1 23

Environmental Monitoring andAssessmentAn International Journal Devoted toProgress in the Use of Monitoring Datain Assessing Environmental Risks toMan and the Environment ISSN 0167-6369 Environ Monit AssessDOI 10.1007/s10661-013-3316-y

Biogeochemical responses to nutrientinputs in a Cuban coastal lagoon: runoff,anthropogenic, and groundwater sources

R. González-De Zayas, M. Merino-Ibarra, M. F. Soto-Jiménez &F. S. Castillo-Sandoval

1 23

Your article is protected by copyright and all

rights are held exclusively by Springer Science

+Business Media Dordrecht. This e-offprint

is for personal use only and shall not be self-

archived in electronic repositories. If you wish

to self-archive your article, please use the

accepted manuscript version for posting on

your own website. You may further deposit

the accepted manuscript version in any

repository, provided it is only made publicly

available 12 months after official publication

or later and provided acknowledgement is

given to the original source of publication

and a link is inserted to the published article

on Springer's website. The link must be

accompanied by the following text: "The final

publication is available at link.springer.com”.

Biogeochemical responses to nutrient inputs in a Cubancoastal lagoon: runoff, anthropogenic, and groundwatersources

R. González-De Zayas & M. Merino-Ibarra &

M. F. Soto-Jiménez & F. S. Castillo-Sandoval

Received: 19 September 2012 /Accepted: 25 June 2013# Springer Science+Business Media Dordrecht 2013

Abstract Laguna Larga, a coastal lagoon in central Cu-ba, has been heavily altered by tourism infrastructureconstruction and sewage disposal. We hypothesize thatthis has decreased the circulation and caused eutrophica-tion of the lagoon. To assess this, 12 bimonthly samplingswere carried out in 2007–2008. Temperature, salinity,oxygen, nutrients and nitrogen, and phosphorous fractions(inorganic, organic, and total) were determined.Water andsalt budgets, as well as biogeochemical fluxes of nitrogenand phosphorus were calculated using the LOICZ budgetmodel for the three sections of the lagoon identified by

morphological constrains and salinity patterns. LagunaLarga is a choked lagoon with restricted water circulation,low exchange, and high residence times that vary signif-icantly along its sections. Residence time was estimated tobe 0.1–0.7 years for the inner section and 1–9 days for theouter one. High levels of total nitrogen (annual means126–137 μM, peaks up to 475 μM) and phosphorus (2.5–4.4 μM, peaks up to 14.5 μM) are evidence of eutrophi-cation of Laguna Larga. During 2007, an average precip-itation year, Laguna Larga exported water (703 m3 d−1)and was a source of nitrogen (9.026 mmol m−2 d−1) andphosphorus (0.112 mmol m−2 d−1) to the adjacent sea.δ15N determinations in the seagrass Thalassia testudinum(−1.83 to +3.02‰) differed significantly between sitesin the lagoon and offshore reference sites located W ofthe inlet, but were similar to those located E of theinlet. δ15N determinations in the seaweed Penicillusdumetosus (+1.02 to +4.2) did not show significantdifferences.

Keywords Budget . Cayo Coco . Isotopes . LOICZ .

Management . Nutrients . Sewage

Introduction

The rapid growth of anthropogenic activities in water-sheds and in adjacent coastal areas has brought about asignificant increase of pollutants (organic matter, nutri-ents, hydrocarbons, and heavy metals) and direct impacts

Environ Monit AssessDOI 10.1007/s10661-013-3316-y

R. González-De ZayasPosgrado en Ciencias del Mar y Limnología, Instituto deCiencias del Mar y Limnología, Universidad NacionalAutónoma de México, Ciudad Universitaria,Coyoacán, Mexico, DF 04510, Mexico

R. González-De ZayasCentro de Investigaciones de Ecosistemas Costeros,Cayo Coco, Morón Ciego de Ávila 69400, Cuba

M. Merino-Ibarra (*) : F. S. Castillo-SandovalUnidad Académica de Ecología y Biodiversidad Acuática,Instituto de Ciencias del Mar y Limnología, UniversidadNacional Autónoma de México, Ciudad Universitaria,Coyoacán, Mexico, DF 04510, Mexicoe-mail: mmerino@cmarl.unam.mx

M. F. Soto-JiménezUnidad Académica Mazatlán, Instituto de Ciencias del Mary Limnología, Universidad Nacional Autónoma de México,Av. Joel Montes Camarena S/N A. Postal 811,82040 Mazatlan, Sinaloa, Mexico

Author's personal copy

on coastal ecosystems, particularly those with limitedwater circulation and carrying capacity (Lapointe andClark 1992, Lucena et al. 2002, Bricker et al. 2008).Accumulation of pollutants in coastal ecosystems de-grades environmental quality and depletes filtering capac-ity (Merino et al. 1992). Thus, knowledge of coastalecosystem functioning is required, particularly in tropicalareas, which have been much less studied than theircounterparts in temperate latitudes (Dale and Prego 2003).

One of the most important impacts on coastal marineecosystems is eutrophication, a process that results in theovergrowth of algae, harmful algal blooms, and oxygendepletion. Eutrophic systems can become instable andundergo dystrophic crisis (e.g., Lesina Lagoon in theAdriatic Sea during 2008, Vignes et al. 2009). Eutrophi-cation is the result of increasing inputs of nutrients,usually carbon, nitrogen, and phosphorus (Lawrenceet al. 1998), elements that normally regulate the trophicstatus in fresh and coastal water bodies. To understandeutrophication in coastal lagoons, it is necessary to knowin detail the processes involving these nutrients and toquantify their flows toward coastal systems and the ma-rine area (Gordon et al. 1996). Information on the routesand destinations of nutrients in tropical areas with oligo-trophic ecosystems such as coral reefs, which are verysensitive to nutrient inputs, is essential to assess theirenvironmental consequences (Leichter et al. 2003).

