinity may change the ecological equilibrium. In the present research, a numeric hydrodynamic model was applied to Araruama Lagoon, to determine the maximum connection depth that promotes reduction of flushing time, however maintaining salinity. Five scenarios were executed in the hydrodynamic model SisBaHiA^© with changing hydraulic area, including channel average depths of 2.5 m, 3.0 m, and 4.0 m. The salinity simulations outlined three segmented sectors, the first associated with the connection channel, but also affecting the proximal portion of the lagoon; the second is an intense evaporation area, with high salinity values, and the third, in the western portion of the lagoon where freshwater inputs can affect salinity. The tested channel bathymetry scenarios promoted a salinity increase with a channel depth of 3.0 meters. With a connection channel depth of 4.0 meters (highest hydraulic area), the threshold limit is attained, and salinities are reduced in the lagoon.

Keywords:
Numerical modeling; Hydrodynamics; Salinity; Water quality; Araruama Lagoon

RESUMO

Araruama é uma laguna hipersalina perene com conexão por onde a água do mar penetra e se difunde, sendo intensamente evaporada. Embora os fluxos de água do mar aumentem a carga de sais para a laguna, aumentando a salinidade, é esperado que um alto fluxo de água do mar venha a reduzir o tempo de renovação, mas decrescendo a salinidade. A redução do tempo de renovação é positiva para a qualidade da água, mas a consequente redução na salinidade pode promover mudanças no equilíbrio ecológico. No presente trabalho, um modelo numérico hidrodinâmico foi desenvolvido a fim de simular a profundidade máxima do canal (área hidráulica) que promova a redução no tempo de residência, mas que mantenha elevada a salinidade. Cinco cenários de simulação foram executados no modelo hidrodinâmico SisBaHiA^©, incluindo profundidades do canal de 2,5 metros; 3,0 metros e 4,0 metros (crescentes áreas hidráulicas. A simulação da salinidade mostrou três setores, um associado com o canal de conexão; o segundo é uma área de intensa evaporação com altos valores de salinidade e a terceira mais a oeste onde aportes de água doce afetam a salinidade. Os cenários testados geraram aumentos na salinidade até 3,0 metros, mas a 4,0 metros o limite é atingido e as salinidades da laguna reduzem o que poderia promover mudanças ecológicas.

Palavras-chave:
Modelagem numérica; Hidrodinâmica; Salinidade; Qualidade da água; Laguna de Araruama

INTRODUCTION

Coastal lagoons are subjected to both natural (extreme weather events) and anthropogenic (domestic sewage discharge) pressures (Meng et al., 2017). Among the anthropogenic activities developed in coastal areas, navigation, dredging, and drainage basin urban occupation have a strong influence over the water quality (Cutroneo et al., 2012; Jung et al., 2021; van Maren et al., 2015). Additionally, these activities increase the complexity of the physical processes related to water circulation (D’Alpaos et al., 2010; Valle-Levinson, 2008), as well as to water exchange between the system and the adjacent ocean (advective transport) (Burchard & Hofmeister, 2008; Venier et al., 2014).

When coastal lagoons are impacted by human occupation, regulatory agencies usually suggest engineering works to restore water quality (Bruun, 1994). The most common action is dredging connection channels to increase the hydraulic area, and to reduce flushing time, by increasing water exchange with the sea (Rodrigues et al., 2012; van Rijn, 2005). This procedure is intended to improve the system's loading capacity (Bruun, 1994). Conversely, changes in the morphology or bathymetry of the connection channel can affect the tidal prism, the structure of flood/ebb flows, hydrodynamics, salinity pattern, and water flushing time (Rynne et al., 2016; Salamena et al., 2016).

The hypersaline Araruama lagoon is a relevant touristic coastal environment located in the easternpart of the state of Rio de Janeiro - Brazil (Figure 1), which has been historically impacted by anthropogenic activities (Souza & Azevedo, 2020). During the last decades, among threats that affected the system, dredging of its connection channel (Canal de Itajurú) and installation of salt evaporation ponds are relevant. However, the extensive urban occupation of its surroundings is the most relevant impact in the system (Bertucci et al., 2016; Wasserman et al., 2019a). Due to these anthropic modifications, the Araruama lagoon has undergone environmental issues such as silting of the connection channel (Garcia-Silva & Rosman, 2016), changes in the primary production patterns (Oliveira et al., 2020), occurrence of diatom blooms, associated with eutrophication processes (Sylvestre et al., 2001). With the years, the trophic degraded from oligotrophic to mesotrophic (Vicente et al., 2021), leading to extensive fish mortalities (during the years 2009, 2010 and 2011; Wasserman et al., 2019b).

Figure 1
Araruama lagoon location (UTM coordinates of the zone 23K; SIRGAS 2000) and its bathymetry obtained from Wasserman et al. (2006).

Araruama Lagoon is a phosphorus limited environment (Souza et al., 2003; Vicente et al., 2021) and intense primary production can be observed whenever higher phosphate inputs occur. Moreira-Turcq (2000) studied a particularly rainy period (1989-1990), when salinities fell to 41 PSU (normally it is 52 PSU, or higher) and observed an intensification of the primary production, with higher phosphate concentrations. The input of phosphate was attributed to a physical chemical modification of the interstitial water that promoted the release of an authigenic phosphate minerals.

In coastal ecosystems, it has been shown that sediments can be a large storage compartment for nutrients (Cotner et al., 2004; Reimer & Huerta-Diaz, 2011) that can be authigenic, from organic phosphorus compounds, biogenic apatite or CaCO₃-bound phosphorus (Zhang et al., 2004). These compounds are partially refractive and can release phosphate into the water column under specific chemical conditions. Guimarães et al. (2021) carried out microcosm experiments (in vitro) with Araruama Lagoon sediments and overlying waters of variable salinities to test the release of nutrients but could not identify any extensive dissolution of phosphate. However, preliminary results of the geochemical speciation of phosphorus in the Araruama sediments confirmed that the nutrient is mainly in the form of authigenic apatite (Silva, 2019). Therefore, severe and long-term modifications in the salinity of Araruama Lagoon may release phosphate to the water column, contributing to eutrophication. This is the main reason why the control of salinity may be important in this type of environment.

The complexity of coastal impacted environments and the multiple processes involved demand an integrated approach (Zarzuelo et al., 2015) that can be done with hydrodynamic and advective transport models. Such approach provides an effective management basis for the improvement of these systems (Ferrarin & Umgiesser, 2005; Zacharias & Gianni, 2008). During the last decades, numerical models were developed and applied in several coastal areas to simulate hydrodynamic patterns (Chen et al., 2017; Colaço et al., 2021; Zhang et al., 2018). Both finite differences and finite elements models constitute prediction tools for many spatial-temporal scales, that can support decision-makers (Bedri et al., 2015; Webster, 2010; Zacharias & Gianni, 2008).

In the case of Araruama Lagoon, Kjerfve et al. (1996) showed that although the salinity of the ocean is lower than that of the lagoon (52 PSU or more), there is a net salt flow from the ocean to the lagoon. As the water flows into the lagoon through the long (7.5 km) and narrow (100 m) connection channel, the intense evaporation in the region increases salinity and as the water reaches the inner compartments of the lagoon, some 16 – 20 km from the ocean inlet, salinities can increase two-fold. A simple calculation of the evaporation based on the area of the lagoon and the hydric deficit (see Study Area) indicates a water loss of 6.6 m³ s^-1 during dryer years. The rise in salinity occurs only when the inflow of seawater is slow enough to enable evaporation. It is possible that slightly higher inflows of seawater may lead to a further increase in salinity, but a very high inflow of seawater should dilute hypersaline water of the lagoon. Considering that salinity levels in Araruama Lagoon are important for the maintenance of the ecosystem equilibrium, preventing the release of phosphate and avoiding algae blooms, the present research aimed to model salinity responses with a numerical hydrodynamic/transport model (SisBaHiA©) to identify the optimal size of the connection channel. The model results are expected to allow the identification of the best depth of the connection channel (hydraulic area) that promotes renewal of the lagoon water, but at the same time maintains the hypersalinity, preserving the characteristics of the ecosystem.

