Red tide development modeling in Persian Gulf and study nutrient effects on algal bloom

Page created by Harry West
 
CONTINUE READING
Red tide development modeling in Persian Gulf and study nutrient effects on algal bloom
Research in Marine Sciences
                                                                                                      289
Volume 3, Issue 2, 2018
Pages 289 -302

                 Red tide development modeling in Persian Gulf
                   and study nutrient effects on algal bloom

                              Madjid Abbaspour1, and Elahe Zohdi 2,*
                       1
                        Professor, Sharif University of Technology, Tehran, Iran
            2
             MSc of shipping engineering, Sharif University of Technology, Tehran, Iran

                    Received: 2018-02-22                      Accepted: 2018-05-10

Abstract

Red tide or algal bloom occurs in estuarine, marine, or fresh water due to accumulating algae rapidly
in water column and results in water coloration in high concentration and widespread serious effects
on public health, fishing, and even tourism industry. Persian Gulf despite rich reserves of oil and
gas, military strategic location, rich resources and valuable diverse aquatic habitats like coral reefs
has been recently confronted with this phenomenon due to human, industrial and agricultural waste,
oil exploration, permanent ships’ traffic and shallow water, geographical and climate conditions and
limited water exchange with external environment. Due to recent broad steady development of red
tides that have irreparable effects on the ecosystem, economic conditions, and public health of the
coastal residents, society’s need for understanding is more pressing than ever. Among the investigation
methods, numerical modeling has great importance due to low time and cost. MIKE software is used
in this paper, considering physical, chemical and hydrodynamics factors in a zone and biological
aspects of this phenomenon and temperature, salinity and nutrient effects on red tide growth. In fact
predicting HABs’ occurrence minimizes their damage by quick preparation of mitigation activity
and prevents economic losses in aquaculture industry, desalination equipment, and also respiratory
and skin irritation problems for swimmers. In this paper, the effects of pollution outflows and
hydrodynamic currents of Persian Gulf on red tide development in 2008-2009 period and variation
of dissolved oxygen amount to chlorophyll-A was simulated and studied by utilizing MIEK21 and
ECOLAB modules. The simulation results showed that the red tide development direction was
according to the Persian Gulf mainstream direction. In addition, chlorophyll-A concentration was
higher near the sewage outflows and persistent in those areas at all times till nutrients present there. In
addition, by increasing chlorophyll-A concentration, the dissolved oxygen concentration increased.

Keywords: Algal bloom; Chlorophyll-A; HABs; MIKE software; Red tide.

* Corresponding Author: ela.zohdi@alum.sharif.edu
Red tide development modeling in Persian Gulf and study nutrient effects on algal bloom
290 Red tide development modeling in Persian Gulf and... / 289 - 302

1. Introduction                                           toxic species have all increased. Indeed, there
                                                          is a conviction among many experts about
Land sourced marine pollution, overexploitation           expanding. Is this expansion real? Is it a global
of living marine resources, destruction of                epidemic? Is it related to human activities? Or is
habitat and introduction of harmful aquatic               it a result of increased scientific awareness and
organisms and pathogens to new environment                improved surveillance or analytical capabilities?
had been identified as the four greatest threats          (Anderson, 1994; Wikipedia, 2014) Of 5000+
to the world’s oceans. Since invasive species             microscopic algae species or phytoplankton
cannot be cleaned or absorbed in the ocean, so,           exist worldwide, about 300 species can cause
their effects on the environment are irreversible.        high concentration of red tides. Only one
Red tide is one of the invasive species and a             fourth of this amount is known to be harmful
colloquial term used to refer to one of a variety         or toxic which include dinoflagellates, diatoms,
of natural phenomena known as algal bloom,                haptophyceae, cyanophyceaes, and some of
in which estuarine, marine, or fresh water algae          cycloflagellate. Studies indicate the presence of
accumulate rapidly in the water column and                21 bloom-former species, are potentially toxic,
resulting in coloration of the surface water,             in Persian Gulf (Malaei Tavana et al., 2008;
varying in color from purple to almost pink,              Wikipedia, 2014). Damage caused by red tides
normally being red or green. These algae are not          is divided into two categories direct poisoning
native to special area, in other word; they have          (death of marine animals, human poisoning,
various species with broad ranges. The first case         respiratory and skin irritation due to producing
of red tide occurred in 1793, British Columbia,           strong toxins like neurotoxin) and indirect
Canada (Mohammadi, 2008; Wikipedia, 2014).                poisoning (marine mortalities by depleting
Whether harmful or harmless kinds of red tide             the oxygen in the water, human poisoning by
has been reported from all over the world such            eating shellfishes contaminated with red tide
as Brazil, Korea, China, Japan, Chile, Canada,            toxin through feeding). A bloom develops
UK, France, Hong Kong, India, the countries               when these single-celled algae photosynthesize
of the Black Sea and the Mediterranean sea and            and multiply; converting dissolved nutrients
recently the Persian Gulf and the Caspian Sea             and sunlight into plant biomass and the
(Malaei Tavana et al., 2008; Seddigh Marvasti,            population of grazing zooplankton and aquatic
2016). Blooms and their environmental                     species is small. It seems that their population
effects in an aquatic ecosystem depend on the             is not affected only by their growth rate, but
species involved, biological and ecological               also wind or ocean currents bring them to the
characteristics of the species, the scale of this         water surface (Anderson, 1994; Malaei Tavana
phenomenon, hydrological and topographical                et al., 2008; Vahedi, 2009). The dominant
status of environment where they are found                mode of reproduction is simple asexual fission.
(Bahrami Rad, 2013). According to scientists              When nutrients are scarce, some species switch
finding, the number of toxic blooms, the scale,           to sexual reproduction, form cysts, and this
and complexity of this natural phenomenon,                transformation facilitates species dispersal
the economic losses from them, the types of               along with currents or the other forces to new
resources affected and the kinds of toxins and            environments. Different phytoplankton species
Red tide development modeling in Persian Gulf and study nutrient effects on algal bloom
Research in Marine Sciences 291