Various approaches have been used to estimate nutrientflow between coastal water bodies and the open ocean(e.g. Wulff and Stigebrandt 1989, Smith and Hollibaugh1997, Witek et al. 2003). However, the complexity of thedynamic physical, chemical, and biological processes in-volved complicates the comparison of observations fromdifferent systems. The program Land–Ocean Interactionsin the Coastal Zone (LOICZ) has proposed an approachbased on water and salt budgets applied to most coastalbodies, which allows the comparison of results (Gordonet al. 1996). In addition, a combination of geochemicaland isotopic techniques is useful to determine the origin ofthe nutrients, tracing them in the environment and study-ing their biogeochemistry. In particular, the stable isotopesof carbon and nitrogen have been frequently used todetermine the origin and flow of nutrient and organicmatter and the trophic structure of coral reef systems(Umezawa et al. 2002, Leichter et al. 2003).

In this study, we analyze the biogeochemical re-sponses of a tropical coastal lagoon (Laguna Larga, offnortheastern Cuba) to the nutrient inputs from runoff,groundwater, and sewage and how the connection

between these processes in the lagoon affects the adjacentmarine coast, characterized by seagrass and coral reefecosystems.

Materials and methods

Study site

Laguna Larga is a tropical (22o32′N, 78o22′W) coastallagoon with limited exchange (Fig. 1). Its natural com-munication with the sea is through a narrow channel(8–15 m wide). The channel is well-defined at the inlet,but it divides into multiple smaller channels inside thesystem. Communication between the three main sec-tions (eastern, central, and western) of the lagoon isalso restricted to a small channel completely coveredwith mangroves. Multiple patches of construction ma-terial left from hotels construction are found through-out the lagoon bottom. Sedimentation in the channels,and the development of mangroves over almost 60 %of the central section, also restrict the circulation of thelagoon waters. The bathymetry is irregular withvery shallow areas in the channels (0.3–0.5 m) andslightly less shallow areas in the widest parts of thewestern and eastern sections (at some sites >1 m due todredging). The tide in the region is regular semidiurnaland with relatively small amplitude. The highest meansea levels occur in the summer months and the lowestin winter, with a maximum difference of 30 cm be-tween the average values. The maximum amplituderecorded was 125 cm in September 1994. The prevail-ing ocean currents in the reef flow westward, withaverage speeds in the order of 10 cm/s. Near the coast,the predominant flow is from the NW, with an averageof 2 cm/s.

Red mangrove (Rhizophora mangle) occupies theinner margin of the lagoon and is colonizing the interiorparts, mainly of the center and eastern sections; plantheight ranges from 3 to 5m, although colonies can reach10 m at the eastern end. Aquatic vegetation is scarce inthe channels with patches of the flowering plantsRuppiamaritima and Halodule wrightii. In the eastern section,the bottom is densely covered by H. wrightii and to alesser extent by Thalassia testudinum; both are denselycovered by epiphytic cyanophytes and Batophoraoerstedii. In the central and western section, seagrasseshave been mostly replaced by phytoplankton andmacroalgae (Guimarais-Bermejo andGonzález-De Zayas

Environ Monit Assess

Author's personal copy

2011); patches of B. oerstedii predominate, localizedmainly at the edges of the lagoon.

The main economic activity in the area is interna-tional tourism (more than 50,000 visitors/year), fo-cused on the exploitation of Playa Larga (PL)(Fig. 1). In the last 16 years, the main impacts on thelagoon have come from clearance of vegetation (most-ly mangrove), filling and compaction of soil, dumpingof rubbish and sewage, and the construction of roads,all linked to tourism development. Paradoxically, im-pacts began with the construction of research facilitiesfor the Centro de Investigaciones de EcosistemasCosteros (CIEC) in 1991. Anthropogenic transforma-tion of Laguna Larga continued with the constructionof four hotels, with more than 400 rooms each, on bothsides (south and north) of the aquatic system (Fig. 1).During the construction of the CIEC in 1991, waterflow between the middle and western sections waspartially interrupted by the construction of a bridgeover the water. During the construction of the Blauand Tryp Hotels in 1993 and 1996, respectively, largeareas from both sides of the lagoon in the westernsection were filled, and a large area was dredgedaround site 12 at the western end of the system. Be-tween 1998 and 2001, the construction of SenatorHotel brought nearly 1,000 additional rooms to theeastern section of the lagoon, including huts mountedon stilts in the water. This significantly affected com-munication between the outer (eastern) section, moreexposed to the tides, and the inner (western) section,since it virtually stopped the flow of water between

them. The hotels and other facilities discharge theirsewage directly or indirectly to the lagoon. In the Tryphotel, groundwater is pumped at times in an attempt toimprove water quality.

Sampling and analytical methods

Twelve bimonthly samplings were carried out during2007–2008 to cover the seasonal variations for 2 years(Table 1). An evenly distributed sampling network wasestablished inside Laguna Larga (12 collecting sites).In addition, three sites were located in the adjacent sea(Playa Larga), three in the coral reefs area, and twosites along the coast east of Playa Larga and the lagooninlet (Fig. 1). Temperature and salinity were deter-mined in situ with a WLW digital salinometer (preci-sion of 0.1 for both parameters), previously calibratedin the laboratory. Water samples were collected for the

Reference siteslocated4 km NW

Fig. 1 Location of LagunaLarga, sampling sites, andsections of the lagoon usedfor biogeochemicalmodeling

Table 1 Periods between samplings in Laguna Larga during2007–2008

Period (2007) Days Period (2008) Days

January–February 54 December–February 68

February–April 63 February–April 63

April–June 69 April–May 28

June–September 89 May–July 57

September–October 46 July–October 89

October–December 44 October–December 60

Total 365 Total 365

Environ Monit Assess

Author's personal copy

determination of dissolved oxygen (DO), nutrients,total nitrogen (TN), and total phosphorus (TP) in thelaboratory. For DO analysis, three replicates were takenat each site, taking special care to totally avoid bubblingand contact with the atmosphere until sample fixation.Water samples for dissolved nutrients were filteredthrough a 0.22-μm (Millipore™ HA), fixed with chlo-roform and frozen in polyethylene bottles until analysis.Samples for TN and TP were not filtered.

DO was determined by the Winkler titration methodin three subsamples for each sample fixed in the field.Dissolved nutrients were analyzed using a Skalar SanPlus segmented flow autoanalyzer, using the standardmethods adapted by Grasshoff et al. (1983) and thecircuits suggested by Kirkwood (1994). The precisionof the analysis with this system was: nitrate 0.1 μM,nitrite 0.02 μM, ammonia 0.1 μM, soluble reactive P(SRP) 0.04 μM, and soluble reactive Si (SRSi) 0.1 μM.TN and TP were analyzed as nitrate and SRP afterhigh-temperature persulfate oxidation followingValderrama (1981). Organic N and P were calculatedas the difference between total and nutrient fractions.