METHODS

Araruama Lagoon

The Araruama lagoon is located between the latitudes -22°49’ and -22°57’ and longitudes -42°00’ and -42°30’. This complex system spreads over nearly 200 km², 40 km east-west extension, a maximum width of 13 km, a perimeter of 190 km, and 618 million m³ water volume (Kjerfve & Oliveira, 2004) (Figure 1). The only connection with the ocean is a natural channel – Canal de Itajurú – located in the eastmost part of the lagoon. The meandering connection channel is 100 to 300 meters wide and 7.5 km long. After Turcq et al. (1999), the origin and evolution of the Araruama Lagoon was dated 120,000 years, during the eustatic rising level of the sea. Subsequent periods of raising and lowering sea level outlined the present bar that separates the lagoon from the ocean.

Kjerfve et al. (1996) classified the study area as a chocked lagoon (with a restricted connection), with an overall water renewal (flushing) time (T50%) of 83.5 days. Later, Rosman (2020) developed the concept of the water age, drawing from the idea of flushing time, and applied it to finite elements models. This approach allowed for the determination of the time at which water is renewed in every spot of the lagoon. The water age is the length of time water remains in a spot before it is exchanged. From this research it was possible to verify that the residence time of the water in the lagoon is shorter, varying between 24 and 52 days.

The regional climate is semiarid, resulting in a negative hydrological balance (Figure 2; Barbieri, 1984). The very small drainage basin (314.3 km²) and the climate conditions promote a perennial hypersalinity in the lagoon, regardless of the fact that the meteorological parameters (wind direction and velocity, precipitation, and evaporation) are seasonally marked (Garcia-Silva & Rosman, 2016). The wet period occurs between November and January and is marked by constant northeast winds. The dry period occurs between July and August and shows a predominance of southwest winds associated with cold fronts (Barbieri & Coe-Neto, 1999).

Figure 2
(A) Mean salinity in Araruama Lagoon (2010-2015) obtained from Prolagos and Águas de Juturnaíba Sanitation Companies; (B) precipitation-evaporation balance (2010-2015) from meteorological station M606.

Due to the extremely high salinity, Araruama Lagoon is under a precarious ecological balance and small disturbances can impact the water quality. The main affluent rivers are Moças, Mataruna, and Salgado, which produce an average overall flow rate of approximately 2.2 m³ s^-1 (19 x 104 m³ d^-1), the other affluents (small creeks) are intermittent and remain dry most of the year (Kjerfve & Oliveira, 2004). The freshwater inputs are significantly reduced, primarily due to the reduced size of the drainage basins, and to a lesser extent, because of the climate (Perrin, 1999).

In the ocean (Figure 1), at the channel mouth, the tide is semi-diurnal, with ranges of 0.80 m at neap tide and 1.30 m at spring tide (Lessa, 1991). In the connection channel, the tide is severely filtered by the meandering narrow path (Kjerfve & Knoppers, 1991), reaching the first large compartment of the lagoon (St3; see Figure 1), at about 8 km from the mouth with a range of not more than 10 cm (Carvalho, 2018). In the inner compartments of the lagoon, the direction, velocity of the currents, and water levels are related to meteorological events, mainly wind, with an average velocity of 8 m s^-1 (Kjerfve et al., 1996).

In the long-term, salinity presents relevant variations depending on the pluviosity (Moreira-Turcq, 2000). This behavior is exemplified in the years between 2010 and (to a lesser extent) 2015 (Figure 2). In 2010 precipitation exceeded evaporation and salinity reached values slightly higher than 40 PSU. In the following years, rainfall reduced and attained normal hydric deficits, promoting salinities of 55 PSU. Seasonal variability is small when compared to interannual oscillations.

The numerical model

SisBaHiA^© (Environmental Hydrodynamic Base System), is a multidimensional (1D, 2D and 3D) numeric model developed by Instituto Alberto Luiz Coimbra de Pós-Graduação e Pesquisa em Engenharia (COPPE) - UFRJ and registered by the Coppetec Foundation (http://www.sisbahia.coppe.ufrj.br/). It consists of several modules: hydrodynamic model, optimized for natural water bodies; eulerian and lagrangian transport models; models of water quality; and waves and tides models. SisBaHiA^© was widely used in Brazil to simulate physical and chemical processes in coastal ecosystems (e.g.: Andrade et al., 2019; Cunha et al., 2017). In the present study, we used the Hydrodynamic Module (HDM) coupled with the Advection-Diffusive Transport model (ADTM). Both modules were two-dimensional (2D) and vertically integrated. The 2D model was used because the lagoon is shallow and stratification is insignificant (Trevisan, 2023).

In SisBaHiA, the HDM resolves its variables by applying the method FIST (Filtered in Space and Time) with the equations of Navier-Stokes and an approximation for shallow water (a mean depth, H). Spatial discretization is done with a finite elements structure and the temporal discretization applies an implicit finite differences scheme (Rosman, 2012). A detailed description of the software can be obtained from SisBaHiA^© Technical Reference on the above-mentioned internet site (in Portuguese).

In hypersaline environments, such as the Araruama lagoon, the concentration of salt exerts a sensible influence on the hydrodynamic circulation (Andutta et al., 2011; Salamena et al., 2016). Thus, evaporation/precipitation must be considered, generating variations in salinities and density, that affect water circulation. In this work, salinity was simulated using the ADTM, and the Eulerian module coupled with the HDM.

HDM and ADTM operate in parallel (coupled), the first simulates the variation of the amount of movement (along the x- and y-axis) at every lapse of time (step of the model), afterward ADTM uses this information to calculate the advection and diffusion of salinity (Rosman, 2012). Then, it is possible to convert salt concentration into density with the equation of the state, as well as in a baroclinic model. SisBaHiA^© calculates the density according to the formulation proposed by Eckart (1958). The resulting salinity is shown in a 2D vertically integrated model that allows better interpretations.

Input data and model configuration

The input values related to geomorphology and freshwater inflows were spatially discretized over a grid of 930 finite elements (Supplementary Materials 1). The bottom roughness, bathymetry, and freshwater inflows of the lagoon were defined throughout the grid for the hydrodynamic model, considering the available data.

Although there are variations in the sediment granulometry, no extensive survey of this parameter has ever been done. In scarce works that collected sediment samples from the lagoon, usually, they are rough sandy (Silva et al., 2019). That is why Garcia-Silva & Rosman (2016) attributed a bottom roughness value of 0.03 m in all elements of the grid. The bathymetry obtained from a detailed survey performed in 2005 (main body of the lagoon; Wasserman et al., 2006) and 2016 (area of the connection channel with the sea) was spatially discretized over the grid (Figure 1 and Supplementary Material 1).

For elements at the river’s outlets, an open border boundary condition was imposed, which allows inflows and outflows, like in a tidal channel. The average flow values of Rio das Moças (1 m³ s^-1), Mataruna (0.7 m³ s^-1), and Salgado (0.5 m³ s^-1) were estimated by Emil Kuichling's Rational Method (1890), which is based on rainfall, drainage basin surface and land use. Empirical data of the region are limited and insufficient to construct a hydrograph.