have different growth rate. Measures taken           sparse observations. To watch red tides, there
pursuant to which the average daily growth rate      are different methods: field observation and
of dinoflagellates is almost 0.3 per day with a      using sampling data, satellite-based studies,
three-day period for regeneration. Diatoms have      laboratory studies, modeling like complex
a daily growth rate more than dinoflagellates;       numerical models, conceptual models, simple
40% of them have regeneration period less            analytic formulae, semi-empirical models,
than three days (Anderson, 1994; Malaei              aggregated box models or zero-dimensional
Tavana et al., 2008; Taghavi and Abbaspour,          models (McGillicuddy, 2010; Park et al., 2013;
2012). Two major contributing factors in the         Seddigh Marvasti, 2016).
growth of phytoplankton are temperature and
                                                     1.1. Red Tide Occurrence in Persian Gulf
nutrients, especially phosphorus and nitrogen
(Wikipedia, 2014). Some of the studies, such as      Persian Gulf despite rich reserves of oil and gas,
Vollenweider (1992), Hodgkiss and Ho (1997),         military and strategic location, rich resources
Peinert et al. (1982), Glibert et al. (2010),        and important and valuable diverse aquatic
showed eutrophication is a major contributor         habitats such as coral reefs, due to sewage and
to HABs development in their growth period           human waste, oil exploration and extraction
(Wong et al., 2007; McGillicuddy, 2010;              operations, permanent ships’ traffic, massive
Sadeghi Mazidi et al., 2011). Long-term              military operations and wars for several years,
studies at the local or regional level do show       industrial and agricultural waste which led
that red tides are increasing as coastal pollution   to the entry of organic materials like NO3,
worsens. Between 1976 and 1986, as the               NH4, PO4 and the other nutrients to this zone,
pollution around Tolo Harbor in Hong Kong            and shallow water, geographical and climate
grew six fold, red tides increased eightfold         conditions and limited water exchange with the
(Anderson, 1994; Malaei Tavana et al., 2008).        external environment, has been confronted with
According to recent wide spread and persistent       numerous hazards. Therefore the probability of
HABs effects, society’s need for understanding       the presence of non-native species or increasing
these phenomena is more pressing than ever.          nitrification phenomenon in this area is not far-
Technological advances have expanded our             fetched. The red tide occurrence is one of the
capabilities for observing the ocean, providing      major environmental issues raised in Persian
unprecedented opportunities not only for the         Gulf in recent years (Taghavi and Abbaspour,
detection of blooms, but also for the physical,      2012; Ebrahimi, 2015). Related to the red tide
chemical, and biological factors that trigger        occurrence, according to statistics of the Iranian
their initiation, development, and ultimate          Fisheries Science Research Institute, in short,
demise. However, despite these rapidly               from 1991 to 1997, close to 30 algal blooms
expanding observational capabilities, HAB            in the Persian Gulf and Oman Sea waters are
processes will continue to be under sampled          recorded but they are short-term, about 48
for the foreseeable future, owing to the wide        hours (Rezaee and Kabiri, 2009; Shakerzadeh
range of space and time scales relevant to these     et al., 2009). Among the identified algae in this
oceanographic phenomena. As such, we must            area, about 12 species can cause red tide bloom
rely on models to help interpret our necessarily     which the eight species of them, mentioned in
Red tide development modeling in Persian Gulf and study nutrient effects on algal bloom
292 Red tide development modeling in Persian Gulf and... / 289 - 302

 Figure 1. The dominant species of red tide phenomenon in the Persian Gulf (Taghavi and Abbaspour, 2012)

the following, have further spread (Taghavi and occurrence. Natural environmental conditions
 The most extensive
Abbaspour,      2012). red tide bloom in the Persian Gulf                happened
                                                                    during            fromfor
                                                                              the bloom      end-August
                                                                                                this type ofto algae,
                                                                                                                Marchsalinity
                                                                                                                         2008
The  most
 which   wasextensive
                 caused red by tide                    Persian is from
                                     bloom in thePolykricoides
                                Cochlodinium                           between the 25   to 30 ppt and
                                                                                    dinoflagellate       water temperature
                                                                                                      category     From late
Gulf   happened
 September             from end-August
                to December        2008, red tide  to was
                                                        March       is between
                                                             observed   in an area32of to    34 ºCkilometers
                                                                                          several     considering fromthat
                                                                                                                         Sirikthe
2008 which was caused by Cochlodinium proper temperature range for the growth of
 port in Oman Sea to Bushehr at the center line of Persian Gulf which was covered approximately
Polykricoides from the dinoflagellate category phytoplankton in the Persian Gulf is in fall and
 750 km
From  lateof   coastline to
            September        (Rezaee
                               December and 2008,
                                              Kabiri,red 2009).
                                                           tide Restriction
                                                                    winter, the which    stop growing
                                                                                     phenomenon          and blooming
                                                                                                       occurrences      wasinin
was  observed
 the Oman     Seainmainly
                       an area  wasof the
                                       several
                                           lack kilometers          that period
                                                  of nutrients. Since                (Manshouri
                                                                           in the north              and Tabiee
                                                                                           of the Strait   of Hormuz, Yeganeh,
                                                                                                                            the
from  Sirik cities
 industrial   port inlikeOman      Sea to
                              Bandar        Bushehr
                                         Abbas     and at           2009;andMohammadi
                                                            the exist
                                                         Qeshm                                   Tahroodi
                                                                                due to the defects      in the et urban
                                                                                                                    al., 2013;
                                                                                                                           and
center line of Persian Gulf which was covered Seddigh Marvasti, 2016). According to the
 industrial wastewater system, the amount of organic materials in coastal waters is high during the
approximately 750 km of coastline (Rezaee and sampling values by Persian Gulf and Oman
 whole of
Kabiri,      the year,
         2009).            and almost
                     Restriction    which algal
                                              stopdensity
                                                     growingwas verySeahigh    all of theResearch
                                                                          Ecological        9-month Institute,
                                                                                                       period in parameters
                                                                                                                    this area.
and  blooming
 In other  areas,inthe thealgal
                             Oman    Sea mainly
                                 density    reducedwas  andthe      variations
                                                             increased             at the
                                                                         periodically        timethese
                                                                                          which     of blooming
                                                                                                        enhancement    wereandin
lack of nutrients.
 reduction             Since ininthe
              were obvious           thenorth   of theand
                                           northern      Strait     the regions
                                                             southern     followingof therange:
                                                                                            Omansalinity      (37-39.2
                                                                                                    Sea (Hamzei,            ppt),
                                                                                                                       2012).
of Hormuz, the industrial cities like Bandar temperature (20.4-33.9 ºC), PH (7.96-8.6),
 Cochlodinium, dinoflagellate with yellowish to brownish green pigment, was able to produce
Abbas and Qeshm exist and due to the defects in Nitrate (0.059-0.707 mg/L), Phosphate (0.035-
 spherical
the urban andor oval    cysts wastewater
                   industrial    and by releasingsystem, active
                                                            the oxygen      causes and
                                                                    1.01 mg/L)       oxidative   fish gillsmg/L).
                                                                                          DO (3.9-9.8         and increases
                                                                                                                      Different
amount   of organic
 the amount              materials
                  of oxygen            in coastal
                                  in the   area atwaters      is ofstudies
                                                       the time       red tide showed     that the
                                                                                  occurrence.         amount
                                                                                                  Natural        of dissolved
                                                                                                             environmental
high  during during
 conditions      the whole      of the for
                          the bloom      year,this
                                                 andtypealmost
                                                           of algae,oxygen
                                                                       salinity(DO)      in the25presence
                                                                                  is between         to 30 ppt of  andalgae
                                                                                                                        wateris
algal density was very high all of the 9-month increased due to photosynthesis and producing
 temperature is between 32 to 34 ºC considering that the proper temperature range for the growth
period in this area. In other areas, the algal oxygen radical by Cochlodinium and salinity
 of phytoplankton
density    reduced and   in theincreased
                                   Persian Gulf      is in fall andequal
                                                periodically          winter,   the phenomenon
                                                                             to 39.2                  occurrences
                                                                                      ppt and temperature         abovewas30inºC
which   these (Manshouri
 that period      enhancement    andand    reduction
                                       Tabiee    Yeganeh, were2009;limits    the Cochlodinium
                                                                      Mohammadi         Tahroodi et growth     (Bahri,
                                                                                                       al., 2013;        2009).
                                                                                                                     Seddigh
obvious   in the northern and southern regions of Based on scientific studies and field evidence,
 Marvasti, 2016). According to the sampling values by Persian Gulf and Oman Sea Ecological
the Oman Sea (Hamzei, 2012). Cochlodinium, the assumptive reasons in this case are
 Research Institute, parameters variations at the time of blooming were in the following range:
dinoflagellate with yellowish to brownish green presented as follows: transport iron-rich
 salinity (37-39.2
pigment,    was able ppt),      temperature
                         to produce      spherical(20.4-33.9
                                                       or oval ºC),suspended
                                                                      PH (7.96-8.6),       Nitrate
                                                                                   particles        (0.059-0.707
                                                                                               by wind      and dustmg/L),
                                                                                                                         storms
cysts  and by
 Phosphate          releasing mg/L)
               (0.035-1.01        activeandoxygen                   fromDifferent
                                                        causes mg/L).
                                               DO (3.9-9.8                  the littoral  statesshowed
                                                                                       studies   around that
                                                                                                           the Persian
                                                                                                                 the amountGulf;
oxidative fish gills and increases the amount Surface disorganization due to sea currents;
of oxygen in the area at the time of red tide Climate changes on a large scale such as El
                                                                293
Red tide development modeling in Persian Gulf and study nutrient effects on algal bloom
Research in Marine Sciences 293