Biogeochemical modeling approach

Biogeochemical fluxes of N and P in the lagoon wereevaluated with the LOICZ budget model, a steady-statebox model based on stoichiometrically linked water,salt, and nutrient budgets (Gordon et al. 1996). Detailsof modeling and carbon-budget derivation are on theLOICZ website (http://www.loicz.org/) and elsewhere(e.g., Smith and Hollibaugh 1997). To apply theLOICZ model, Laguna Larga was divided in threerelatively isolated sections (Fig. 1), as evidenced bymorphology and salinity distribution. The principalfluxes of water were calculated as follows:

Precipitationvolume

was calculated bymultiplying the rainfallof each period between samplings(m3 d−1) by the area of each section.Rain data used were measured at themeteorological station 339 of theCuban Net of Meteorological Stationsat Cayo Coco, located 50 m fromLaguna Larga.

Runoffvolume

was calculated following Schreiber(1904), taking into consideration thoseareas that drain into Laguna Larga(Fig. 1) and the monthly averages of

rainfall and temperature reported bymeteorological station 339.

Evaporationvolume

was calculated multiplying evaporationby each section area. Evaporation wascalculated using the Turc method (Xuand Singh 2000), which combinesmaximum possible radiation bymonth and latitude, monthly averageinsolation, and monthly mean temperature.

Groundwatervolume

was estimated from the pumping rate(reported at 168 m3 d−1 in 2007). Noevidence of additional natural groundwater inputs in the study area could bedetermined.

Sewagevolume

was assessed by direct measurement ofwastewater flows at five dischargepoints to Laguna Larga detected in theSenator and Tryp hotels. Monito-ring was done every 2–4 h during a24-h period for two conditions: low(September) and high tourist occupancy(February).

N and Pinputs

were calculated multiplying TN and TPby the water volume of each source foreach period between samplings. Watersamples for the analysis of TN, TP, andnutrients were collected from each ofthe water sources bimonthly or whenpossible throughout 2007–2008; runoffsamples immediately after rainfall; rainsamples whenever the rain volume wasenough for the analysis, and groundwater samples, when it was pumped tothe lagoon during 2007.

N isotopic composition of T. testudinum and Penicillusdumetosus

The stable N-isotopic composition of the seagrass T.testudinum and the seaweed P. dumetosus was used toassess the dominant sources of N in the area becausethey are common both within Laguna Larga and in thecoastal zone outside the lagoon. Samples were collect-ed during the dry (February) and rainy (October) sea-sons of 2007. Sample sites (Fig. 1) were located bothwithin Laguna Larga and outside the lagoon, in theseagrass meadows along PL and along the coral reef

Environ Monit Assess

Author's personal copy

Fig. 2 Seasonal distributionof mean temperature, meansalinity, and mean dissolvedoxygen (DO) at LagunaLarga in 2007–2008

Environ Monit Assess

Author's personal copy

track (CR) located W of the lagoon inlet and also atcoastal sites located E of the lagoon inlet. Two addi-tional reference samples were collected at a pristine sitelocated 4 km W of the study zone. Plant tissue wasrinsed with distilled water and dried to constant weightat <80 °C. The dry samples were finely grounded,exposed to HCl–saturated atmosphere for 4 h, anddried again in an oven. Homogenized sample aliquots(5 mg) were packed in aluminum vials and sent to theisotope analysis laboratory of the University of Cali-fornia at Davis for N isotope determination. These datawere previously used in a different context to assess Ndeposition in the coastal region of Cayo Coco(González-De Zayas et al. 2012).

Results

Hydrological variability

Temperature in Laguna Larga averaged 27.7±2.7 °C(range 21.6–33.1 °C) in 2007–2008, with no significantdifference between 2007 (27.3±2.6 °C) and 2008 (27.2±2.8 °C). Within each year, water temperature in thelagoon ranged from low during the winter season to highduring the summer season (Fig. 2). The widest variabilitywas observed in the shallow sites (sites 6 to 10) withoutinfluence of the adjacent marine and sewage inputs.

Average salinity was 34.5±5.9 (range 8.0–59.0) forthe whole study, with a significant difference (F=33.48,

Fig. 3 Salinity distributionfor each section of LagunaLarga in 2007–2008

Fig. 4 Time variation of ni-trogen fractions: organic ni-trogen (ON), nitrite, nitrateand ammonium at LagunaLarga in 2007–2008

Environ Monit Assess

Author's personal copy

p<0.05) between 2007 (33.0±5.6) and 2008 (36.0±5.9).Salinity showed a high spatial and temporal variabilityin the middle (31.4–59.0) and western (8.0–48.0) sec-tions of the lagoon (Figs. 2 and 3), mainly at the end ofdry seasons. Salinity in eastern section (34.2–37.6) wassimilar to the adjacent sea (36.0–38.0).

High seasonal variability in DO was observed(Fig. 2) with maximum values in February (8.9 ml/L)and minimum in September and October 2007 (3.2–3.5 ml/L). Lowest DO concentrations (including anox-ia) were observed mainly at site 9 in front of the TrypHotel. Mean DO concentration was lower in 2008 (4.8±2.6 ml/L) than in 2007 (5.5±3.2 ml/L), but the differ-ence was not significant (p>0.05).

Nutrient dynamics

TN in water averaged 137.0±98.0 μM (23.1–326.7 μM)in 2007 and 126.0±95.6 μM (17.7–474.8 μM) in 2008,but there were no significant differences (p>0.05) be-tween years (Fig. 4). Organic nitrogen (ON) averaged122.1±94.5 μM (5.9–317.7 μM) and 109.4±83.9 μM(9.2–456.8 μM) in 2007 and 2008, and the differencewas not significant either. ONwas the dominant fraction(23–96 %) of TN at Laguna Larga, and they shared asimilar (r=0.87, p<0.05) spatial pattern, increasing fromthe western towards the eastern section (Fig. 5).Dissolved inorganic N (DIN) concentration in LagunaLarga averaged 15.0±26.2 μM (1.9–167.0 μM) in 2007and 18.8±46.8 μM (0.5–335.7 μM) in 2008. Ammoni-umwas the dominant DIN species (>50%), followed bynitrate (19–40 %). No significant differences in DINconcentrations were found between years (p>0.05), butvariations were observed between samplings (Fig. 4),for example, in May 2008, when DIN peaked due to anitrate increase.