A tidal forcing was applied in the mouth of the ocean connection channel with local tidal harmonics (Supplementary Material 2). The astronomical tidal constituents were obtained from an oceanographic station operated by the DHN (Diretoria de Hidrografia e Navegação, Brazilian Navy) located at Forno Port (Figure 1). Based on these tidal harmonics, the model automatically generated tidal curves for the mouth of the connection channel. Based on these data, a model simulation period: 01/01/2015 to 31/01/2015 (Supplementary Material 3) was set. This period was considered sufficient to encompass general changes in the salinity of the system, achieving the research objectives. Moreover, within a month, the system experienced the full range of astronomical tidal variations, including two spring and neap tides, resulting in both higher and lower volumes of seawater entering the system.

Meteorological data such as wind velocity and direction, precipitation, and evaporation were included in the HDM. Wind velocity, direction, and precipitation data were obtained from the meteorological stations M831 and M606 (Figure 1) operated by the INMET (Instituto Nacional de Meteorologia, 2024) and the evaporation data were estimated according to Linacre (1993), using data from station M606. Besides, values derived from Barbieri (1984) were used as presented in Table 1, summarizing the evaporation and precipitation values used in the HDM and ADTM modules. The period of simulation was the month of January, a worst-case scenario that was selected because it was the hottest, with the largest evaporation. The difference of salinity during this period between the lagoon and the ocean is the highest and it is expected that an increase in exchange rates will affect more significantly the lagoon.

Table 1
Evaporation and precipitation data used as meteorological forcing for the modules HDM and ADTM in the period 01/01/2015 to 31/01/2015.

Stations	Precipitation (mm month -1)	Evaporation (mm month -1)	Precipitation – Evaporation (mm month -1)
M831	136	140*	-4
M606	159	188**	-27

• ^*
  Values from Barbieri (1984);
• ^**
  Values estimated with temperature alone (Linacre, 1993).

Salinity data used for the ADTM were obtained from the sanitation companies PROLAGOS (2024) and Águas de Juturnaíba (2024). These companies perform daily measurements at St2, St3, St4, St5, St6 and St7 (Figure 1). The salinity data were spatially discretized over the grid to define the initial condition of the ADTM.

In the simulated period, real and modeled meteorological data (wind speed and direction, precipitation and evaporation), oceanographic data (harmonic constants and water level variation), and physical and chemical data (salinity) allowed to evaluate the model performances. two virtual monitoring stations were implemented in the model in the same locations as real monitoring sites (Figure 1).

The time step employed in the model was 30 seconds, resulting in an average courant number of 2.6, a value close to the optimum of 3.0 that avoids numerical instability (Rosman, 2012).

The HDM and ADTM models were executed simultaneously (coupled), starting at the moment 28,800^th second (8 hours), as the elevation of the water throughout the system was close to zero. To achieve this condition, the HDM was executed for a short period to generate values of elevation, and velocities in U and V that can be used in the model as an initial condition to avoid numerical errors and “turbulence” during the simulation of the scenarios. The initial condition for salinity was established through a gradient varying between 37 PSU in the ocean and 50 PSU in the center of the system and reducing to 40 PSU in the westernmost portion of the lagoon (due to freshwater inputs). This initial condition was applied in all scenarios. The salinities in the ocean open boundary were set at 37 PSU and 0 PSU in the affluent rivers (freshwater).

Scenarios setup

Six scenarios were defined, simulating the hydrodynamic circulation due to tides alone (A) and two different wind directions (B – NNE and C – SSW); salinity behavior in the typical climatic situation (D); and how the salinity would respond to bathymetric changes (3.0- and 4.0-meters depth) in the ocean connection channel (E and F). Table 2 summarizes the configuration of the scenarios developed for the present study.

Table 2
Configuration of five developed scenarios.

Scenarios	Forcing Functions					Bathymetry
	Wind	Precipitation	Evaporation	Freshwater flow	Tides *
A	NO	NO	NO	NO	YES	Unchanged
B	NNE (8 m s-1)	NO	NO	NO	NO	Unchanged
C	SSW (8 m s-1)	NO	NO	NO	NO	Unchanged
D	YES	YES	YES	YES	YES	Unchanged
E	YES	YES	YES	YES	YES	(-3.0m)
F	YES	YES	YES	YES	YES	(-4.0m)

• ^*
  Obtained from harmonic constants.

To achieve the objective of evaluating the effect of the channel depth on salinity, the original bathymetric values (scenario D) initially assigned to the model mesh nodes were manually altered in the Mesh & Model Domain section of SisBaHiA^© (Figure 3A). To change the original bathymetric values. A set of nodes was first selected in order to create a “channel/passage” aiming to increase the inflow/outflow of seawater. The nodes selected began at the entrance of the Canal de Itajurú and ended in the inner portion of the Araruama lagoon (Figure 3B). Second, the selected nodes had their bathymetric values changed to 3 meters (scenario E) and then to 4 meters (scenario F). In this “channel/passage” only the shallower values were changed (Figure 3C), like as if there were dredging operations to increase the hydraulic area along the connection.

Figure 3
Explanation on how the connection channel bathymetry was virtually altered to simulate scenarios D and E. (A) procedure in the software; (B) dark dots indicate elements that were modified; (C) Configuration of the table of values, after modifications.

Calibration/Comparison of modelled and measured salinity

SisBaHiA^© calibration modules were performed in three levels: (a) Geometric calibration; (b) Hydrodynamic calibration; and (c) Calibration of the scalar transport model (Supplementary Material 4). Only after checking all three levels, the coherence between the measured data and modeled data were evaluated according to the “Coherence Index of Measured and Computed Values” (CIM) (Equation 1) and the correlation coefficients (Rosman, 2012).

$CIM=\left(1-\frac{\sum \left|V_{obs}-V_{model}\right|}{\sum \left|V_{obs}\right|}\right)*100$ (1)

where CIM is the Coherence Index of Measured and Computed Values (%); V_obs is measured data; V_model is modeled data.

The measured values of salinity are from St4 and St5 (Supplementary Material 5) located inside the lagoon system (Figure 1). These stations were selected because they had daily salinity values (measured by refractometry) in the calibration period (January 2015). The modeled data were obtained from virtual monitoring stations implemented in the same region as the above-mentioned salinity measurement points (which were sampled in the margin of the lagoon).

A statiscal comparative evaluation of the measured and modeled data was carried out applying the following tests: basic statistics, skewness (as a measure of bias), root mean square error (RMSE), scatter index (SI - %) and correlation coefficient (r –Pearson’s).

RESULTS

Calibration

The results of the calibration showed that HDM and ADTM adequately reproduced real measurements since all requirements were met. The CIM results between measured and modeled data at St4 was 95% (Figure 4A) and at St5 was 99% (Figure 4B).

Figure 4
Calibration of the modeled (blue line) and measured (black line) salinities for 31 days in January 2015. (A) represents salinity variations in station St4; (B) represents salinity variations in station St5.

In addition, the salinity time series shown in Figure 4A and 4B indicates that measured values were more variable compared to the modeled values. It is usual for measured data to be more “turbulent” or “irregular” than modeled one since the former usually incorporates smaller scale forcing elements (e.g.: groundwater seepings) that cannot be adequately reproduced in the model, because measurements were done in the margin of the lagoon. Such an intense variation in measured salinity was unexpected, considering the dimension of the body of water that should rapidly buffer these variations.