                     (a)                                (b)                      (c)                       (d)

Figure 2. (a) Coral reefs mortality due to suffocation and loss of light- the layer of algae covering their
surface, (b) foam (sewage) due to aquatic mortality, (c) Dead fish due to red tide in the Persian Gulf,
Autumn 2008, (d) Water discoloration due to Cochlodinium bloom in north of the strait of Hormuz,
Qeshm in November 2008 (Rezaee and Kabiri, 2009; Taghavi and Abbaspour, 2012)

Niño; transport species through ships ballast the water quality and damage to desalination
 In general,
waters;     Heavyto decrease
                       rainfall growth    of redtotides
                                 in addition              different
                                                    raising          methodsand
                                                                 equipment       including
                                                                                     systems,physical    (likedesalination
                                                                                                especially      spraying
the
 clay,temperature,
         reducing sourcesreducing   the amount
                                of nutrients,        of salt
                                                 etc.),  chemicalequipment       (RO) hypochlorite
                                                                     (using sodium       of Qeshm, and          ships by
                                                                                                          produced       and
and    increasing
 electrolysis          the wateretc.)
                  of seawater,      nutrients   especially
                                        and biological     (usingoffshore     facilities,
                                                                   of zooplankton           damage tobacteria,
                                                                                       and anti-algae        the seafood
                                                                                                                      etc.)
vitamin B12; Entering abundant volume of industry, especially aquacultures in Hormozgan
 are used. Since prevention methods in comparison to decrease methods is preferable, so it is
nutrients and increased water temperature due and Bushehr (Rezaee and Kabiri, 2009; Taghavi
 essential
to  geologic  to find    a method
                   activities  and for    predicting
                                     upwelling         red tide and
                                                   (Qeshm         occurrence
                                                                       Abbaspour,and 2012),
                                                                                       observing   the trend
                                                                                              extinction         of itshells
                                                                                                            of pearl    by
earthquake
 consideringoccurrence
                  all factors and
                               suchyearly    upwelling
                                      as salinity,        on andprecipitation,
                                                    temperature,        sea cows of       Persian
                                                                                       current       Gulf and
                                                                                                velocity     (Hormozgan
                                                                                                                  relative
the  east coast
 humidity.     For ofthisOman   Sea abetween
                          purpose,               February
                                        monitoring               news, 2011),
                                                      network system       can besocial   andand
                                                                                    created    economic
                                                                                                  used forproblems
                                                                                                              recordingfor
and March); entering abundant volume of lowlander and stop tourism industry in Kish
 the required information. An integrated predicting model could be an important goal for research
nutrients, urban and industrial sewage, from and Qeshm, producing foam with fetid smell
 programs
coastal        on ecology
           region            and algal
                     of Bandar     Abbas.bloom
                                            The time    (Vahedi,caused
                                                  southern         2009; byTaghavi
                                                                              bacteriaanddecomposition
                                                                                           Abbaspour, 2012). which results
cities without treatment plant pour their sewage in lowlander harassment and a sharp decline in
 2. 2. Materials
directly     into the sea   andmore
                                  methodsthan 50 years. commutes (Mohammadi, 2008; Mohammadi
Discharge of Bandar Abbas sewage was equal Tahroodi, 2013).
to  300 2.1.
 2.1.      lit/s Introduction
                  in that time (Shakerzadeh
                                    to MIKE software et al., and In governing
                                                                     general, to theories
                                                                                      decrease growth of red tides
2009; Vahedi, 2009; Seddigh Marvasti, 2016).                     different methods including physical (like
Consequences
 MIKE softwareofisCochlodinium
                          able to simulate algal           in spraying
                                                 bloomchemical,
                                              physical,                       clay, reducing
                                                                      or biological    processessources
                                                                                                   in lakes,ofestuaries,
                                                                                                                  nutrients,
the
 bays,Persian
        coastalGulfareassummarized
                           and seas withasthefollows:
                                                highestfish      etc.),Forchemical
                                                          reliability.                  (using sodium
                                                                            red tide modeling     in MIKEhypochlorite
                                                                                                               software,
mortality due to deposition in the fish gills and produced by electrolysis of seawater, etc.)
 the hydrodynamics module-MIKE21 and the biological module-ECOLAB have been used.
suffocation, increase anaerobic respiration and and biological (using of zooplankton and anti-
 MIKE21
the  depletionis suitable
                    of DO intool
                               thefor two-dimension
                                    water  and poisoning  (2D) simulation     of physical,
                                                                 algae bacteria,     etc.) arechemical
                                                                                               used. Since or biological
                                                                                                                prevention
aquatic,
 processesespecially
                in coastalfishorand   benthic
                                  marine        organisms
                                            areas                methods
                                                   and it is based       on in
                                                                             thecomparison
                                                                                   numerical to    decrease
                                                                                                solution    of methods
                                                                                                                 the two is
who    feed planktons
 dimensional       shallow(Rezaee      and Kabiri,
                              water equations         2009;
                                                  - the          preferable,incompressible
                                                         depth-integrated       so it is essentialReynolds
                                                                                                    to find a averaged
                                                                                                               method for
Farman Ara et al., 2012), injury and severe predicting red tide occurrence and observing
 Navier-Stokes equations. Thus, the model consists of continuity, momentum, temperature,
damage to light-dependent aquatic ecosystems, the trend of it by considering all factors such
 salinity, and
especially         density
               coral         equations.
                       reef in  Hormozgan In the  horizontal
                                               waters    due plane     an unstructured
                                                                 as salinity,              grid is
                                                                                 temperature,        used. The spatial
                                                                                                 precipitation,      current
to   suffocationof and
 discretization              reduce light
                        the primitive          penetration
                                        equations   is performed velocity
                                                                    using aandcellrelative humidity.
                                                                                    centered            For thismethod.
                                                                                              finite volume        purpose,
(Mohammadi,          2008; Vahedi,
 MIKE 21 has different                2009;
                                modules       Taghavi
                                           which         andwater
                                                   are for       a monitoring
                                                                    quality andnetwork      system
                                                                                   ecological         can beECOLAB
                                                                                                modeling,      created and
Abbaspour, 2012), discoloration and changing used for recording the required information.
                                                            295
Red tide development modeling in Persian Gulf and study nutrient effects on algal bloom
294 Red tide development modeling in Persian Gulf and... / 289 - 302