Total P averaged 2.5±1.6 μM (0.05–9.3 μM) in 2007and 4.4±3.1 μM (1.2–14.5 μM) in 2008, although thedifference was not statistically significant (p<0.05). TPshowed an increasing trend towards the western sectionof the lagoon, with the highest peaks at site 9 (Fig. 5).Organic phosphorus constituted nearly 90 % of TP. SRPwas significantly higher (0.52±0.66μM) in 2008 than in2007 (0.09±0.1 μM). SRSi showed an inverse changeamong years, increasing from an average of 13.2±11.4 μM (1.2–57.7 μM) in 2007 to a mean of 25.9±115.5 μM (2.1–980.6 μM) in 2008. Spatially, SRSialso showed an increase trend from the eastern to thewestern section, but its main feature was the marked

peak at site 9 (Fig. 5) due to sewage and fresh waterdischarge, mainly from Tryp Hotel. SRSi concentrationswere higher in the wet season (>90.0 μM) than in therest of the year (<16.0 μM).

Water and salt budgets

Differences in salinity among western, middle, and east-ern sections support the use of the LOICZ budget ap-proach to estimate mixing volumes and residence timesthrough water and salt budgets, although the differenceswere smaller between the eastern section of the lagoon

Fig. 5 Total nitrogen (TN), total phosphorus (TP), and soluble reactivesilicate (SRSi) from inlet to head of Laguna Larga in 2007–2008

Environ Monit Assess

Author's personal copy

and the adjacent sea. Residual and mixing volumes andresidence times were highly variable at the bimonthlyscale (Table 2) but showed defined trends in the overalland annual budgets (Table 3). In 2007–2008, LagunaLarga was on average a net exporter of water (mean337 m3 d−1) to the adjacent sea. However, net flow washigher in 2007, when the lagoon exported 703 m3 d−1,and reversed in 2008, when it imported 28 m3 d−1 ofwater from the sea.

Mixing exchange was higher (38,210 m3 d−1 overallaverage) in the outer (eastern) section of the lagoon andlowest (948 m3 d−1 overall average) in the innermost(western) section. Consequently, residence time (τ)was much longer in the western section, in the orderof years (up to 1,187 days, mean 142 days, Table 2),than in the eastern section of the lagoon (maximum29 days, mean 5 days). Because of the reductions inrainfall and in groundwater pumping, the exchangebetween sections and with the adjacent sea was muchlower in 2008 than in 2007, and consequently, resi-dence times were longer for all sections in 2008 than in2007 (Table 3).

N and P budgets

Runoff and sewage were the largest N and P sources toLaguna Larga, mainly at the eastern section. The fluxesto the western and middle sections were significantlyhigher in 2007 than in 2008 (p<0.05), but the oppositeoccurred for the eastern section (Table 4), which

received the highest sewage fluxes in 2008 (24.4 molof P per day and 275.0 mol of N per day). Overall, the Nflux to the lagoon as a whole decreased by 15 % from2007 to 2008, but the P flux in 2008 almost doubled theintegrated flux for the previous year.

During 2007, Laguna Larga was a net source of N tothe adjacent sea. Its eastern section exported 9.026-mmolm−2 d−1, themiddle section 6.180mmolm−2 d−1,and only its western section behaved as a sink (Table 5).In contrast, during 2008, all the sections of the lagoonbehaved as moderate net N sinks, with rates rangingfrom −2.682 to −0.193 mmol m−2 d−1. At the bimonth-ly scale, the eastern section, which connects to the opensea, remained as a net source of N throughout 2007(Table 6). In contrast, during 2008, the alternation ofsource or sink periods were observed for all sections.Overall, during 2007–2008, Laguna Larga behaved asa net source of N; on average, its eastern sectionexported 3.172 mmol m−2 d−1 and its middle section2.990 mmol m−2 d−1, while the western section was anet sink of N with a rate of −0.418 mmol m−2 d−1

(Table 5).In terms of P, Laguna Larga was also a source during

2007,when the eastern section exported 0.112mmolm−2-d−1, but during 2008, it behaved as a net P sink (−0.179-mmol m−2 d−1, Table 5). Over the full 2007–2008

period, it was also a sink, but the P flux from the adjacentsea was much smaller, averaging −0.067 mmol m−2 d−1.P sink periods were more frequent than P exportationperiods at the bimonthly scale (Table 6).

Table 2 Residual water flow (VR, m3 d−1), mixing volume (VX, m

3 d−1), and residence time (τ, days) from the water and salt budgets forthe periods between samplings during 2007–2008 at Laguna Larga

2007

Section Jan–Feb(54 days)

Feb–Apr(63 days)

Apr–Jun(69 days)

Jun–Sept(89 days)

Sept–Oct(46 days)

Oct–Dec(44 days)

VR VX τ VR VX τ VR VX τ VR VX τ VR VX τ VR VX τ

Western −239 110 191 −15 60 892 −1161 6200 9 −151 600 89 −494 1930 28 −186 250 153

Middle −204 4 46 +64 650 19 −1375 5530 1 −141 220 15 −573 5540 2 −184 890 11

Eastern −270 35880 1.4 −6 13700 3 −2587 84340 0.5 −132 5100 9 −927 148300 0.3 −200 77700 0.6

2008

Section Dec–Feb(68 days)

Feb–Apr(63 days)

Apr–May(28 days)

May–Jul(57 days)

Jul–Oct(89 days)

Oct–Dec(60 days )

VR VX τ VR VX τ VR VX τ VR VX τ VR VX τ VR VX τ

Western −10 50 1187 −20 100 540 −60 150 303 −20 160 385 −20 460 140 +20 400 157

Middle −5 540 24 +1 20 122 +80 380 23 +20 290 31 +40 570 13 +20 510 15

Eastern +20 4420 9 −20 1270 29 +600 12610 4 −20 5580 8 −60 2240 15 +40 7230 6

Environ Monit Assess

Author's personal copy

Nitrogen isotope composition of T. testudinum and P.dumetosus

The stable N isotope ratio (δ15N) measured in T.testudinum leaves ranged from −1.83 to +3.02‰ andaveraged 0.55‰ (Table 7). The δ15N values were slightlylower in the dry season sampling (mean +0.37‰) thanduring the wet season sampling (average +0.72‰), al-though this difference was not statistically significant(p>0.05). The δ15N values were highest (mean +2.37‰) in the seagrasses sampled within the lagoon,which were significantly (p>0.05) different to the sam-ples from the Playa Larga meadows (mean +0.26‰), thecoral track (mean +0.24‰), and the reference sites (mean−0.12‰) located 4 kmwest of the study area. In contrast,the coastal sites located E of the lagoon inlet exhibited

high values (mean +2.12‰), similar to those in thelagoon (p=0.36).