Table 3 shows the basic and comparison statistics applied to both modeled and measured data. The results indicate that although the mean values are quite close, a larger variability in the salinities given by standard deviation, was observed in the measured data, corroborating the above-mentioned explanation for Figure 4A and 4B. On the other hand, skewness is higher for measured values, indicating a greater deviation from the normal behavior, which is natural from in situ measurements, compared to modeled. Cael (2021) showed that comparisons of data obtained from satellites and measured values indicate higher skewness in the measured values, which is attributed to small scale variations (spatial and temporal) that cannot be detected by remote sensors. RMSE and SI (%) follow this behavior showing relevant differences between both sets of data.

Table 3
Basic statistics and skewness of the measured and modeled salinities. Root mean standard error (RMSE), scatter index (SI (%)) and correlation coefficients compare each data series, measured and modeled.

	n	Mean	Minimum	Maximum	Std.Dev.	Skewness	RMSE	SI (%)	Correlation
Meas St4	31	47.323	40.000	50.000	2.625	-1.192	19.180	40.529	-0.3786
Mod St4	31	49.472	48.930	49.997	0.338	0.034
Meas St5	31	50.355	43.000	55.000	3.136	-0.982	16.419	32.607	0.3153
Mod St5	31	49.850	47.564	51.950	1.276	-0.129

Although correlation coefficient between measured and modeled data from station St4 is significant (p<0.05), the fact it is negative is relevant and cannot be accepted. The problem of this non-parametric analysis is that the range of variation is quite small (which is not considered by the correlation test) and the errors expected between both measured and modeled are random. The graph in Figure 5 (St5) shows how this random error behavior may produce a spread distribution, but within a small range of values.

Figure 5
Distribution graph showing measured and modeled data from station St5. Although the correlation coefficient was not significant (p<0.05), the range of the measured and modeled samples was quite narrow.

Hydrodynamics

In order to study the water circulation of the Araruama Lagoon, the HDM was first configured with the ocean semidiurnal tidal oscillation alone (scenario A) and with two different wind directions (scenarios B - NNE and C - SSW). The HDM simulated 36 hours to encompass more than one tidal cycle.

In scenario A, tidal range and current velocity decreased drastically after station St2 (Figures 6 and 7A). The averaged tidal range decreased from 0.70 m ± 0.10 at St2 to 0.07 m ± 0.01 at St3. Then, the tidal range decreased to average values close to zero, 0.01 m ± 0.001, at St4.

Figure 6
Free surface elevation results at stations St2, St3 and St4 from scenario A.

Figure 7
Velocity fields for different simulation scenarios. (A) tide only; (B) 8 m s^-1, wind from the NNE; (C) 8 m s^-1 SSW wind. Time: 115,200 seconds.

Numerous circulation cells were formed all over the lagoon in response to wind, which is the most noticeable result produced in scenarios B and C (Figure 7B and C). With a simulated 8 m s^-1 wind from the North-North-East (scenario B), it was possible to observe the formation of cyclonic (south margin) and anticyclonic (north margin) cells (Figure 7B). These circulation cells are produced by wind and currents reached a velocity of up to 0.11 m s^-1.

With the same 8 m s^-1 wind, but blowing from South-South-West (scenario C), the direction of circulation was inverted (Figure 7C). In scenarios B and C (with both wind directions), it was possible to observe the formation of cyclonic circulation in the northern margin and anticyclonic circulation in the southern margin of the lagoon respectively. Once again, the forcing wind produced circulation reaching velocities of up to 0.11 m s^-1 near the northern margin of the lagoon. It is noteworthy that this type of circulation does not favor vertical stratification because the water pulled by wind can easily return in a horizontal circulation movement. This should explain why Trevisan (2023) could not find any significant stratification along the lagoon.

Salinity responses to bathymetric changes in the channel

Scenarios D, E, and F (described in Table 2) were conceived to depict salinity modifications under a typical climatic situation and modified hydraulic area (channel depths of 3.0 and 4.0 meters). These three scenarios were simulated in the same period of the A, B, C, and calibration (01/01/2015 until 01/31/2015), with the coupled HDM/ADTM, under the same climatic conditions, freshwater flows, and tides, but with changes in the hydraulic area (bathymetry) of the connection channel.

In scenario D (unchanged bathymetry), there were significant spatial variations in salinities with a maximum value of 50.2 PSU at St5 and a minimum of 39.3 PSU at St3 (closer to the channel). After thirty-one simulation days, the western portion of the lagoon, near the riverine inputs, showed salinities of 50.1 and 45.3 PSU at St6 and St7, respectively (Figure 8), showing that Moças River alone can impress salinity reduction on station St7. Furthermore, station St7 also showed the largest standard deviation (±1.75; Figure 8), which indicates that rainfall has a larger effect on its salinity. Stations St2 and St3 (located closer to the ocean) presented lower salinities but also a broad variation (given by a standard deviation up to ±1.1) that can be attributed to the mixing with seawater. It is interesting to note that, regardless of the fact that water level variation with the tide at St4 is smaller than 5 cm (even in spring tide; Figure 5), mixing still reduces its salinities. At stations St5 and St6 the average salinity values are higher, and the variations are smaller when compared to the other stations (Figure 8). High salinities observed in these two stations were attributed to the fact that marine waters (with lower salinities) penetrate through the channel and move slowly to the inner portions, slowly evaporating to become hypersaline, generating a controversial behavior, i.e., regardless of the fact that hypersalinity is fed by marine waters, higher volumes of seawater into the lagoon would dilute hypersalinity.

Figure 8
Average and standard deviation (SD) for the six virtual monitoring stations implemented in HDM coupled models with ADTM, for simulated scenarios C, D, and E (01/01/2015 to 01/31/2015).

Looking closely at the results of scenario D, three main points stand out. First, although the tide promotes minor oscillation of the water level in the lagoon, the salt that feeds the hypersalinity of the lagoon comes from seawater. Second, this process promotes small variations of salinity in the center of the lagoon, probably due to the constant evaporation and the absence of freshwater inputs. Third, the presence of tributaries in the farther western portion (that were simulated in the climatic scenario) regulates salinity variations and causes continuous desalination of the hypersaline water that is formed in the larger central portion of the lagoon (St7, Figure 9A).

Figure 9
Modeled salinity results in stations St2, S4, and St7, obtained in scenarios D (A), E (B), and F (C).

In scenario E (channel with 3 meters depth), the variations of salinity among the three compartments significantly changed. Overall, there was an increase in the mean value of salinity in all stations. The most noticeable response to the deepening of the connection channel was the salinity increase at St7, which varied from 45.3 PSU to 50.0 PSU (Figure 9A). Moreover, the salinity at St7 in scenario E was far more constant than in scenario D, which should be explained by the higher salinity, demanding larger amounts of freshwater to promote oscillations in this parameter. On the other hand, more intense inflow/outflow of marine water in scenario D promoted larger variability in stations 3 and 4 (Figure 9A). In scenario E, the freshwater inputs do not seem to be able to cause desalination in any station (including St7) as observed in scenario C (Figure 9B).

In scenario F (Figure 9C), results indicate a salinity decrease in the whole system when compared to scenarios D and E (Figures 9B and 9C). In this scenario, the inner stations (St5, St6, and St7) presented a gradual decrease in the salinity while those located in the eastern compartment presented salinities close to the average values observed in scenario D.

DISCUSSION

The tides were severely filtered by the channel (Figures 6 and 7A). Our results corroborate Lessa (1991), who observed a tidal prism of 3.71 x 10⁷ m³, which was largely confined to the connection channel. This behavior is prevalent in choked lagoons, where the connection channel serves as a dynamic filter, responsible for eliminating oscillations of tidal currents and fluctuations (Kjerfve & Magill, 1989). Moreover, the current velocities forced by the tides alone were practically null after the connection channel (Figure 7A), beyond station St3. The maximum simulated current velocity of 0.12 m s^-1 occurred within the channel, associated with the tides, but velocities of 0.11 m s^-1 were identified in the lagoon which were exclusively forced by the winds.

