Colored dissolved organic matter signature and phytoplankton ...

Marine Environmental Research xxx (2014) 1e11

Contents lists available at ScienceDirect

Marine Environmental Research journal homepage: www.elsevier.com/locate/marenvrev

Colored dissolved organic matter signature and phytoplankton response in a coastal ecosystem during mesoscale cyclonic (cold core) eddy Gundala Chiranjeevulu a, K. Narasimha Murty a, Nittala S. Sarma a, *, Rayaprolu Kiran a, N.V.H.K. Chari a, Sudarsana Rao Pandi a, Pragada Venkatesh a, C. Annapurna b, K. Nageswara Rao c a

Marine Chemistry Laboratory, Andhra University, Visakhapatnam 530 003, India Department of Zoology, Andhra University, Visakhapatnam 530 003, India c Department of Geo-Engineering, Andhra University, Visakhapatnam 530 003, India b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 7 January 2014 Received in revised form 22 February 2014 Accepted 7 March 2014 Available online xxx

Chromophoric dissolved organic matter (CDOM) and hydrochemical parameters were measured in the nearshore region of the western Bay of Bengal with and without significant terrestrial influence. A mesoscale cyclonic eddy that occupied the northern part of the study area set up a nutrient enriched distinct ecosystem in April (premonsoon) attended with increased levels of DOM fluorescence, particularly the protein tyrosine like fluorescence (B). A new (minor) fluorescence component, attributed to land source was revealed which contained two fluorophores, the red-shifted tryptophan-like (TU) hypothesized as the “unfolded protein” and the petroleum hydrocarbon-like (P). During the eddy, pennate diatom population increased, bringing the centric:pennate diatom ratio to half of what it was during the remaining period (monsoon season). The nutrients distribution suggested that when pennates are favored (premonsoon), orthophosphate and silicate are the limiting nutrients and that when centric diatoms are favored (monsoon season), the limitation is by nitrate. Ó 2014 Elsevier Ltd. All rights reserved.

Keywords: Optical properties Marine ecology Coastal waters Fluorescent dissolved organic matter (FDOM) EEM spectra PARAFAC analysis Petroleum hydrocarbons Bay of Bengal

1. Introduction The Bay of Bengal receives 6.6% of the global river flux (UNESCO, 1979) although it occupies an order of magnitude lower surface area of the world oceans. On its western side, where the influx is from the large and medium sized rivers of India and Bangladesh, the surface freshening leads to a surface barrier layer that limits upwelling and productivity (Vinayachandran et al., 2002). But pulses of increased productivity happen in the aftermath of the episodic tropical cyclones (Rao et al., 2006; Maneesha et al., 2011; Chen et al., 2013) that occur frequently during monsoon season when nutrients are injected in to the surface. Higher productivity will also happen when mesoscale cyclonic (cold core) eddies form at surface which due to positive sea surface height anomaly (SSHA)

* Corresponding author. Tel.: þ91 9866746405. E-mail addresses: [email protected], [email protected] (N.S. Sarma).

promote upwelling, cf. Prasanna Kumar et al. (2004, 2007) and Chen et al. (2013), a phenomenon that has been reported for other parts of the ocean as well (Havry, 1984; Chu et al., 1998; Webster et al., 1999; Oke and Griffin, 2011; Cabrera et al., 2011). Wind driven upwelling has been reported but is confined to isolated areas away from river mouths during the monsoon season in the western Bay of Bengal, when the winds are strong and blow from the southwest (Murty and Varadachari, 1968; Shetye et al., 1991; Narasimha Rao, 2002). Extracellular metabolites are excreted by the growing phytoplankton averaging to 13% of the photosynthetically fixed carbon, a global estimate into seawater (Myklestad, 2000). Most of this is labile and undergoes removal within the euphotic layer by bacterial respiration (del Giorgio and Duarte, 2002). The DOM which had been bacterially degraded/modified while at depth and now emplaced at surface, in case of upwelling, alters the properties of the surface DOM. Although refractory for further bacterial metabolism, the DOM of upwelled water has been shown to be photo-

http://dx.doi.org/10.1016/j.marenvres.2014.03.002 0141-1136/Ó 2014 Elsevier Ltd. All rights reserved.

Please cite this article in press as: Chiranjeevulu, G., et al., Colored dissolved organic matter signature and phytoplankton response in a coastal ecosystem during mesoscale cyclonic (cold core) eddy, Marine Environmental Research (2014), http://dx.doi.org/10.1016/ j.marenvres.2014.03.002