The δ15N values were higher in P. dumetosus than inT. testudinum, ranging from +1.02 to +4.17‰ and aver-aging +2.67‰, and were slightly, though not signifi-cantly, higher (mean +3.14‰) in the dry sampling (Feb-ruary 2007) than in the rainy sampling (mean +2.25‰).Mean values for the sampling sites followed a patternsimilar to those found in the seagrass, but the smallerdifferences found were not significant (p>0.05) in anycase. The mean δ15N found in P. dumetosus was alsohighest (+3.50‰) in the lagoon, lowest at the sites Wofthe inlet (PL +2.51‰, CR +2.48‰) and again high (+3.21‰) at the sites E of the inlet. In contrast to T.testudinum, the δ15N found in P. dumetosus at the ref-erence sites was also relatively high (+3.19‰).

Table 4 Mean N and P inputs (mol d−1) to Laguna Larga during 2007–2008

Input (mol d−1) 2007 2008 2007-2008

Western Middle Eastern Whole Lagoon Western Middle Eastern Whole Lagoon Whole Lagoon

Sewage N 38.8 10 123.3 172.1 33.5 10 275 318.5 245.3

P 1.81 0.3 6.5 8.61 0.7 0.3 24.4 25.4 17.0

Runoff N 154.4 7.2 94.2 255.8 37.1 3 25.3 65.4 160.6

P 1.5 0.7 2.5 4.7 0.4 0.3 0.6 1.3 3.0

Precipitation N 8 2.8 9.4 20.2 6.5 2.4 7.8 16.7 18.5

P 0.2 0.08 0.3 0.58 0.2 0.08 0.3 0.58 0.6

Groundwater N 18.2 0 0 18.2 0 0 0 0 9.1

P 0.4 0 0 0.4 0 0 0 0 0.2

Total N 219.5 20.1 226.9 466.5 77.1 15.4 308 400.5 433.5

P 4 1.1 9.3 14.4 1.3 0.7 25.3 27.3 20.9

Table 3 Mean annual residualvolume (VR), mixing volume(VX), and residence time (τ) inLaguna Larga during 2007–2008

Period Section VR (m3 d−1) VX (m3 d−1) τ (days)

2007 Western −378 1,650 33

Middle −407 2,010 4

Eastern −703 51,800 1

2008 Western −13 247 251

Middle +37 654 16

Eastern +28 4,620 9

Average 2007–2008 Western −195 948 142

Middle −185 1,332 10

Eastern −337 38,210 5

Environ Monit Assess

Author's personal copy

However, in the case of P. dumetosus, none of thesedifferences were statistically significant (p>0.05).

Discussion

The very high levels of dissolved and total N and P foundin Laguna Larga are indicative of the eutrophication ofthis lagoon. TP in Laguna Larga reached levels as high asfound in Lesina lagoon (1–5 μM) during its dystrophiccrisis of 2008, when it changed from a macrophytes-based system toward a phytoplankton-based one (Vigneset al. 2009). In fact, Laguna Larga is already halfwaythrough this process, as its western section is now dom-inated by phytoplankton, and macrophytes are dominantonly in its eastern section (Guimarais-Bermejo andGonzález-De Zayas 2011). The eutrophication of LagunaLarga is likely a result of human activities in the area,mainly direct sewage discharge to the lagoon, which isthe most important nutrient input to Laguna Larga,

mainly because of its high nutrient content, that can reachup to >8,000 μM of TN.

The impact of discharges on water bodies dependslargely on their water exchange and residence time, par-ticularly in coastal lagoons (Bricker et al. 2008). Themorphology of Laguna Larga renders it particularly vul-nerable to nutrient loads and pollution in general. Itselongated shape and shallowness restrict circulation andexchange with the adjacent sea through its single inlet. Itis a paradigmatic case of the “choked lagoon,” as pro-posed by Kjerfve (1994) in its classification of coastalLagoons. Circulation is further restricted in Laguna Largaby the multiple obstructions and fillings derived fromhotel constructions in and around the lagoon. Although,overall, Laguna Larga is a net exporter of water to theadjacent sea, its net water export to sea is very small,representing <1 % of the lagoon volume. As a result, itsresidence time is very high, particularly in its inner sec-tion (western), in the order of years (mean 0.7, maximum3.3, fromTables 2 and 3), similar to other choked lagoons

Table 6 N and P net fluxes (mmol m−2 day−1) in Laguna Larga between samplings during 2007–2008; positive values indicate an excess ofproduction, through remineralization, over biological consumption. Negative values indicate export to adjacent section(s)

2007

Section Jan–Feb (54 days) Feb–Apr (63 days) Apr–Jun (69 days) Jun–Sept (89 days) Sept–Oct (46 days) Oct–Dec (44 days)

ΔN ΔP ΔN ΔP ΔN ΔP ΔN ΔP ΔN ΔP ΔN ΔP

Western +0.385 −0.003 −2.802 −0.125 −8.995 −0.003 −0.244 −0.005 +0.786 +0.005 −0.360 −0.003Middle −1.781 −0.026 −0.229 −0.011 +36.47 −0.169 −1.952 −0.057 +0.997 −0.288 −0.513 −0.019Eastern +3.605 +0.502 +1.346 −0.071 +8.175 −0.740 +3.868 +0.026 +28.79 +0.739 +17.77 +0.752

2008

Section Dec–Feb (68 days) Feb–Apr (63 days) Apr–May (28 days) May–Jul (57 days) Jul–Oct (89 days) Oct–Dec (60 days )