The results of scenarios B and C (with winds NNE and SSW) shown in Figures 7B and 7C agreed with the observations of Kjerfve & Oliveira (2004). These authors applied a coupled two-dimensional circulation-dispersion tidal simulation model that showed that water level variations, direction, and velocity of the currents are rather due to meteorological events (mainly winds). Moreover, even using a different numerical model, these authors’ results also identified the formation of horizontal vorticed circulation patterns.

Based on the results of the present study, and observations from Kjerfve & Oliveira (2004) it was possible to suggest two hydrodynamic compartments in Araruama lagoon. The first was dominated by wind action (stations 4 – 7; Figure 1), which was not affected by tides, because of the strong filtration in the connection channel. The second was controlled by tidal action (stations 2 and 3; Figure 1) but did not receive any influence from the strong winds in the region, because the narrow morphology hinders the formation of a fetch.

The results of scenario D (Figure 9A) agreed with the observations made by Kjerfve et al. (1996) and Moreira-Turcq (2000). According to these authors, the high salinity variability in the connection channel of the Araruama Lagoon is influenced by the mixing of hypersaline water from the interior and the adjacent marine waters. Besides, the salinity in the central portion was constant and their oscillation in the farther western portion was regulated by river discharges (mainly in the last compartment). This behavior is distinct from what was observed in the Mar Menor Lagoon (Spain) where García-Oliva et al. (2019) identified a considerable increase in salinity range when dredging of the connection channels were simulated in a 3D hydrodynamic model. The difference between both situations is that in Mar Menor Lagoon the channels are considerably shorter than in Araruama, promoting immediate changes in the salinity. These authors developed a concept they call openness of the channel, that was well applied in their system but would have to be modified in Araruama, due to the length of the channel.

With the aim of confirming the compartmentation of the lagoon, the results of scenario D were evaluated in a mode-Q cluster analysis (Ward's method and City-block, Manhattan distances; Figure 10). The results showed three distinct groups, the first was formed by St2 and St3 influenced by the connection channel, the second was formed by St4, St5 and St6. The third was formed by St7 which was somehow isolated from the previous stations, but presented lower salinity (closer to stations St2 and St3)

Figure 10
Cluster Analysis for stations St2, St3, St4, St5, St6 e St7 with data from scenario D.

Considering the results of scenarios D, E, and F (Figure 9), it was possible to observe that due to the deepening of the connection channel, the tide promoted changes beyond the eastern compartment (Figure 11). The deepening of the connection channel down to 3.0 meters (Figures 9B and 11B) would increase the salinity benefiting mainly the western region since the exchange of water between the ocean and the lagoon was not large enough to dilute salinity and the flux of freshwater entering the lagoon was low enough to let evaporation promote increases in salinity. The deepening of the connection channel down to 4.0 meters (Figures 9C and 11C) would cause a general reduction of salinity throughout the system because the seawater inflow was so intense that dilution of the salinity overcame, evaporation.

Figure 11
Salinity distribution within the Lagoon for scenario D (A), scenario E (B), and scenario F (C).

In the simulations presented in this research, it was observed that there was a breakthrough point (at 4.0 meters; scenario F), beyond which salinity reduces due to dilution with seawater. With shallower depths, the channel exchange was not intense enough to reduce salinity, and ocean waters entering the lagoon still promoted an increase in salinity. With further dredging of the channel, or with the opening of multiple connections with the sea (as suggested by Garcia-Silva & Rosman, 2016), it is probable that the salinity would further reduce, attaining values close to the seawater. The benthonic primary production that yields transparent waters may shift to opaque waters with phytoplanktonic production during periods of low salinity (Moreira-Turcq, 2000).

Hypersaline environments exert strong pressure on the organisms that must develop expensive osmotic regulations (Breaux et al., 2019). Therefore, adapted organisms in this type of environment tend to be extremely sensitive, bearing a reduced resilience to variations, particularly nutrient inputs (Asencio, 2013). Therefore, long-term modifications of salinity may promote unbalance of the ecosystem.

CONCLUSION

Regardless of the difficulties in obtaining a precise calibration of the model, due to small scale forcing which cannot be adequately introduced in the model, the identified differences were sufficient small to give an idea of the circulation and to determine the effects of the modifications in the channel depth.

The hydrodynamic patterns depicted in the present research showed that Araruama Lagoon can be divided into two main compartments. The internal compartment where hydrodynamics displayed circulation cells that were controlled by wind direction. The second compartment was restricted to the connection channel, where tides were considerably more important than wind.

Salinity simulations showed that with the present water exchange rates, Araruama Lagoon can be compartmentalized into three sectors. The eastern compartment showed salinity variations forced by flood and ebb tides; the central compartment, with higher salinities and little oscillation whether due to tide or freshwater inputs. In the western compartment, no influence of the connection channel could be identified, but salinity was influenced by freshwater inputs.

The simulation of the depth alteration in the connection channel (to 3.0 meters) caused a generalized increase in salinity throughout the compartments of the lagoon system. This was due to the fact that the increase of flow (circa 432,000 m³ per tidal cycle) of marine waters was not sufficient to dilute the lagoon water and evaporation was able to promote a general increase in salinity. On the other hand, the second depth alteration of the connection channel (to depths of 4.0 meters) caused a general salinity reduction in all compartments of the lagoon system. This was because a threshold point was attained (above 3.0 meters) where the exchange of marine waters was so important that dilution of the hypersaline water overwhelmed its evaporation, promoting the reduction of salinity.

Finally, it is essential to underline that the application of numerical models to investigate such important parameters as hydrodynamics and salinity in a hypersaline ecosystem can be a useful tool to assist the management and prediction of impacts generated by human and natural interventions, thus making decision-makers more assertive.

Supplementary Material

Supplementary material accompanies this paper.

Supplementary Material 1 –

Map showing the applied grid of Araruama Lagoon for the construction of the hydrodynamic model in SisBaHiA.

Supplementary Material 2 –

Tidal constants, period, amplitude and phase inserted in the oceanic frontier of the hydrodynamic module. Data from DHN, Forno Port, Arraial do Cabo – RJ.

Supplementary Material 3–

Tide at the open edge for the interval analyses of 01/01/2015 a 31/01/2015. The medium level (0.0 m) is the NMM at oceanographic station located in Forno Port.

Supplementary Material 4 –

Calibration levels, items for assess, requirements that must be considered for each calibration level and if those requirements were complied with.

Supplementary Material 5 –

Raw data used for calibration, obtained in stations 4 and 5.

This material is available as part of the online article from https://doi.org/10.1590/2318-0331.292420240078

ACKNOWLEDGEMENTS

JCW is grateful to the Brazilian Council of Scientific and Technological Development (CNPq) for a research grant (grant # 305374/2023-0). ACBC is grateful to the Brazilian Council of Scientific and Technological Development (CNPq) for a M.Sc. Scholarship and MVC is grateful to CAPES for a M.Sc. Scholarship. The authors are also grateful to CAPES for financial support (grant # 001).