2

G. Chiranjeevulu et al. / Marine Environmental Research xxx (2014) 1e11

labile at the illuminated surface; conversely, the surface DOM considered as photo-refractory is bacterially labile (Nieto-Cid et al., 2006). These are important mechanisms in the production of CO2, supply of nutrients and trace metals and smaller DOM molecules that are better bio-available (Blough and Sulzberger, 2003). Colored DOM (CDOM) which has absorption in the UVeVisible region constitutes a significant fraction of DOM of natural waters. The UV-fluorescent DOM (FDOM) is a further fraction of CDOM. The CDOM (and FDOM) are principally contributed by terrestrial inputs in lakes, estuaries and the coastal region. When these regions become productive the CDOM is also significantly contributed by the living processes e.g., phytoplankton excretion, zooplankton and bacterial metabolism (Coble, 2007). Optical characterization through the absorbance spectra and more recently EEM spectra and their parallel factor (PARAFAC) analysis have been immensely helpful in tracing the DOM inputs and removal mechanisms (Coble, 2007). To our knowledge, there are no reports of how an eddy influences the optical characters of DOM. The objective of this paper was to investigate whether CDOM and FDOM can track a coastal meso-scale eddy and how the phytoplankton responds in the new ecosystem. We also investigated whether any CDOM anthropogenic marker such as petroleum hydrocarbons occur in the coastal waters. 2. Study area The study area is off a major portion of the Indian central east coast (292 km in length) in the Bay of Bengal. Water samples were collected at 21 stations, 3 each (10 m, 20 m and 30 m isobaths) in seven transects (T1eT7, Fig. 1). The T7 is opposite R. Godavari, a major river with annual discharge of 105 km3 (UNESCO, 1979) and which is monsoonal in character. Nearly 90% of its flow (into the Bay of Bengal) takes place during the 4 months JulyeOctober coinciding with the peak monsoon season due to rains in catchment areas of the upper reaches of the river. The river is nearly dry during the non-monsoon particularly pre-summer and summer months (JanuaryeMay). The T6eT2 transects are opposite small rivers. These rivers are minor and their annual discharge is insignificant, compared to R. Godavari (T7). But whenever their inputs take place, they happen in the form of flash flood. The transect T6 is opposite R. Tandava, an ephemeral stream. A sandbar had formed at the mouth of this river preventing exchange of water between the

river and the sea. During the course of this study, the stream remained isolated from the sea. The transect T5 is opposite the Sarada-Varaha estuary. Visakhapatnam (T4) is a major harbor of India, natural, located in the Meghadrigedda estuary. The harbor receives besides riverine inputs during monsoon season, industrial and urban effluents due to the city’s industrialization. Transects T3 and T2 are off Vamsadhara estuary. The T1 is not in the vicinity of any point source of water. 3. Materials and methods 3.1. Vessels and instruments The cruises were conducted on the coastal research vessels (CRV) Sagar Paschimi or Sagar Purvi. The water samples were collected using teflon inner lined 5 L Niskin bottles. A calibrated CTD (Seabird Electronic) onboard measured the salinity. The pH was measured onboard with Thermo Scientific Orion 2Star bench top pH meter with accuracy of 0.01. All spectrophotometric measurements were made with Shimadzu 1800 double beam spectrophotometer using quartz cuvettes. The Excitation Emission Matrix (EEM) spectra were taken on a Horiba Jobin Yvon Fluoromax-4 spectrofluorometer equipped with 150 W ozone-free xenon arclamp and R928P detector. 3.2. Collection of water samples Water samples were collected from surface (1 m) seasonally, representing pre-monsoon (spring inter-monsoon, March 28eApril 1 2010), and monsoon (July 23e27, October 12e16 and November 27e30, 2010). Immediately after retrieval of the sampling bottle, subsamples of the water were drawn into corning bottles separately for dissolved oxygen (DO) and for CDOM in amber colored bottles (pre-washed with chromic acid and combustion-treated for 4 h and provided with teflon screw cap). The CDOM sample was immediately filtered under low light using 0.22 m Millipore membrane filters, pre-equilibrated for 5 min in MilliQ water, and the filtrates were preserved at 4 C in 100 ml amber colored corning bottles (with teflon screw cap) rinsed thoroughly with filtered seawater. For chlorophyll estimation, GF/F (47 mm) filters containing particulate matter of 1 L water of each sample, preserved at 20 C

Fig. 1. Study area with transects T1 (Nuvalarevu), T2 (Bhavanapadu), T3 (Kalingapatnam), T4 (off Visakhapatnam), T5 (Bangarammapalem), T6 (Pentakota) and T7 (Kakinada) and stations (B).