ΔN ΔP ΔN ΔP ΔN ΔP ΔN ΔP ΔN ΔP ΔN ΔP

Western −0.711 −0.011 −0.756 −0.017 −0.908 −0.022 −0.989 −0.013 −0.591 −0.008 −0.454 −0.024Middle +1.064 −0.011 −0.387 −0.025 −1.503 −0.031 −0.611 −0.034 +0.357 −0.131 −1.223 +0.068

Eastern −3.785 −0.352 −3.512 −0.248 +0.218 +0.358 −0.661 +0.015 −4.413 −0.286 −1.264 −0.188

Table 5 N and P net fluxes(mmol m−2 d−1) at Laguna Largaduring 2007–2008

Period Section ΔN (mmol m−2 d−1) ΔP (mmol m−2 d−1)

2007 Western −2.131 −0.023Middle +6.180 −0.090Eastern +9.026 +0.112

2008 Western −0.706 −0.015Middle −0.193 −0.035Eastern −2.682 −0.179

Average2007–2008

Western −0.418 −0.019Middle +2.990 −0.062Eastern +3.172 −0.067

Environ Monit Assess

Author's personal copy

in the Caribbean, like Bojórquez and Nichupté lagoons inCancun (Merino et al. 1992). Although the middle andouter (eastern) sections of Laguna Larga have shorterresidence times, in the scale of days, these are similar tothose in other choked lagoons in Brazil (Knoppers et al.1991) and slightly higher than in other eutrophic coastallagoons, like Chiku (5 days, Lin et al. 2001, Hung andKuo 2002) and the hypertrophic Tapong Bay (10–11-days, Lin et al. 2006, Hung and Hung 2003).

Water exchange in Laguna Larga is subject both totropical rainfall variations and changes in anthropic waterdischarge. During 2008, because of rainfall decrease (by35 % relative to 2007) and groundwater pumping shut-down, Laguna Larga became a net importer of water fromthe sea. In spite of nutrient load variations, the N and Pbudgets in the lagoon followed the variations in the waterbudget. In 2007, Laguna Larga exported water and was asource of N and P to the adjacent sea, but in 2008, it was a

Table 7 N isotopic composition (δ15N,‰) of Thalassia testudinum and Penicillus dumetosus at Laguna Larga and reference sites in thecoastal zone of Cayo Coco, Cuba compared to other sites in the Caribbean and to common N sources to the coastal zone

Sampling site δ15N (‰) References

Mean Range

Thalassia testudinum

LL–Laguna Larga sites +2.37PL–Seagrass meadows outside Laguna Larga +0.26

CR–Coral track outside Laguna Larga +0.24 This study

E–Seagrass and coral sites E of Laguna Larga +2.12R–Reference sites (4 km W of Laguna Larga −0.12Coastal zone of Cayo Coco, Cuba +0.79 −1.83 to +3.16 Gonzalez de Zayas et al. 2012

Bahamas −0.2 Kieckbusch et al. 2004

Bahamas (Union Creek, Great Inagua) +1.2 0.4 to 1.8 Vander Zanden et al. 2013

Bermuda +1.4 Vaslet et al. 2012

Biscaine Bay, Florida +1.4 Kieckbusch et al. 2004

Puerto Morelos Reef, Mexican Caribbean +2.4 2.4 to 2.5 Sánchez et al. 2013

Pearl Cays, Nicaragua +3.2 2.6 to 4.3 Vander Zanden et al. 2013

Yum Balam Reserve, Yucatán +4.2 Sánchez et al. 2013

Florida, St. Joe Bay +5.6 Vander Zanden et al. 2013

Sian Ka’an Reserve +5.8 4.9 to 6.7 Mutchler et al. 2010

Bahia Akumal, Mexican Caribbean +7.1 7.0 to 7.3 Sánchez et al. 2013

Nichupté lagoon, Mexican Caribbean +9.6 8.8 to 10.8 Sánchez et al. 2013

Penicillus dumetosus

LL–Laguna Larga sites +3.50

PL–Seagrass meadows outside Laguna Larga +2.51

CR–Coral track outside Laguna Larga +2.48 This study

E–Seagrass and coral sites E of Laguna Larga +3.21

R - Reference sites (4 km W of Laguna Larga +3.19

Coastal zone of Cayo Coco, Cuba Cayo +2.80 +1.02 to +5.55 Gonzalez de Zayas et al. 2012

Bahamas +0.9 Kieckbusch et al. 2004

Biscaine Bay, Florida +1.7 Kieckbusch et al. 2004

Bahia Akumal, Mexican Caribbean +4.0 Mutchler et al. 2007

Florida Keys +3.10 to 5.32 Dillon et al. 2007

N sourcesSewage +6.0 to +22.0

N fixation −3.0 to 0.0 Sherwood et al. 2010

Atmospheric deposition −12.0 to +4.0

Environ Monit Assess

Author's personal copy

net sink ofN and P. Similar variability has been reported forother coastal lagoons, like San Quintin Bay in México(Camacho-Ibar et al. 2003) and around the world (Hungand Hung 2003, Gupta et al. 2006, McGuirk-Flynn 2008).In other cases, like in Chiku Lagoon, tropical coastal la-goons behave as net sinks of bothN and P (Lin et al. 2001).Apparently, the behavior as a sink or source depends largelyon seasonal and interannual variability. When compared tothemean historical average, rainfall at Laguna Larga during2007 was very close to average (Gonzalez-De Zayas et al.2012), while in 2008, it was 35 % below average. There-fore, it is likely that Laguna Larga behaves, on the average,as a source of N and P to the adjacent sea.

Our stable N isotope data are consistent with this andshow the relative importance of sewage as a source of N inLaguna Larga, where δ15N values in T. testudinum leaveswere significantly higher (>2.1 difference) than outside thelagoon and at the reference sites located 4 km away (>2.4difference, Table 7). Only the reference sites located alongthe coast E of the lagoon inlet had δ15N as high as withinthe lagoon, which suggests that this might be the mainroute followed by the water exported from the lagoon tothe adjacent sea. Although smaller and not significant, inthe case of P. dumetosus, the δ15N values followed asimilar pattern and were also higher within Laguna Largathat outside.