REFERENCES

• Águas de Juturnaíba. (2024). Retrieved in 2024, August 6, from https://www.grupoaguasdobrasil.com.br/aguas-juturnaiba/
  » https://www.grupoaguasdobrasil.com.br/aguas-juturnaiba/
• Andrade, V. S., Rosman, P. C. C., & Azevedo, J. P. S. (2019). Evaluation of baroclinic effects on mean water levels in Guanabara Bay. Revista Brasileira de Recursos Hídricos, 24, e48. http://doi.org/10.1590/2318-0331.241920180112
  » http://doi.org/10.1590/2318-0331.241920180112
• Andutta, F. P., Ridd, P. V., & Wolanski, E. (2011). Dynamics of hypersaline coastal waters in the Great Barrier Reef. Estuarine, Coastal and Shelf Science, 94(4), 299-305. http://doi.org/10.1016/j.ecss.2011.06.009
  » http://doi.org/10.1016/j.ecss.2011.06.009
• Asencio, A. D. (2013). Permanent salt evaporation ponds in a semi-arid Mediterranean region as model systems to study primary production processes under hypersaline conditions. Estuarine, Coastal and Shelf Science, 124, 24-33. http://doi.org/10.1016/j.ecss.2013.03.006
  » http://doi.org/10.1016/j.ecss.2013.03.006
• Barbieri, E. (1984). Cabo Frio e Iguaba Grande, dois microclimas distintos a um curto intervalo espacial. In L. D. Lacerda, D. S. D. Araújo, R. Cerqueira, & B. Turcq (Eds.), Restingas: origem, estrutura, processos (pp. 3-13). Niterói: CEUFF.
• Barbieri, E., & Coe-Neto, R. (1999). Spatial and temporal variation of rainfall of the East Fluminense Coast and Atlantic Serra do Mar, State of Rio de Janeiro, Brazil. In B. A. Knoppers, E. D. Bidone, & J. J. Abrão (Eds.), Environmental Geochemistry of Coastal Lagoon Systems of Rio de Janeiro, Brazil (pp. 47-56). Rio de Janeiro: UFF/FINEP.
• Bedri, Z., O’Sullivan, J. J., Deering, L. A., Demeter, K., Masterson, B., Meijer, W. G., & O’Hare, G. (2015). Assessing the water quality response to an alternative sewage disposal strategy at bathing sites on the east coast of Ireland. Marine Pollution Bulletin, 91(1), 330-346. http://doi.org/10.1016/j.marpolbul.2014.11.008
  » http://doi.org/10.1016/j.marpolbul.2014.11.008
• Bertucci, T. C. P., Silva, E. P., Marques Júnior, A. N., & Monteiro Neto, C. (2016). Tourism and urbanization: environmental problems of the Araruama Lagoon, State of Rio de Janeiro, Brazil. Ambiente & Sociedade, 19(4), 59-80. http://doi.org/10.1590/1809-4422asoc137111v1942016
  » http://doi.org/10.1590/1809-4422asoc137111v1942016
• Breaux, N., Lebreton, B., Palmer, T. A., Guillou, G., & Pollack, J. B. (2019). Ecosystem resilience following salinity change in a hypersaline estuary. Estuarine, Coastal and Shelf Science, 225(12), 106258. http://doi.org/10.1016/j.ecss.2019.106258
  » http://doi.org/10.1016/j.ecss.2019.106258
• Bruun, P. (1994). Engineering projects in Coastal Lagoons. Elsevier Oceanography Series, 60, 507-533. http://doi.org/10.1016/S0422-9894(08)70022-9
  » http://doi.org/10.1016/S0422-9894(08)70022-9
• Burchard, H., & Hofmeister, R. (2008). A dynamic equation for the potential energy anomaly for analysing mixing and stratification in estuaries and coastal seas. Estuarine, Coastal and Shelf Science, 77(4), 679-687. http://doi.org/10.1016/j.ecss.2007.10.025
  » http://doi.org/10.1016/j.ecss.2007.10.025
• Cael, B. B. (2021). Variability-based constraint on ocean primary production models. Limnology and Oceanography Letters, 6(5), 262-269. http://doi.org/10.1002/lol2.10196
  » http://doi.org/10.1002/lol2.10196
• Carvalho, A. C. B. (2018). Trocas de sal entre a laguna de Araruama e o mar, utilizando um modelo hidrodinâmico (SiSBaHiA) (Salt exchange betwwen Araruama Lagoon and the sea, applying a hydrodynamic model - SiSBaHiA) (Dissertação de mestrado). Universidade Federal Flumiense, Niterói, RJ.
• Chen, Y., Zou, R., Han, S., Bai, S., Faizullabhoy, M., Wu, Y. Y., & Guo, H. C. (2017). Development of an integrated water quality and macroalgae simulation model for tidal marsh eutrophication control decision support. Water (Basel), 9(4), 277. http://doi.org/10.3390/w9040277
  » http://doi.org/10.3390/w9040277
• Colaço, G., Violante-Carvalho, N., Arruda, W. Z., D’Avila, V., Vilhena, H., Filippo, A., & Candella, R. (2021). Mid-long term characterisation and short term modelling of a semi-protected pocket sandy beach in the Southern Coast of Brazil. Regional Studies in Marine Science, 41, 101593. http://doi.org/10.1016/j.rsma.2020.101593
  » http://doi.org/10.1016/j.rsma.2020.101593
• Cotner, J. B., Suplee, M. W., Chen, N. W., & Shormann, D. E. (2004). Nutrient, sulfur and carbon dynamics in a hypersaline lagoon. Estuarine, Coastal and Shelf Science, 59(4), 639-652. http://doi.org/10.1016/j.ecss.2003.11.008
  » http://doi.org/10.1016/j.ecss.2003.11.008
• Cunha, A. C., Mustin, K., Santos, E. S., Santos, E. W. G., Guedes, M. C., Cunha, H. F. A., Cunha, H. F. A., Rosman, P. C. C., & Sternberg, L. S. L. (2017). Hydrodynamics and seed dispersal in the lower Amazon. Freshwater Biology, 62(10), 1721-1729. http://doi.org/10.1111/fwb.12982
  » http://doi.org/10.1111/fwb.12982
• Cutroneo, L., Castellano, M., Pieracci, A., Povero, P., Tucci, S., & Capello, M. (2012). The use of a combined monitoring system for following a turbid plume generated by dredging activities in a port. Journal of Soils and Sediments, 12(5), 797-809. http://doi.org/10.1007/s11368-012-0486-0
  » http://doi.org/10.1007/s11368-012-0486-0
• D’Alpaos, A., Lanzoni, S., Marani, M., & Rinaldo, A. (2010). On the tidal prism-channel area relations. Journal of Geophysical Research, 115(F1), 2008JF001243. http://doi.org/10.1029/2008JF001243
  » http://doi.org/10.1029/2008JF001243
• Eckart, C. H. (1958). Properties of water: II. The equation of state of water and sea water at low temperatures and pressures. American Journal of Science, 256(4), 225-240. http://doi.org/10.2475/ajs.256.4.225
  » http://doi.org/10.2475/ajs.256.4.225
• Ferrarin, C., & Umgiesser, G. (2005). Hydrodynamic modeling of a coastal lagoon: the Cabras lagoon in Sardinia, Italy. Ecological Modelling, 188(2-4), 340-357. http://doi.org/10.1016/j.ecolmodel.2005.01.061
  » http://doi.org/10.1016/j.ecolmodel.2005.01.061
• García-Oliva, M., Marcos, C., Umgiesser, G., McKiver, W., Ghezzo, M., De Pascalis, F., & Perez-Ruzafa, A. (2019). Modelling the impact of dredging inlets on the salinity and temperature regimes in coastal lagoons. Ocean and Coastal Management, 180(33), 104913. http://doi.org/10.1016/j.ocecoaman.2019.104913