An integrated predicting model could be                   ecosystem into a reliable numerical model for
an important goal for research programs on                accurate predictions. The predefined ECOLAB
ecology and algal bloom time (Vahedi, 2009;               templates containing the mathematical
Taghavi and Abbaspour, 2012).                             descriptions of ecosystems can be used as a
                                                          basis and edited arbitrarily or the new template
2. Materials and methods                                  can be defined based on processes taking
2.1. Introduction to MIKE software and                    place in the ecosystem. The demand for tailor
governing theories                                        made ecosystem descriptions is great, because
                                                          ecosystems vary. A template is independent
MIKE software is able to simulate physical,               of grid systems and the same template can be
chemical, or biological processes in lakes,               integrated with one, two and three dimensional
estuaries, bays, coastal areas and seas with              or hydrology modeling. The template can
the highest reliability. For red tide modeling            describe dissolved substances, particulate
in MIKE software, the hydrodynamics                       matter of dead or living material, living
module-MIKE21 and the biological module-                  biological organisms and other components.
ECOLAB have been used. MIKE21 is suitable                 State variables included in ECOLAB can
tool for two-dimension (2D) simulation of                 either be transported by advection-dispersion
physical, chemical or biological processes in             processes based on hydrodynamics, or have
coastal or marine areas and it is based on the            a more fixed nature (e.g. rooted vegetation or
numerical solution of the two dimensional                 mussels). The description of the ecosystem
shallow water equations - the depth-integrated            state variables in ECOLAB is formulated as a
incompressible Reynolds averaged Navier-                  set of ordinary coupled differential equations
Stokes equations. Thus, the model consists of             describing the rate of change for each state
continuity, momentum, temperature, salinity,              variable based on processes taking place in the
and density equations. In the horizontal plane            ecosystem. Besides the pure process orientated
an unstructured grid is used. The spatial                 description (Eulerian modeling), this module
discretization of the primitive equations is              also supports the characterization of individual
performed using a cell centered finite volume             spatial entities or particles with associated
method. MIKE 21 has different modules                     attributes (agent based modeling or Lagrangian
which are for water quality and ecological                modeling). After creating or picking an
modeling, ECOLAB module is used.                          ECOLAB template for the ecosystem, the
ECOLAB is a complete numerical laboratory                 ECOLAB model must be set up and run in
for customizing aquatic ecosystem, water                  MIKE 21 FLOW MODEL (FM). In this paper,
quality, and ecological studies by describing             Eutrophication 1 ECOLAB template had
chemical, biological, ecological processes                been used. This template considers dispersion
and interactions between state variables and              in Eulerian framework by default, and in
physical processes. By ECOLAB exactly the                 order to simulate the simultaneous processes
required model for the specific ecosystem can             of transport, dispersion and biological/
be developed and its processes can be described           biochemical processes considers a system of 12
by using standard and transform any aquatic               differential equations describing the variations
Red tide development modeling in Persian Gulf and study nutrient effects on algal bloom
is great, because ecosystems vary. A template is independent of grid systems and the same
                  template can be integrated with one, two and three dimensional or hydrology modeling. The
                  template can describe dissolved substances, particulate matter of dead or living material, living
                  biological organisms and other components. State variables included in ECOLAB can either be
                  transported by advection-dispersion processes based on hydrodynamics, or have a more fixed
                  nature (e.g. rooted vegetation or mussels). The description of the ecosystem state variables inResearch in Marine Sciences                             295
                  ECOLAB is formulated as a set of ordinary coupled differential equations describing the rate of
                  change for each state variable based on processes taking place in the ecosystem. Besides the pure
             for
             process12 orientated
                           components                  (11 movable
                                              description                         components
                                                                (Eulerian modeling),           this module Based also on     the the
                                                                                                                        supports     above, red tide modeling in
             in both hydrodynamic and biological modules Persian Gulf has been done (MIKE manuals,
             characterization      of    individual     spatial    entities  or  particles  with   associated attributes  (agent  based
             modeling or Lagrangian modeling). After creating or picking an ECOLAB template for the
             and    a fixed nature component belonging to the 2012).
             ecosystem, the ECOLAB model must be set up and run in MIKE 21 FLOW MODEL (FM). In
             benthic system). The processes describing the
             variations of the components in time and space 2.2. Red Tide Modeling Process
             this paper, Eutrophication 1 ECOLAB template had been used. This template considers
             dispersion in Eulerian framework by default, and in order to simulate the simultaneous processes
             are    dependent on external factors such as the
             of transport, dispersion and biological/ biochemical processes considers                      Generally,
                                                                                                                    a systemtheofsteps
                                                                                                                                     12 to carry out an environmental
             salinity,         water           temperature,               the    light
             differential equations describing the variations for 12 components (11project influx,
                                                                                                                         consist of:
                                                                                                             movable components       in project analysis (analysis
             and    the      discharges.              In    this    paper,       the
             both hydrodynamic and biological modules and a fixed nature component     variation                    belonging to of the the ecosystem problem and
                                                                                                           and evaluation
             of    chlorophyll-A,
             benthic   system). The processes        inorganic
                                                          describing the   phosphorus
                                                                              variations of the andcomponents in time and space are
                                                                                                           solution strategy); collection of data, field
             nitrogen
             dependent on     and     alsofactors
                                 external        DO are   suchstudied.        The water
                                                                 as the salinity,     dynamics
                                                                                             temperature, the light influx, and the
                                                                                                           measurements, and laboratory analysis; set
             of     advective
             discharges.     In this paper,  ECOLAB
                                                  the variation state            variables
                                                                     of chlorophyll-A,     inorganicis phosphorus and nitrogen and
                                                                                                           up and run the model; calibration; solution
             calculated
             also DO are studied. by: The dynamics of advective ECOLAB state variables is calculated by:
                                                                                                           (running the production simulations and
             డ௖      డ௖         డ௖          డ௖         డమ ௖          డమ ௖       డమ ௖
                ൅ ‫ ݑ‬൅ ‫ ݒ‬൅ ‫ ݓ‬ൌ ‫ܦ‬௫ మ ൅ ‫ܦ‬௬ మ ൅ ‫ܦ‬௭ మ ൅ ܵ௖ ൅ ܲ௖ (1)                                             presenting the results). (1)          For red tide modeling
             డ௧      డ௫         డ௬          డ௭         డ௭            డ௭         డ௭