When compared to other systems (Table 7), the P.dumetosus δ15N values in Laguna Larga are above areasconsidered unpolluted, like Bahamas, and similar to areaswith sewage influence (i.e., Akumal Bay). Unfortunately,data for P. dumetosus are relatively scarce, and a thor-ough literature search did not yield any recent reports.Fortunately, for T. testudinum, there are several recentuseful reports (e.g., Sánchez et al. 2013, Vander-Zandenet al. 2013). When compared to δ15N values recentlymeasured at other sewage polluted areas of the Caribbean(e.g., Akumal Bay +7.1, Nichupté lagoon +9.6, Sánchezet al. 2013), the values found in Laguna Larga (+2.37),and Cayo Coco coastal zone in general (−1.83 to +3.16),are relatively low. This couldmean that pollution is not ashigh in Laguna Larga yet, but it could also be due to therelative importance of other N sources in the area. In fact,Gonzalez-De Zayas et al. (2012) found that N depositionseems to be nowadays the most important N source to thecoastal zone of Cayo Coco, and that atmospheric sources(deposition and N2 fixation) comprise 70–90 % of the Nbudget in the area. Since the δ15N of N deposition (−12 to+4‰, Sherwood et al. 2010) and of fixation (−3.0 to0.0‰, Table 7) are much lower than that of sewage (+6.0

to +22.0, Table 7), the apparently low values found inLaguna Larga could result from the high dilution of thesewage signal with these atmospheric sources, in spite ofthe high magnitude of sewage discharge and the evidenteutrophication of the lagoon.

In any case, because of its ecological value and eco-nomic importance, management strategies need to beimplemented urgently to reduce eutrophication in LagunaLarga. The control and reduction of N and P inputs toLaguna Larga is evidently desirable to manage eutrophi-cation from its origin. Sewage and groundwater sourcesare likely easier to control that runoff, and the regulationsand actions should be concentrated there. However, be-cause of the large pools of N and P that sediments andbiomass constitute in coastal lagoons (Merino et al. 1992;Gumpta et al. 2006), input control might not be enough toimprove water quality in the short term, as has occurred inother systems (Bricker et al. 2008). Because of the chokednature of the system, management actions directed todecreasing the residence time within the lagoon and en-hancing water exchange might be particularly efficient toreduce eutrophication impacts in the case of LagunaLarga, as proposed for other choked lagoons (Merinoet al. 1992). Because of its small width and depth, dredg-ing and vegetation trimming along themain channel couldbe a feasible and economic action in this direction. For amuch higher enhancement of water exchange, the optionof developing a flushing system to clean up coastal la-goons in low tide areas (Alatorre-Mendieta et al. 2004) isbeing considered for Laguna Larga by the Cuban govern-ment. Such a system, discharging a net flow of 300–700 m3 d−1 of clean sea water at the inner (western) endof Laguna Larga, could produce a regular water flowalong the axis of the lagoon and reduce the full system’sresidence time to 1–2 days.

Acknowledgments The authors are grateful to R. Almira fornutrient analysis and to R. Almira, G. Vera, L. Rodríguez, and R.Guerra for water sampling and L. Castellanos, A. Fernández, M.Reiné, L. Vázquez, and A. Fernández for rain recollecting. ToAnn Grant, Marcos Merino, and Vicente Rodríguez for theEnglish revision of the document.

References

Alatorre-Mendieta, M., Silva-Casarín, R., Ruiz-Renteria, F., &Merino-Ibarra, M. (2004). A flushing system to clean upcoastal lagoons. Coastal Structures, 2003, 902–910.

Bricker, S., Longstaff, B., Dennison, W., Jones, A., Boicourt, K.,Wicks, C., &Woerner, J. (2008). Effects of nutrient enrichment

Environ Monit Assess

Author's personal copy

in the nation’s estuaries: a decade of change.Harmful Algae, 8,21–32.

Camacho-Ibar, V., Carriquiry, J. D., & Smith, S. (2003). Non-conservative P and N fluxes and net ecosystem productionin San Quintin Bay, México. Estuaries, 26(5), 1220–1237.

Dale, A. W., & Prego, R. (2003). Tidal and seasonal nutrientdynamics and budget of the Chupa Estuary, White Sea(Russia). Estuarine, Coastal and Shelf Science, 56, 377–389.

Dillon, K. S., Chanton, J. P., & Smith, L. K. (2007). Nitrogensources and sinks in a wastewater impacted saline aquiferbeneath the Florida Keys, USA. Estuarine, Coastal andShelf Science, 73(1), 148–164.

González-De Zayas, R., Merino-Ibarra, M., Matos-Pupo, F., &Soto-Jiménez, M. F. (2012). Atmospheric deposition ofnitrogen to a Caribbean coastal zone (Cayo Coco, Cuba):temporal trends and relative importance as a nitrogensource. Water, Air, and Soil Pollution, 223(3), 1125–1136.

Gordon, D. C., Boudreau, P. R., Mann, K. H., Ong, J. E., Silvert,W. L., Smith, S. V., Wattayakorn, G.,Wulff, F., & Yanagi, T.(1996). LOICZ biogeochemical modeling guidelines.LOICZ Reports & Studies, 5, 1–96.

Grasshoff, K., Kremling, K., & Ehrhardt, M. (1983).Methods ofseawater analysis. Weinheim: Verlag Chemie.

Guimarais-Bermejo, M., & Gonzalez-De, Z. R. (2011).Productividad primaria en Laguna Larga, Cayo Coco, Cu-ba. Revista Ciencias Marinas y Costeras, 3, 31–41.

Gupta, G. V. M., Usha Natesan, Ramana Murthy, M. V., SravanKumar, V. G., Viswanathan, S., Bhat, M. S., Ajay Kumar Ray,& Subramanian, B. R. (2006). Nutrient budgets for Muthupetlagoon, southeastern India. Current Science, 90(7), 967–972.

Hung, J. J., & Kuo, F. (2002). Temporal variability of carbon andnutrients budgets from a tropical lagoon in Chiku, southwesternTaiwan. Estuarine, Coastal and Shelf Science, 54, 887–900.

Hung, J. J., & Hung, P. J. (2003). Carbon and nutrients dynamicsin a hypertrophic lagoon in southwestern Taiwan. Journalof Marine Systems, 42, 97–114.