  » http://doi.org/10.1016/j.ocecoaman.2019.104913
• Garcia-Silva, R. A., & Rosman, P. C. C. (2016). Hydro-sedimentologic viability of a tidal channel project in the ouest of Araruama Lagoon - RJ. Revista Brasileira de Recursos Hídricos, 21(1), 25-35. http://doi.org/10.21168/rbrh.v21n1.p25-35
  » http://doi.org/10.21168/rbrh.v21n1.p25-35
• Guimarães, T. C. S. M., Montenegro, K. S., Wasserman, M. A. V., & Wasserman, J. C. (2021). Innovative microcosm experiments for the evaluation of the regeneration rates of nutrients in sediments of a hypersaline lagoon. Marine Pollution Bulletin, 166, 112252. http://doi.org/10.1016/j.marpolbul.2021.112252
  » http://doi.org/10.1016/j.marpolbul.2021.112252
• Instituto Nacional de Meteorologia – INMET. (2024). Retrieved in 2024, August 6, from https://tempo.inmet.gov.br/TabelaEstacoes
  » https://tempo.inmet.gov.br/TabelaEstacoes
• Jung, H. B., Richards, J., & Fitzgerald, A. (2021). Temporal and spatial variations of water quality in the Newark Bay Estuary. Regional Studies in Marine Science, 41(12), 101589. http://doi.org/10.1016/j.rsma.2020.101589
  » http://doi.org/10.1016/j.rsma.2020.101589
• Kjerfve, B., & Knoppers, B. (1991). Tidal choking in a coastal lagoon. In B. B. Parker (Ed.), Tidal hydrodynamics (pp. 169-179). New York: John Wiley & Sons.
• Kjerfve, B., & Magill, K. E. (1989). Geographic and hydrodynamic characteristics of shallow coastal lagoons. Marine Geology, 88(3-4), 187-199. http://doi.org/10.1016/0025-3227(89)90097-2
  » http://doi.org/10.1016/0025-3227(89)90097-2
• Kjerfve, B., & Oliveira, A. M. (2004). Modelling of Circulation and Water Exchange in a Hypersaline Coastal Lagoon: Lagoa de Araruama, Brazil. In L. Drude de Lacerda, R. E. Santelli, E. K. Duursma, & J. J. Abrão (Eds.), Environmental geochemistry in tropical and subtropical environments (pp. 235-251). Berlin: Springer. http://doi.org/10.1007/978-3-662-07060-4_18
  » http://doi.org/10.1007/978-3-662-07060-4_18
• Kjerfve, B., Schettini, C. A. F., Knoppers, B., Lessa, G., & Ferreira, H. O. (1996). Hydrology and salt balance in a large, hypersaline coastal lagoon: lagoa de Araruama, Brazil. Estuarine, Coastal and Shelf Science, 42(6), 701-725. http://doi.org/10.1006/ecss.1996.0045
  » http://doi.org/10.1006/ecss.1996.0045
• Lessa, G. C. (1991). Dinâmica de maré e transporte de sedimentos no canal de Itajuru - Laguna de Araruama (RJ). Revista Brasileira de Geociencias, 21(4), 378-386. http://doi.org/10.25249/0375-7536.1991378386
  » http://doi.org/10.25249/0375-7536.1991378386
• Linacre, E. T. (1993). Data-sparse estimation of lake evaporation, using a simplified Penman equation. Agricultural and Forest Meteorology, 64(3-4), 237-256. http://doi.org/10.1016/0168-1923(93)90031-C
  » http://doi.org/10.1016/0168-1923(93)90031-C
• Meng, P.-J., Tew, K. S., Hsieh, H.-Y., & Chen, C.-C. (2017). Relationship between magnitude of phytoplankton blooms and rainfall in a hyper-eutrophic lagoon: a continuous monitoring approach. Marine Pollution Bulletin, 124(2), 897-902. http://doi.org/10.1016/j.marpolbul.2016.12.040
  » http://doi.org/10.1016/j.marpolbul.2016.12.040
• Moreira-Turcq, P. F. (2000). Impact of a low salinity year on the metabolism of a hypersaline coastal lagoon (Brazil). Hydrobiologia, 429(1-3), 133-140. http://doi.org/10.1023/A:1004037624787
  » http://doi.org/10.1023/A:1004037624787
• Oliveira, M. M., Ventura, A. M. F., Fraga, A. S., Neves, M., Dantas, R. F., Hauser-Davis, R. A., Salles, J. B., Bastos, F., Bastos, V. C., & Bastos, J. (2020). Mugil liza (Valenciennes, 1836) biomarker responses in a hypersaline lagoon in southeastern Brazil before and after a microalgal bloom. International Aquatic Research., 12(4), 231-241. http://dx.doi.org/10.22034/iar.2020.1896881.1030.
• Perrin, P. (1999). Physiography and surface formations of the East Fluminense Coast, State of Rio de Janeiro, Brazil. In B. A. Knoppers, E. D. Bidone, & J. J. Abrão (Eds.), Environmental geochemistry of coastal lagoon systems of Rio de Janeiro, Brazil (Vol. 6, pp. 11-23). Rio de Janeiro: UFF/FINEP.
• PROLAGOS. (2024). Retrieved in 2024, August 6, from http://www.prolagos.com.br/
  » http://www.prolagos.com.br/
• Reimer, J. J., & Huerta-Diaz, M. A. (2011). Phosphorus speciation and sedimentary fluxes in hypersaline sediments of the Guerrero Negro Salt Evaporation Area, Baja California Sur, Mexico. Estuaries and Coasts, 34(3), 514-528. http://doi.org/10.1007/s12237-010-9308-z
  » http://doi.org/10.1007/s12237-010-9308-z
• Rodrigues, A. M., Quintino, V., Pereira, F., & Freitas, R. (2012). Alterations in macroinvertebrate spatial patterns in coastal lagoons: Óbidos (NW coast of Portugal) 1984 versus 2002. Estuarine, Coastal and Shelf Science, 110, 176-189. http://doi.org/10.1016/j.ecss.2012.05.024
  » http://doi.org/10.1016/j.ecss.2012.05.024
• Rosman, P. C. C. (2012). Referências técnicas do SisBaHiA®. Retrieved in 2024, August 6, from https://www.sisbahia.coppe.ufrj.br/assets/downloads/SisBaHiA_RefTec_v11c.pdf
  » https://www.sisbahia.coppe.ufrj.br/assets/downloads/SisBaHiA_RefTec_v11c.pdf
• Rosman, P. C. C. (2020). Estudos de hidrodinâmica ambiental para suporte e planejamento de intervenções na Lagoa de Araruama - RJ, em prazos curto e médio Retrieved in 2024, August 6, from https://www.sisbahia.coppe.ufrj.br/
  » https://www.sisbahia.coppe.ufrj.br/
• Rynne, P., Reniers, A., van de Kreeke, J., & MacMahan, J. (2016). The effect of tidal exchange on residence time in a coastal embayment. Estuarine, Coastal and Shelf Science, 172, 108-120. http://doi.org/10.1016/j.ecss.2016.02.001
  » http://doi.org/10.1016/j.ecss.2016.02.001
• Salamena, G. G., Martins, F., & Ridd, P. V. (2016). The density-driven circulation of the coastal hypersaline system of the Great Barrier Reef, Australia. Marine Pollution Bulletin, 105(1), 277-285. http://doi.org/10.1016/j.marpolbul.2016.02.015
  » http://doi.org/10.1016/j.marpolbul.2016.02.015
• Silva, K. M. d. (2019). Geochemistry of phosphorus in the sediments of Araruama Lagoon, RJ for an assessment of eutrophication and its impact on environmental health (Dissertação de mestrado). Fundação Oswaldo Cruz, Rio de Janeiro.