                                                                                                           in MIKE software, the hydrodynamics module-
             which C is the concentration of 296                               the ECOLAB
                                                                                                           MIKE21 and the biological module-ECOLAB
             state variable; u, v, w are flow velocity
                                                                                                           have been used (MIKE manuals, 2012). Red
             components;
              which C is the              Dx, concentration
                                                    Dy, Dz are the           of the  dispersion
                                                                                          ECOLAB tide       statemodeling
                                                                                                                     variable; process
                                                                                                                                    u, v, winare        flow Gulfvelocity
                                                                                                                                                     Persian           zone is
             coefficients; Sc is sources and sinks; and Pc
he ECOLAB     components;
                   state variable;      Dx, Du,y, D           are
                                                        v,zprocesses.
                                                              w are  the dispersion
                                                                            flowThe   velocitycoefficients;dividedSc isinto    four and
                                                                                                                          sources       sections:
                                                                                                                                             sinks; and Pc indicates
             indicates             ECOLAB                                                     mass
             which
              ECOLAB     C isprocesses.
                                   the concentration   Thesinks;mass  of balance
                                                                            the ECOLAB                     1) Define
                                                                                                   state variable;
                                                                                       for chlorophyll-A                 u, Persian
                                                                                                                    reads:   v, w are Gulf  flow topography
                                                                                                                                                    velocity        by using
 rsion coefficients;
             balance      Sfor
                             c is chlorophyll-A
                                    sources and                    reads:and Pc indicates
             components; Dx, Dy, Dz are the dispersion coefficients; Sc shoreline                           is sources and andsinks;
                                                                                                                                 waterand  depth    data of Persian Gulf in
                                                                                                                                               Pc indicates
 e for chlorophyll-A
              ௗ஼ு            reads:
             ECOLAB ൌ ‫݊݋݅ݐܿݑ݀݋ݎ݌‬
                               processes. The      െ ݀݁ܽ‫݄ݐ‬          െ ‫݊݋݅ݐܽݐ݊݁݉݅݀݁ݏ‬
                                                          mass balance          for chlorophyll-A   ൌ ܴܲ‫ܪܥ‬ Longitude/Latitude
                                                                                                          reads: െ ‫ ܪܥܧܦ‬െ ܵ‫ܪܥܧ‬projection     ൅ ܵ‫ܪܥܧ‬௡ିଵbetween (2)      23.7 to
               ௗ௧
             ܴܲ‫ ܪܥ‬െ ‫ ܪܥܧܦ‬െ ܵ‫ ܪܥܧ‬൅ ܵ‫ܪܥܧ‬௡ିଵ
 ݊‫ ݊݋݅ݐܽݐ‬ൌ ௗ஼ு                                                                              (2)(2)         33 ºC north and 47.7 to 58 ºC east in xyz format
              where
             which
               ௗ௧
                       C  n-1
                   ൌ ‫݊݋݅ݐܿݑ݀݋ݎ݌‬
                             is   denotes
                                 the                the
                                                െ ݀݁ܽ‫݄ݐ‬
                                          concentration    input
                                                              െ of    from
                                                                 ‫݊݋݅ݐܽݐ݊݁݉݅݀݁ݏ‬
                                                                      the       the
                                                                             ECOLAB   above
                                                                                          ൌstate layer
                                                                                              ܴܲ‫ܪܥ‬      െ (n>1).
                                                                                                          ‫ܪܥܧܦ‬
                                                                                                    variable;      െ The
                                                                                                           of u,ETOPO-1
                                                                                                                    v,  w chlorophyll-A
                                                                                                                       ܵ‫ܪܥܧ‬ are൅flow
                                                                                                                                   ܵ‫ܪܥܧ‬    ௡ିଵ
                                                                                                                                          velocity
                                                                                                                                    website             (2)
                                                                                                                                                   production
                                                                                                                                                   including     amount
                                                                                                                                                                  Longitude/
             where
             components;   n-1   D denotes
                                      ,  D    , D     the
                                                    are   the  input
                                                                dispersion from      the    above
                                                                                coefficients;    S   is sources  and   sinks;  and  P    indicates
  above layer is:(n>1).
             where      n-1   The   x
                               denotes  chlorophyll-A
                                            y     z
                                                the input       fromproduction
                                                                          the above   amount       c       Latitude position and depth of every point and
                                                                                                                                      c
                                                                                        layer (n>1). The chlorophyll-A production amount
             layer
             ECOLAB      (n>1).
                             processes.    The        chlorophyll-A
                                                The mass                            production
                                                              balance for chlorophyll-A            reads: generate proper mesh for computational grid:
             is:
             amount        ൌis:ሺ ೘೔೙െሻ݀݁ܽ‫݄ݐ‬
                                 ஼ு
              ܴܲ‫ܪܥ‬
             ௗ஼ு
                   ൌ ‫݊݋݅ݐܿݑ݀݋ݎ݌‬
                                                 ൈ ‡š’      ሺ‫ܨ‬ଷ ሺܰǡ ܲሻሻ ൈ ൌܴܲܲ‫ܥ‬
                                                         െ ‫݊݋݅ݐܽݐ݊݁݉݅݀݁ݏ‬               ܴܲ‫ ܪܥ‬െ ‫ ܪܥܧܦ‬െ       import
                                                                                                              ܵ‫ ܪܥܧ‬൅ shoreline
                                                                                                                         ܵ‫ܪܥܧ‬௡ିଵ           data,
                                                                                                                                              (2) modify boundaries,
                                                                                                                                                                       (3)
                                     ூ௄
              ௗ௧               ஼ு೘೔೙
 ܴܲܲ‫ܥ‬        ܴܲ‫ ܪܥ‬ൌ ሺ                     ሻ ൈ ‡š’ሺ‫ܨ‬ଷ ሺܰǡ ܲሻሻ ൈ ܴܲܲ‫ܥ‬                         (3)(3)        define specifications of boundaries           (3)     for2 the   area
              wheren-1CH
             where        					   ூ௄
                              denotes   is   the
                                            the     coefficient
                                                  input    from     the   of  determining
                                                                          above    layer  (n>1).   the
                                                                                                    Theminimum
                                                                                                         chlorophyll-A  chlorophyll-A
                                                                                                                           production amount  production      (E/m   /d -1
                                                                                                                                                                          ),
                                 min
                                                                                                           by creating closed domain,2 define                    position of
 rmining thewhere
               minimum
             where
             is:
              CH   max
                       CH CH
                          is  the
                              minmin
                                         is the
                                 chlorophyll-A
                                   iscoefficient
                                         the          coefficient
                                                coefficient   production
                                                             of   determiningof (E/m
                                                                   of determining determining
                                                                                         2 -1
                                                                                          /d
                                                                                         the
                                                                                       the      ),
                                                                                               minimum
                                                                                             maximum         chlorophyll-A
                                                                                                              chlorophyll-A      production
                                                                                                                                    production
                                                                                                                                                         -1
                                                                                                                                                (E/min/dthe
                                                                                                           Jask (Lat.=25/63، Long.=57/77) and Majis        ), absence of
             the
             CHmaxminimum
 g the maximum         is the
                     chlorophyll-Acoefficient chlorophyll-A
                                                       of determining
                                                  production           in   production
                                                                            the the   maximum
                                                                                  absence      of(E/chlorophyll-A production in the absence of
              nutrient
             ܴܲ‫ܪܥ‬
                             ஼ு೘೔೙
                       ൌ ሺlimitation, ሻ ൈ ‡š’ሺand   ‫ ܨ‬ሺܰǡ F3ܲሻሻ (N)     is calculated by the following
                                                                     ൈ ܴܲܲ‫ܥ‬                                (Lat.=24/518،formula:Long.=56/606)  (3)          and draw arc,
             m    /d ),limitation,
                 2 -1
             nutrient       CH  ூ௄              the Fଷcoefficient
                                           is and          (N)    is         of determining
                                                                     calculated       by  the  following     formula:
 lated by the following formula:   max                  3                                                  and define a unique integer value for boundaries
             the
             where maximum
                      CH is the chlorophyll-A
                                            coefficient௉ே of determining production         in thechlorophyll-A production (E/m2/d-1),
                                                                                    the minimum
              ‫ܨ‬ଷ ሺܰሻ ൌmin‫ܪܥ‬௠௔௫ ൈ௉ே                  ቄሺ െ ܲܰ௠௜௡ ሻȀሺܲܰ௠௔௫ െ ܲܰ௠௜௡                            (open
                                                                                                              ሻቅ boundaries and land boundaries mark                   (4) 2
             ‫ܨ‬ଷ ሺܰሻ
             absence
             CH   max isൌthe  of
                             ‫ܪܥ‬      nutrient
                                coefficient            limitation,
                                           ൈ ቄሺ of determining
                                                       െ௉஼ܲܰ௠௜௡ ሻȀሺܲܰ      theand     F
                                                                                maximumെ ܲܰ(N)      is
                                                                                                     ሻቅ
                                                                                             chlorophyll-A      production    in the  absence   of       (4)
                                                                                        3 (4)
                                  ௠௔௫                                            ௠௔௫           ௠௜௡
ܲܰ௠௔௫ െ ܲܰ௠௜௡ ሻቅ                                   ௉஼                                                      and 1, respectively) to create a mesh, define
             calculated
             nutrient
              SECH        and     by
                         limitation,
                                   DECH  theand  following
                                                  F   (N)
                                                  read:
                                                    3        is      formula:
                                                                calculated      by  the following formula:
             SECH and DECH read:                                                                           polygons for closed domains and exclude the
             ‫ܨ‬ଷ ሺܰሻ ൌ ‫ܪܥ‬௠௔௫ ൈ ቄሺ஼ு ஼ு
                                                ௉ே
                                                    െ ܲܰ௠௜௡ ሻȀሺܲܰ௠௔௫ െ ܲܰ௠௜௡ ሻቅ (4)                        small islands from the(4)mesh domain, import
              ܵ‫ܪܥܧ‬
             ܵ‫ܪܥܧ‬  			  ൌൌܵ‫ܥܲܧ‬ ܵ‫ܥܲܧ‬      ൈൈ   ሺ ሺሻ௉஼ ሻ
                                                ௉஼
                                                                                                                                                         (5)           (5)
                                                ௉஼
                                                                                             (5)           scatter data to interpolate water                depths in mesh
             SECH
             SECH and     and DECH DECH    read: read:
                                                                    ஼ு                                     elements, generate depth-independent mesh,
             ‫ܪܥܧܦ‬       ൌ
              ‫ ܪܥܧܦ‬ൌ ሺ‫ܥܲܧܦ‬   ሺ‫ܥܲܧܦ‬           ൅ ‫ܥܴܲܩ‬ሻ
                                                 ൅ ‫ܥܴܲܩ‬ሻ      ൈ ቀ ൈ ቁቀ஼ுቁ                                                                                (6)           (6)
             ܵ‫ ܪܥܧ‬ൌ ܵ‫ ܥܲܧ‬ൈ ሺ ሻ
                                            ஼ு                      ௉஼ ௉஼                        (5) then smoothing and refining               (5)      and    analyzing the
                                            ௉஼                                               (6)
             Based on the above, red tide ஼ு                   modeling in Persian Gulf has been           meshdone  to (MIKE
                                                                                                                         ensure manuals,
                                                                                                                                     that the2012).
                                                                                                                                                  mesh is optimal, carry
              Based
             ‫ܪܥܧܦ‬  		  ൌon      the above,
                            ሺ‫ܥܲܧܦ‬         ൅ ‫ܥܴܲܩ‬ሻ     redൈ ቀtideቁ modeling in Persian            (6) Gulf has been done (MIKE(6)manuals, 2012).
g in Persian Gulf has been done (MIKE manuals, 2012).          ௉஼                                          out mesh-interpolation by interpolating scatter
             2.2.     2.2.     Red        Tide      Modeling           Process
             Based on the above, red tide modeling in Persian Gulf has been done (MIKE manuals, 2012).
                  2.2. 2.2. Red Tide Modeling Process
ocess
                  Generally,
                  2.2.       the Tide
                       2.2. Red  stepsModeling
                                      to carry out an environmental project consist of: project analysis (analysis
                                                Process
           2.2. 2.2. Red Tide Modeling Process
           Generally,
           andproject
                         the of
                evaluation
                              steps
                                  the
                                      to carry outproblem
                                       ecosystem
                                                     an environmental       project consist of: project analysis (analysis
                                                             and solution strategy); collection of data, field
environmental           consist   of: project  analysis (analysis
           Generally, the steps to carry out an environmental project consist of: project analysis (analysis
           and evaluation
           measurements,    andof    the ecosystem
                                 laboratory  analysis;  problem
                                                       set up field andthesolution
                                                              and run                 strategy); solution
                                                                             model; calibration;    collection    of data, field
                                                                                                             (running
 roblem andand
             solution
               evaluationstrategy);    collection
                            of the ecosystem       of data,
                                                problem and solution   strategy); collection of data, field
Red tide development modeling in Persian Gulf and study nutrient effects on algal bloom
mesh elements, generate depth-independent mesh, then smoothing and refining and analyzing
    the mesh to ensure that the mesh is optimal, carry out mesh-interpolation by interpolating
    scatter data amounts and represent the bathymetry in mesh nodes according to the input data,
296 and
     Redfinally   export modeling
         tide development the mesh    file inGulf
                                  in Persian  order
                                                  and...to  use- 302
                                                         / 289    it in MIKE21 hydrodynamic module.
    According to the above steps, the results of this part are shown in Figure 3.