Kieckbusch, D. K., Koch, M. S., Serafy, J. E., & Anderson, W. T.(2004). Trophic linkages among primary producers andconsumers in fringing mangroves of subtropical lagoons.Bulletin of Marine Science, 74(2), 271–285.

Kirkwood, D. S. (1994). SanPlus segmented flow analyzer andits applications seawater analysis. Amsterdam: Skalar.

Kjerfve, B. (1994). Coastal lagoon processes. Elsevier Ocean-ography Series, 60.

Knoppers, B., Kjerfve, B., & Carmouze, J. P. (1991). Trophicstate and water turnover time in six choked coastal lagoonsin Brazil. Biogeochemistry, 14, 149–166.

Lapointe, B. E., & Clark, M. W. (1992). Nutrient inputs from thewatershed and coastal eutrophication in the Florida Keys.Estuaries, 15, 465–476.

Lawrence, E., Jackson, A. R. W., & Jackson, J. M. (1998).Eutrophication, in Longman Dictionary of EnvironmentalScience (pp. 144–145). London England: Addison WesleyLongman Limited.

Leichter, J. J., Wing, S. R., Miller, S. L., & Denny, M. W. (2003).Pulsed delivery of subthermocline water to Conch Reef(Florida Keys) by internal tidal bores. Limnology andOceanography, 41, 1490–1501.

Lin, H. J., Hung, J. J., Shao, K. T., & Kou, F. (2001). Trophicfunctioning and nutrient flux in a highly productive tropicallagoon. Oecologia, 129, 395–406.

Lin, H.-J., Dai, X.-X., Shao, K.-T., Huei-Meei, S., Lo, W.-T.,Hsieh, H.-L., Fang, L.-S., & Hung, J.-J. (2006). Trophicstructure and functioning in a eutrophic and poorly flushedlagoon in southwestern Taiwan.Marine Environmental Re-search, 62, 61–82.

Lucena, J. R., Hurtado, J., & Comín, F. (2002). Nutrients related tothe hydrologic regime in the coastal lagoons of Viladecans(NE Spain). Hydrobiologia, 475(476), 413–422.

McGuirk-Flynn, A. (2008). Organic matter and nutrient cyclingin a coastal plain estuary: carbon, nitrogen, and phosphorusdistributions, budgets, and fluxes. Journal of Coastal Re-search, 55(SI), 76–94.

Merino, M., Gonzalez Reyes, A., Gallegos, M. E. & Czitrom, S.(1992). Eutrophication in the lagoons of Cancún, México.Science of the Total Environment, Supp., 861–870.

Mutchler, Y., Dunton, K. H., Townsend-Small, A., Fredriksen,S., & Rasser, M. K. (2007). Isotopic and elemental indica-tors of nutrient sources and status of coastal habitats in theCaribbean Sea, Yucatan Peninsula, Mexico. Estuarine,Coastal and Shelf Science, 74, 449–457.

Mutchler, T., Mooney, R. F., Wallace, S., Podsim, L., Fredriksen,S., Dunton, K. H. (2010). Origins and fate of inorganic-nitrogen from land to coastal ocean on the Yucatan Peninsula,Mexico. In: M. J. Kennish & H. W Paerl (Eds.), Coastallagoons critical habitats of environmental change (pp. 285–305). Florida: CRC Press.

Sánchez, A., Ortiz-Hernández, M. C., Talavera-Sáenza, A., &Aguíñiga-García, S. (2013). Stable nitrogen isotopes in theturtle grass Thalassia testudinum from the Mexican Carib-bean: implications of anthropogenic development. Estua-rine, Coastal and Shelf Science. doi:10.1016/j.ecss.2013.01.021.

Schreiber, P. (1904). Über die Beziehungen zwischen demNiederschlag und der Wasserführung der Flüsse inMitteleuropa. Meteorologische Zeitschrift, 21, 441–452.

Sherwood, O., Brian, A., Lapointe, M., Risk, E., &Jamieson, R. E. (2010). Nitrogen isotopic records ofterrestrial pollution encoded in Floridian and BahamianGorgonian corals. Environmental Science and Technology,44(3), 874–880.

Smith, S. V., & Hollibaugh, J. T. (1997). Annual cycle andinterannual variability of ecosystem metabolism in a tem-perate climate embayment. Ecological Monographs, 67,509–533.

Umezawa, Y., Miyajima, T., Yamamuro, M., Kayanne, J., &Koike, I. (2002). Fine-scale mapping of land-derived nitro-gen in coral reefs by δ15N in macroalgae. Limnology andOceanography, 47, 1405–1416.

Valderrama, J. C. (1981). The simultaneous analysis of totalnitrogen and total phosphorus in natural waters. MarineChemistry, 10, 109–122.

Vander-Zanden, H. B., Arthur, K. E., Bolten, A. B., Popp, B. N.,Lagueux, C. J., Harrison, E., Campbell, C. L., & Bjorndal,K. A. (2013). Trophic ecology of a green turtle breedingpopulation. Marine Ecology Progress Series, 476, 237–249.

Vaslet, A., Phillips, D. L., France, C., Feller, I. C., & Baldwin, C.C. (2012). The relative importance of mangroves andseagrass beds as feeding areas for resident and transientfishes among different mangrove habitats in Florida andBelize: evidence from dietary and stable-isotope analyses.

Environ Monit Assess

Author's personal copy

Journal of Experimental Marine Biology and Ecology, 434,81–93.

Vignes, F., Barbone, E., Breber, P., D’Adamo, R., Leonilde, R.,Ungaro, N., & Basset, A. (2009). Spatial and temporaldescription of the dystrophic crisis in Lesina lagoon duringsummer 2008. Transitional Waters Bulletin, 3(2), 47–62.

Witek, Z., Humborg, C., Savchuk, O., Grelowski, A., & Lysiak–Pastuszak, E. (2003). Nitrogen and phosphorus budgets of

the Gulf of Gdansk (Baltic Sea). Estuarine, Coastal andShelf Science, 57, 239–248.

Wulff, F., & Stigebrandt, A. (1989). A time-dependent budgetmodel for nutrients in the Baltic Sea. Global Biogeochem-ical Cycles, 3, 63–78.

Xu, C. Y., & Singh, V. P. (2000). Evaluation and generalizationof radiation-based methods for calculating evaporation. Hy-drological Processes, 14, 339–349.

Environ Monit Assess

Author's personal copy