• Silva, K. M., Guimarães, T. C., & Wasserman, J. C. (2019). Distribuição espacial da concentração de nutrientes em sedimentos na laguna hipersalina de Araruama – RJ, Brasil. Geochimica Brasiliensis, 33(1), 16-27. http://doi.org/10.21715/GB2358-2812.2019331016
  » http://doi.org/10.21715/GB2358-2812.2019331016
• Souza, F. P., & Azevedo, J. P. S. (2020). An overview of urban lagoons in Rio de Janeiro: relevant aspects in the management of Rodrigo de Freitas and Araruama Lagoons and the Jacarepagua lagoon system. Engenharia Sanitaria e Ambiental, 25(1), 197-204. http://doi.org/10.1590/s1413-41522020147571
  » http://doi.org/10.1590/s1413-41522020147571
• Souza, M. F. L., Kjerfve, B., Knoppers, B., Souza, W. F. L., & Damasceno, R. N. (2003). Nutrient budgets and trophic state in a hypersaline coastal lagoon: lagoa de Araruama, Brazil. Estuarine, Coastal and Shelf Science, 57(5-6), 843-858. http://doi.org/10.1016/S0272-7714(02)00415-8
  » http://doi.org/10.1016/S0272-7714(02)00415-8
• Sylvestre, F., Beck-Eichler, B., Duleba, W., & Debenay, J. P. (2001). Modern benthic diatom distribution in a hypersaline coastal lagoon: the Lagoa de Araruama (RJ), Brazil. Hydrobiologia, 443(1-3), 213-231. http://doi.org/10.1023/A:1017558914971
  » http://doi.org/10.1023/A:1017558914971
• Trevisan, C. L. (2023). Biogeochemical processes in a tropical hypersaline lagoon: an approach to dystrophy (Doctoral dissertation) University Federal Fluminense, Niterói, RJ.
• Turcq, B., Martin, L., Flexor, J. M., Suguio, K., Pierre, C., & Tasayco-Ortega, L. (1999). Origin and evolution of the Quaternary coastal plain between Guaratiba and Cabo Frio, State of Rio de Janeiro, Brazil. In B. A. Knoppers, E. D. Bidone, & J. J. Abrão (Eds.), Environmental geochemistry of Coastal Lagoon Systems of Rio de Janeiro, Brazil (Vol. 6, pp. 25-46). Rio de Janeiro: UFF/FINEP.
• Valle-Levinson, A. (2008). Density-driven exchange flow in terms of the Kelvin and Ekman numbers. Journal of Geophysical Research, 113(C4), 2007JC004144. http://doi.org/10.1029/2007JC004144
  » http://doi.org/10.1029/2007JC004144
• van Maren, D. S., van Kessel, T., Cronin, K., & Sittoni, L. (2015). The impact of channel deepening and dredging on estuarine sediment concentration. Continental Shelf Research, 95, 1-14. http://doi.org/10.1016/j.csr.2014.12.010
  » http://doi.org/10.1016/j.csr.2014.12.010
• van Rijn, L. C. (2005). Estuarine and coastal sedimentation problems. International Journal of Sediment Research, 20(1), 39-51.
• Venier, C., D’Alpaos, A., & Marani, M. (2014). Evaluation of sediment properties using wind and turbidity observations in the shallow tidal areas of the Venice Lagoon. Journal of Geophysical Research. Earth Surface, 119(7), 1604-1616. http://doi.org/10.1002/2013JF003019
  » http://doi.org/10.1002/2013JF003019
• Vicente, M. C., Carvalho, A. C. B., Trevisan, C. L., Soares, F. F. L., & Wasserman, J. C. (2021). Spatial-temporal distribution of dissolved inorganic nutrients in the hypersaline Araruama Lagoon, Rio de Janeiro, Brazil. Regional Studies in Marine Science, 44, 101787. http://doi.org/10.1016/j.rsma.2021.101787
  » http://doi.org/10.1016/j.rsma.2021.101787
• Wasserman, J. C., Alves, A. R., Costa, A. M. K. d., Retto, M. L. S. N., Pereira, L. F. M., & Barros, S. R.. (2006). Avaliação ambiental da laguna de Araruama, para subsídiar um programa de recuperação e gestão (Environmental evaluation of the Araruama Lagoon to subsidize a recovery program and management). Retrieved in 2024, August 6, from http://docs.wixstatic.com/ugd/87caa6_a66729ef1d1a465fa304c0fad5184b76.pdf
  » http://docs.wixstatic.com/ugd/87caa6_a66729ef1d1a465fa304c0fad5184b76.pdf
• Wasserman, J. C., Carvalho, Â. C., Silva, K. M. d., Vicente, M. C., & Guimarães, T. C. S. M. (2019a). A qualidade da água e as mortandades de peixes na laguna de Araruama. In M. I. P. Ferreira, G. S. Barreto, J. L. Junior, J. A. F. Silva, & M. P. F. Barros (Eds.), Engenharia e Ciências Ambientais: Contribuições à Gestão Ecossistêmica (pp. 222-244). São Paulo: Essentia Editora. http://doi.org/10.19180/978-85-99968-58-1.11
  » http://doi.org/10.19180/978-85-99968-58-1.11
• Wasserman, J. C., Carvalho, Â. C., Silva, K. M. d., Vicente, M. C., & Guimarães, T. C. S. M. (2019b). Water quality an fish mortalities in Araruama Lagoon. In M. I. P. Ferreira, G. S. Barreto, J. L. Junior, J. A. F. Silva, & M. P. F. Barros (Eds.), Engineering and environmental sciences: contributions to ecosystem management (pp. 222-244). São Paulo: Essentia Editora. http://doi.org/10.19180/978-85-99968-58-1.11
  » http://doi.org/10.19180/978-85-99968-58-1.11
• Webster, I. T. (2010). The hydrodynamics and salinity regime of a coastal lagoon – The Coorong, Australia – Seasonal to multi-decadal timescales. Estuarine, Coastal and Shelf Science, 90(4), 264-274. http://doi.org/10.1016/j.ecss.2010.09.007
  » http://doi.org/10.1016/j.ecss.2010.09.007
• Zacharias, I., & Gianni, A. (2008). Hydrodynamic and dispersion modeling as a tool for restoration of coastal ecosystems. Application to a re-flooded lagoon. Environmental Modelling & Software, 23(6), 751-767. http://doi.org/10.1016/j.envsoft.2007.09.007
  » http://doi.org/10.1016/j.envsoft.2007.09.007
• Zarzuelo, C., Díez-Minguito, M., Ortega-Sánchez, M., López-Ruiz, A., & Losada, M. Á. (2015). Hydrodynamics response to planned human interventions in a highly altered embayment: The example of the Bay of Cádiz (Spain). Estuarine, Coastal and Shelf Science, 167, 75-85. http://doi.org/10.1016/j.ecss.2015.07.010
  » http://doi.org/10.1016/j.ecss.2015.07.010
• Zhang, J. Z., Fischer, C. J., & Ortner, P. B. (2004). Potential availability of sedimentary phosphorus to sediment resuspension in Florida Bay. Global Biogeochemical Cycles, 18(4), 2004GB002255. http://doi.org/10.1029/2004GB002255
  » http://doi.org/10.1029/2004GB002255
• Zhang, M. L., Xu, Y. Y., & Qiao, H. T. (2018). Numerical study of hydrodynamic and solute transport with discontinuous flows in coastal water. Environmental Modeling and Assessment, 23(4), 353-367. http://doi.org/10.1007/s10666-017-9585-z
  » http://doi.org/10.1007/s10666-017-9585-z

Edited by

• Editor-in-Chief:
  Adilson Pinheiro

• Associated Editor:
  Iran Eduardo Lima Neto

Publication Dates

• Publication in this collection
  09 Dec 2024
• Date of issue
  2024

History

• Received
  06 Aug 2024
• Reviewed
  27 Sept 2024
• Accepted
  03 Oct 2024