                                                 (a)

                                                   (b)
 Figure 3. (a) Persian Gulf bathymetry, (b) mesh file and the defined boundary codes

data amounts and represent the bathymetry in rose plot for during the year, hydrodynamic
                                                298
mesh nodes according to the input data, and sources data like the average flow of Mond,
finally export the mesh file in order to use it Dalaki, Arvand rivers respectively equal to
in MIKE21 hydrodynamic module. According 4.39, 5.13, 1750 cubic meter per second and
to the above steps, the results of this part are inputting wastewater flow of Bandar Abbas
shown in Figure 3.                                  equal to 300 liter per second: in this modeling
2) Definition of hydrodynamic flow in the Evaporation-precipitation over the Persian
Persian Gulf, according to water level data in Gulf, the dominant wave system and density
the open boundary between Jask and Majis variations based on salinity and temperature
from IOC website in the period of 1 July 2008 are ignored. Thus, time series and wind rose
till 30 June 2009, and prevailing hydrodynamic plot are as Figure 4.
forcing including wind force data in station The wind rose in Figure 5 is in accordance with
(long.=56, lat.=26) from Coastal and Port the studies about winds pattern over the Persian
Engineering Department of Ports and Maritime Gulf that states this area is in the pattern of the
Organization in the form of time series and wind summer monsoon winds (southwest) and winter
Red tide development modeling in Persian Gulf and study nutrient effects on algal bloom
sources data like the average flow of Mond, Dalaki, Arvand rivers respectively equal to 4.39,
  5.13, 1750 cubic meter per second and inputting wastewater flow of Bandar Abbas equal to
  300 liter per second: in this modeling Evaporation-precipitation over the Persian Gulf, the
  dominant wave system and density variations based on salinity and temperature
                                                                  Research        areSciences
                                                                           in Marine  ignored.297
  Thus, time series and wind rose plot are as Figure 4.