onboard were extracted with 90% acetone overnight at 4 C, and the absorption spectrum of the clear supernatant measured using Shimadzu 1800 double beam UV Visible spectrophotometer. The concentration of chl-a was calculated using the Jeffrey equations (Jeffrey and Humphrey, 1975). Filtered water, soon after collection, was used for the estimation of nutrients (nitrite, ammonium, phosphate, and silicate) onboard, and nitrate in the shore laboratory using standard methods (Grasshoff et al., 1999). The DO was estimated by the modified Winkler’s method (Grasshoff et al., 1999). 3.3. Optical methods Just before CDOM (and FDOM) measurement, the samples were equilibrated to room temperature (22 1 C), filtered again as above and both optical absorbance and excitation emission matrix (EEM) spectra were measured within 1e4 days after collection. For blank, ultrapure water from the Millipore Q-3 model water purification system having conductivity C1. The Components C1, C2, C3 and C5 contained a fluorophore each that correspond with the marine humic-like (M), tryptophan protein-like (T), UV humic-like (A) and tyrosine protein-like (B) fluorophores respectively as per the nomenclature of Coble (1996), and reported since extensively in marine waters (Coble et al., 1998; Yamashita and Tanoue, 2003; Stedmon and Markager, 2005a,b; Yamashita et al., 2008; Dubnick et al., 2010). The B fluorophore (C5) attributed to degraded protein of autochthonous DOM, often reported as a major fluorophore (Mayer et al., 1999; Stedmon and Markager, 2005a, 2005b; Murphy et al., 2008; Yamashita et al., 2008; Kowalczuk et al., 2009; Gao et al., 2010; Para et al., 2010; Chari et al., 2012) represented 54% and 50% of

a

BIX = -0.005*S300-500 + 0.018, R²

= 0.522

0.02

the total fluorescence in pre-monsoon and monsoon season respectively. The B is known to be masked by the T fluorescence by internal quenching (Lakowicz, 1999; Mayer et al., 1999). Since T is the second most major fluorescence, and does not apparently mask B in the present study, a laboratory experiment on the aerobic degradation of natural phytoplankton assemblage was performed and our (unpublished) results suggest that B reached up to 0.0084 RU without being quenched even as T reached a maximum concentration of 0.0087 RU. Quenching did prevail in field samples at T > 0.012 RU (Pandi, 2013). Since in the present study, T 0.0117 RU, the B fluorescence is inferred as not undergoing quenching. The C2 (T) was significantly more enriched during the monsoon season than during pre-monsoon season (average of monsoon:premonsoon ratio, 2.3). Chlorophyll a reached higher values in the monsoon season (range, 0.1e10.2 mg L1) than during pre-monsoon (2.2e3.9). The BIX, an indicator of recent biological activity correspondingly was higher during monsoon (2.2e3.9) than during premonsoon (0.72e1.78). The BIX had a negative linear relationship with S300e500 during monsoon season (Fig. 5a). Since spectral slope is inversely related to molecular weight (Helms et al., 2008), the DOM (of lower molecular weight, higher BIX) is formed of in situ, more during the monsoon season when fresh nutrient from external source is added than the pre-monsoon season when nutrients had just shown their presence consequent to the (episodic) eddy and the ecosystem was unstable due to the (turbulent) eddy. During the premonsoon season, the relationship of BIX was positive with several constituents including S300e500 (Fig. 5b), salinity, nutrients (particularly nitrate), and fluorescence Components (Table 3), and it appears BIX is likely dependent on the strength of upwelling (see Table 4). 4.3.3. Fluorophores enrichment at the eddy centre Transect averages of intensities (Fig. 6) show that the fluorophores are enriched in the T2eT3 region compared to the remaining region (except T7 where the R. Godavari influence is seen) by 760, 560, 380, 230 and 170% for M, B, (TU þ P), A and T respectively. The enrichment due to the eddy is more prominently seen for B which is the most major (Fig. 7). Moreover, for the B fluorophore, there is a significant linear correlation with nitrate. The slope of the linear regression equation is about an order of 1 magnitude steeper for NO1 3 < w7 mM than for NO3 > w7 mM (Fig. 7). The upwelled water in which nitrate liberation took place by organic matter mineralization (premonsoon) is thus a stronger source of the B fluorophore (degraded protein) than the terrestrial sourced nitrate (OM) during the monsoon season. The other fluorophores are also formed simultaneously during bacterial respiration (Nieto-Cid et al., 2006). But their enrichment was not as much as in the case of B. The A formed recently is photolabile (Nieto-Cid

b

0.018 BIX = 0.005*S300-500 + 0.005, R² = 0.483

0.018 0.016

0.016 S300-500

S300-500

7

0.014 0.012 0.01

0.014 0.012

0.008 0.5

1

1.5 BIX

2

0.01 1

1.5 BIX

2

Fig. 5. Scatter plots for seasonal distribution of S300e500 against BIX: (a) during monsoon and (b) during pre-monsoon.


8


4.4. The new protein like and petroleum-like fluorescence in one component

Table 3 Spearman’s rank correlation coefficients of high two-tailed significance. Correlation of

C1 (M)

C2 (T) C3 (A) C4 (TU þ P) BIX

DO

Correlation with

C2(T) C3 (A) C4 (TU þ P) aCDOM(350) C3 (A) C4 (TU þ P) C5(B) C1 (M) C2 (T) C3 (A) C4 (TU þ P) C5 (B) aCDOM(350) NO1 3 Chl a NHþ 4 Chl a 1 NO3 IP NHþ 4

Premonsoon

Monsoon

r

p

r

p

0.85 0.95 0.90 0.76 0.83 0.96 0.89 0.86 0.67 0.78 0.70 0.84 0.79 0.52 e e 0.65 0.58 0.58 e