                                                    (a)

                                                    (b)
Figure 4. Water level time series (m) in (a) Jask and (b) Majis stations

                                                  299

                                 (a)                                                     (b)
Figure 5. (a) Wind time series [speed (m/s) and direction (deg)], (b) wind rose plot in station (Long =56,
Lat = 26)

monsoon  winds
  The wind  rose(northeast)
                  in Figure and
                              5 isthe
                                    in strength of with
                                        accordance     2016).  After defining
                                                          the studies          water
                                                                      about winds      level over
                                                                                    pattern    boundary
                                                                                                  the
winter monsoon
  Persian        winds
          Gulf that     arethis
                    states  mucharealess  than
                                      is in     patterncondition,
                                            thethe                wind monsoon
                                                        of the summer    forcing, winds
                                                                                   rivers,(southwest)
                                                                                             topography
summer monsoon winds (Seddigh Marvasti, and determining initial conditions in MIKE 21
  and winter monsoon winds (northeast) and the strength of winter monsoon winds are much
   less than the summer monsoon winds (Seddigh Marvasti, 2016). After defining water level
   boundary condition, wind forcing, rivers, topography and determining initial conditions in
   MIKE 21 Flow Model (FM), the hydrodynamic model is created and a two-dimensional
Red tide development modeling in Persian Gulf and study nutrient effects on algal bloom
298 Red tide development modeling in Persian Gulf and... / 289 - 302

Figure 6. Comparison of simulated water level (m) with actual water level data in Kharg station (eddy
viscosity=0.34, bed resistance=50)

FlowTheModel
         observed    difference
                 (FM),            in the figures model
                         the hydrodynamic          may be due to to
                                                                  50consider
                                                                      and 0.34only
                                                                                 which   are used
                                                                                     the wind   force.in Inthegeneral,
                                                                                                                ultimate
is created   and are
    four factors   a two-dimensional
                        important in the output
                                             Persianand       hydrodynamic
                                                       Gulf Stream               model
                                                                       circulation:     to define
                                                                                     wind,   tides, red
                                                                                                     water tide.
                                                                                                               density
a point output in Kharg station are defined. In The observed difference in the figures may be
    variation and turbulent diffusion caused by vertical mixing processes. (Taghavi and
order to convergence and stability of the model due to consider only the wind force. In general,
    Abbaspour,
during   solving 2012)
                     partialThis difference equations
                              differential     can be due to     notfactors
                                                              four    considerareother factorsinlike
                                                                                   important      the tides,
                                                                                                          Persianwater
                                                                                                                     Gulf
using   explicit
    density         method, Courant number is Stream circulation: wind, tides, water density
             variation.
defined as 0.5. The simulation time step is variation and turbulent diffusion caused
    4) to4)10800
equal        Defining    trend of
                     seconds,    andalgal
                                       the bloom
                                            numberinofPersian by Gulf     zone
                                                                   vertical      by utilizing
                                                                              mixing            data (Taghavi
                                                                                       processes.      variation in   and
timesalinity, temperature,
      steps due    to cover chlorophyll-A,
                              the year is definednutrient
                                                        as concentration
                                                              Abbaspour,(nitrate
                                                                              2012) and
                                                                                     Thisphosphate)
                                                                                            differenceand   canDO be in
                                                                                                                      due
2919.
    formLow    order method
          of average     monthlyis in
                                    selected                  to not consider
                                                to solve ECOLAB
                                       open boundary,                             other factors
                                                                        forces including           like tides,equal
                                                                                            solar radiation        water
the equations. Two-dimensional            results show density variation.
    to constant value 26.5 E/m2/d, ECOLAB loading concentration including concentration of
counterclockwise flow of the Persian Gulf.                    4) Defining trend of algal bloom in Persian
    Bandar    Abbas   sewage     pollution   equal
3) Persian Gulf hydrodynamic model calibration Gulf  to  300  lit/s inzone
                                                                        eutrophication   1 ECOLAB
                                                                               by utilizing                 template: in
                                                                                                data variation
by Horizontal
     using actual diffusion
                      water coefficients,
                               level data growth
                                              of Kharg and death    rate, temperature,
                                                              salinity,    sedimentationchlorophyll-A,
                                                                                           and grazing rate,        the
                                                                                                                 nutrient
station  which
    growing       received
               rate          fromon
                    dependence      Coastal    and Port
                                        temperature,          concentration
                                                        salinity  and nutrients(nitrate  and phosphate)
                                                                                  are changeable                and DO
                                                                                                     to investigate
Engineering Department of Ports and Maritime in form of average monthly in open boundary,
    the effect and calibrate the model. This template only considers diatoms and green algae. In
Organization: in order to model calibration, ECOLAB forces including solar radiation
bedthis  modeling
     resistance   anddiatoms    growth values
                      eddy viscosity      rate equal   to 0.8 equal
                                                  during       per dayto and  greenvalue
                                                                         constant    algal26.5
                                                                                            growth
                                                                                                 E/mrate
                                                                                                       2
                                                                                                         /d, equal
                                                                                                              ECOLAB to
the 1.4 per day, sedimentation
     simulation                      rate in theFinally,
                    have been changed.             depth of less   than concentration
                                                              loading    2 meters equalincluding
                                                                                           to 0.25 m/day        and in
                                                                                                       concentration
the optimal values of them respectively equal of Bandar Abbas sewage pollution equal to 300
                                                         301
the depth of greater than 2 meters equal to 0.5 m/day and maximum grazing rate equal to 1.5
   per day and death rate equal to 0.02 per day are defined. In output part of the simulation, a
                                                                          Research in Marine Sciences 299
   two-dimensional output including chlorophyll-A, inorganic nitrogen and phosphorus, and DO
     variation is defined.
lit/s in eutrophication 1 ECOLAB template: the depth of less than 2 meters equal to 0.25 m/
Horizontal diffusion coefficients, growth and day and in the depth of greater than 2 meters
 3. 3. Results and Discussion
death rate, sedimentation and grazing rate, equal to 0.5 m/day and maximum grazing
the growing rate dependence on temperature, rate equal to 1.5 per day and death rate equal
 Finally, and
salinity    the results  of simulation
                  nutrients                show that
                              are changeable      to chlorophyll-A
                                                       to 0.02 per day concentration
                                                                          are defined.isIn increased  byof
                                                                                            output part
 enhancing the amount
investigate                of calibrate
                  effect and  nitrate andthephosphate,
                                              model. and
                                                       the increasing  in chlorophyll-A
                                                              simulation,                 concentration
                                                                            a two-dimensional       output
This
 causestemplate    onlyin considers
          increasing                  diatoms and
                           DO concentration.           including
                                                 Also the          chlorophyll-A,
                                                            trend of   chlorophyll-Ainorganic
                                                                                       developmentnitrogen
                                                                                                       is
green algae. In this modeling diatoms growth and phosphorus, and DO variation is defined.
 according to Persian Gulf stream circulation as well as increasing growth rate due to nutrient
rate equal to 0.8 per day and green algal growth
 concentration.
rate  equal to 1.4 per day, sedimentation rate in

                                                   302

 Figure 7. The trend of chlorophyll-A (mg/l) by utilizing eutrophication 1 template and dominant stream in
 Persian Gulf (chlorophyll-A dispersion and diffusion is in accordance with main direction of dominant
 stream and also location of contamination sources)
300 Red tide development modeling in Persian Gulf and... / 289 - 302

3. Results and Discussion                                 Conclusion

Finally, the results of simulation show that             In the recent years, Persian Gulf due to special
chlorophyll-A concentration is increased by              geographical and climate conditions has been
enhancing the amount of nitrate and phosphate,           confronted with red tide phenomenon and it
and increasing in chlorophyll-A concentration            gives seriously impact on ecosystem, public
causes increasing in DO concentration. Also              health and economy condition. Due to the facts
the trend of chlorophyll-A development is                that through the investigating methods of this
according to Persian Gulf stream circulation as          phenomenon, numerical modeling has great
well as increasing growth rate due to nutrient           importance because of the need to lower time
concentration.                                           and cost, in this paper, the effects of pollution
Figure 7. The trend of chlorophyll-A (mg/l) by utilizing eutrophication 1 template and dominant stream in
                                                         outflows and hydrodynamic currents of Persian
Persian Gulf (chlorophyll-A dispersion and diffusion is in accordance with main direction of dominant
stream and also location of contamination sources)

                        (a)                                                  (b)
Figure 8. Comparison the proceed of (a) chlorophyll-A (mg/l) to (b) nitrogen and phosphorus dispersion
(mg/l)

                         (a)                                                   (b)
Figure 9. Comparison of (a) DO variation (mg/l) to (b) chlorophyll-A concentration (mg/l) by utilizing
eutrophication 1 template

                                                    303
Research in Marine Sciences 301

Gulf on Red Tide development in 2008-2009              the 10th International Conference on Coasts,
period and variation of amount of DO to                Ports and Marine Structures, Tehran, Iran.
chlorophyll-A was simulated and studied by           Glibert, P. M., Allen, J.I., Bouwman, A. F.,
utilizing MIEK21 FLOW MODEL (FM) and                   Brown, C.W., Flynn, K. J., and et al. 2010.
EUTROPHICATION 1 ECOLAB TEMPLATE                       Modeling of HABs and eutrophication:
modules. The simulation results showed that            Status, advances, challenges. Journal of
the direction of red tide development was in           Marine Systems, 83(3-4): 262-75.
accordance with the mainstream direction of          Hamzei, S. 2012. Field Observation and
Persian Gulf. Considering nitrate concentration        numerical modeling of red tide development
(0.059-0.707 mg/l) and phosphate concentration         in the north part of Hormoz Strait, Science
(0.035-1.01 mg/l) in hydrodynamic boundary             and Technology University, Iran.
and Bandar Abbas swage pollution equal to 300        Hodgkiss, I.J., and Ho, K.C. 1997. Are changes
lit/s, chlorophyll-A concentration was increased       in N: P ratios in coastal waters the key to
with increasing nitrate and phosphate as well          increased red tide blooms? Hydrobiologia,
as persistent in those areas at all times till the     352: 141e147.
presence of nutrients over there. In addition,       Hormozgan news. 2011. Available at http://
due to increasing chlorophyll-A concentration,         www.hormozgannews.com/Pages/News-
the dissolved oxygen concentration increased.          2325.html [Accessed on July 2017].
                                                     Malaei Tavana, H., Behpoor, S., Changizi,
References                                             M., and Karimi, H. 2008. Investigate the
                                                       reinforcing factors in forming and occurrence
Anderson, D. M. 1994. Red tides. Scientific            of harmful algal bloom, National Conference
  American Journal, 271(2): 52-58.                     on Human, Environment and Sustainable
Bahrami Rad A. 2013. Algae bloom and its               Development, Hamedan, Iran.
  importance in aquatic ecosystems, http://          Manshouri, M., and Tabiee Yeganeh, S. 2009.
  www.newposeidon.blogfa.com/post-57.                  Investigate the red tide causes and controlling
  aspx.                                                it in Persian Gulf, 12th National Congress
Bahri, A. 2009. Bloom and its importance in            on Environmental Health, Shahid Beheshti
  aquatic ecosystems, Persian Gulf and Oman            University, Tehran, Iran
  Sea Ecological Research Institute. Available       McGillicuddy, D.J. 2010. Models of harmful
  at: http://pgoseri.ac.ir [Accessed on March          algal blooms: conceptual, empirical, and
  2017].                                               numerical approaches, Preface to a Special
Ebrahimi, M. 2015. Monitoring algae bloom in           Issue of Journal of Marine Systems, DRAFT
  coastal waters of Persian Gulf and Oman Sea          MANUSCRIPT.
  (Hormozgan coastal waters), Persian Gulf           MIKE manuals. 2012. ECOLAB Short
  and Oman Sea Ecological Research Institute.          Scientific Description; MIKE 21 - FLOW
Farman Ara M., Hosseini Balam F., and                  MODEL FM-Particle Tracking Module;
  Hasanzade E. 2012. Numerical simulation              MIKE21/3        PARTICLE         TRACKING;
  of subsurface flow effects on the distribution       GEODESY          IN     MIKE       ZERO-Map
  of red tide phenomenon in the Persian Gulf,          Projection and Datum Conversion; MIKE21
302 Red tide development modeling in Persian Gulf and... / 289 - 302

  FLOW         MODEL        FM-ECOLAB/OIL                   Aquaculture Journal, 5: 13-21.
  SPILL; MIKE ZERO-PREPROCESSING&                         Seddigh Marvasti S. 2016. Numerical
  POSTPROCESSING; MIKE ZERO-Project                         simulation to study effective factors and
  Oriented Water Modelling; MIKE ZERO-                      different mechanisms of red tide in Oman
  mesh generator; MIKE ZERO-The Common                      Sea, Physical Oceanography PhD thesis,
  DHI User Interface for Project Oriented                   Islamic Azad University, Science and
  Water Modelling, available at https://www.                Research Branch, Tehran, Iran.
  mikepoweredbydhi.com        [Accessed      on           Shakerzadeh, F., Bavandi, S., Bahmanmanesh,
  March 2017].                                              A., and Hedayatifard, M. 2009. Red tide
Mohammadi, H. 2008. Red tide, http://                       and its destructive effects on the marine
  hamedmohamadi63.blogfa.com/post-27.                       environment with emphasis on marine
  aspx                                                      mollusks, The International Conference of
Mohammadi Tahroodi, M. Fatemi, S. M.,                       Persian Gulf in Islamic Azad University of
  Nabavi, S.M., and Kazemi Tabar, S. M.                     Bushehr, Iran.
  2013. Study the occurrence of Cochlodinium              Taghavi, L., and Abbaspour, M. 2012. The
  algal bloom and environmental parameters                  determinative factors for modeling and
  in the coastal waters of Qeshm-                           management of red tide in the Persian
  Hormozgan, International Congress on                      Gulf, The First International Conference
  Environmental Crises and its Solutions,                   of Environmental and Geopolitical Persian
  Kish Island, Iran                                         Gulf.
Park, S., Lee, Y., and Lee, S.R. 2013.                    Vahedi, M. 2009. Red tide threat. Port & Sea
  Forecasting Red Tide using Ensemble                       magazine, 23: 116-119.
  Method. International Journal of Software               Vollenweider, R.A. 1992. Coastal marine
  Engineering and Its Applications, 7(5): 145-              eutrophication. In: Vollenweider, R.A.,
  152.                                                      Marchetti, R., Viviani, R. (Eds.), Marine
Peinert, R., Saure, A., Stegmann, P., Stien, C.,            Coastal Eutrophication. London: Elsevier.
  Haardt, H., and Smetacek, V. 1982. Dynamics             Wikipedia. 2014. Red tide, https://en.wikipedia.
  of primary production and sedimentation in                org/wiki/Red_tide.
  a coastal ecosystem. Netherlands Journal of             Wong, K.T., Lee, J. H., and Hodgkiss, I. J. 2007.
  Sea Research 16: 276-289.                                 A simple model for forecast of coastal algal
Rezaee, H., and Kabiri, K. 2009. Study red tide             bloom. Estuarine, Coastal and Shelf Science,
  phenomenon tracking in the Bandar Abbas                   74(1-2): 175-196.
  coastal waters and Qeshm and Hormuz
  Islands, Research Project of National Ocean
  Graphic Center.
Sadeghi Mazidi, S., Ahmadi, M.R., and
  Taherizadeh, M.R. 2011. The seasonal
  changes in phytoplankton population and
  environmental factors in winter and spring in
  Bandar Abbas coastal waters. Fisheries and
You can also read