Influence of Discharge of River Daya in Chilika Lagoon: A Brief Study using Mohid Simulation Model

pp .10-21

Subham Mukherjee

I. Introduction
Chilika, the largest brackish water lagoon of Orissa, is the hotspot of biodiversity and a Ramsar site. It’s highly productive lagoon ecosystem supports the livelihood of more than 0.2 million local communities (Panigrahi, 1998). With the passage of time due to ecological and anthropogenic pressure the lagoon ecosystem had been degrading (Figure 1).

Figure 1 : Map of Chilika Lagoon within Mahanadi river system (Source: State Pollution Control Board of Orissa, Bhubaneswar, India)
For preservation of its ecology and conservation as well and to bring an all round development in and around the lake which calls for multi-dimensional and multi- disciplinary developmental activities, the Chilika Development Authority (CDA)was created in 1991 (CDA, 2005).
Chilika eco-system is very dynamic and it is considered to be a very productive eco-system. It supports more than 200,000 people, mostly fishermen living in and around Chilika, for their livelihood. Chilika Development Authority is constantly working for the restoration and conservation of Chilika eco-system. CDA carried out several ameliorating measures for restoration of the lake based on the targeted studies to trace the root cause of the degradation of the lake. Due to successful restoration

measure Chilika was removed from the Monteux Record (threatened list of Ramsar site) in 2002. Chilika lake ecosystem is very dynamic and ecosystem and needs meticulous monitoring with all precession. This is highly essential for sustainable management of the Lake.
Integrated modelling of physical and biochemical processes serves as a powerful tool for assessing quantitatively the water quality in coastal regions (Taguchi, 1998). MOHID is a three-dimensional water modeling system, developed by MARETEC (Marine and Environmental Technology Research Center) at Instituto Superior Técnico, Lisboa, Portugal. The integration of MOHID different tools (MOHID Water, MOHID Land and MOHID Soil) can be used to study the water cycle in an integrated approach. MOHID GIS is a geographical information system specifically designed to support MOHID Water Modelling System which handles spatial and temporal variable data required or produced by the MOHID numerical programs.
The objective of the present work is to study, briefly, to monitor the most probable effect of the coliform discharge along with fresh water by the river Daya into the lagoon system and its extent through MOHID for a 3 days period, which is representative of the long-term effect of the contaminated river discharge.
II. Study Area
Chilika Lagoon is situated on the east coast of India. It is one of the largest brackish water wetland of Asia with estuarine character (Figure 1). It is the largest wintering ground for migratory water-fowl found anywhere on the Indian sub-continent. It is one of the hot spot of biodiversity, and some rare, vulnerable and endangered species listed in the IUCN Red List of threatened Animals inhabit the Lake area for at least part of their life cycle (Panigrahi, 1998).
There are 52 rivers falling on Chillika out of which River Daya is the most influential in the lagoon ecosystem. From the analysis it is observed that the water qualities of rivers falling in Chilka confirms to Class-C use category (contains more than 25000 MPN/100ml faecal coliform) (CDA, 2005).

The pear-shaped Lake is about 64.3 km long and its width varies from 18 km to 5 km (Das and Samal, 1988). The Lake was
connected to the sea by a 33 km long, irregular channel with several small sandy and usually ephemeral islands. About 1.5 km wide, the channel runs parallel to the Bay of Bengal and is separated from it by a narrow spit, 150 – 2000 m wide, the channel was connected to the sea by an extremely narrow mouth close to its northeast end during pre-new mouth period. Numerous islands are present in the Lake, especially near the channel (Mohapatra, 2005).
III. Methodology
Geographic information data set for the study area was created using MOHID GIS 9.4.2 program (Braunschweig and Fernandes, 2005).
The Satellite image and the bathymetry data of the whole Mahanadi Estuarine region were received from the Sri Subhas Chanda Santra, who is one of the Ecosystem Scientists under CDA (State Pollution Control Board of Orissa, Bhubaneswar, India). Then this image is been processed and shortened in ArcGIS 10 in order to focus only on the Chilika area (Figure 2).

The simulation using MOHID software represents the lagoon with two freshwater discharges channels (branches of River Daya) without treatment, which can be related approximately to 2000’s years. Interpolated bathymetry and the simple tide model for Mahanadi basin were used for the model. The beginning of the simulation time period was matched with the beginning of low tide, so both simulations were run for 73 hours from 13.00 of 1st October 2002 until 13.00 of 4th October 2002. The time step was 6 seconds due to small size of grid cells.
For the simulation the same atmospheric pressure and absence of wind stress were considered. Hence, Atmosphere and InterfaceWaterAir files were left empty. For Geometry file the sample data was used: 2D model of one layer from bottom to surface with sigma coordinates minimum depth of 0.1 m and code for land of 99. In InterfaceSedimentWater, the default rugosity value of 0.0025 was used for bottom fluxes. For Turbulence the given sample data of vertical viscosity of 0.001 and horizontal of 10 were used.
For Water Properties value of density was 1026.72546 mg/m3 and only one water property of salinity was used with default initial values of 36 and value for boundary was 36 as well. For Hydrodynamic a simple model was used with default values with barotrophic conditions, tides, no radiation of waves, upwind condition, implicit vertical advection and diffusion, Abbot discretization for coastal water, Coriolis force, horizontal and vertical advection and diffusion, cold start without initial velocities, no tide potential, no wind, no rain and no difference in atmospheric pressure, bottom stress of 0.1, no momentum discharge, usual start, minimum horizontal advection of 0.5, local density calculation, no submodels, no ground water bottom fluxes, no slipping condition, hydrostatic condition, no barotrophic force at the boundary, and the preliminary created time series.
For the simulation two water discharges were modelled in the lagoon, which are the channels of river Daya. Therefore, in Discharges file two discharges were included, both of which had a width of one grid cells. For the both freshwater discharges approximate flow values of 90 m3/s was assigned. In Discharges only one water property of salinity with constant value of 0.0001 for fresh water and with possibility of alternative location because of gently sloping shore around the lagoon were used. Eulerian observational method was used for water property of salinity and Lagrangian observational method for faecal coliform.
In Lagrangian, two origins represent two freshwater discharges with coliform with point spatial emission, continuous temporal emission,flow values of 90 m3/s for the each origin as it has in Discharges, default sample values of turbulence, advection, trajectory step the same as the time step, 1 particle, depth cells of 0.5, time of doubling of volume of particles of 7200 seconds, velocity way of volume increasing, volume factor for particle deletion of 1000, only one water property of feacal coliforms with concentration of 25000 MPN/100 ml in the each of the two origins, time of 7200 seconds to inactivate 90 % of coliform, and considering that ambient water was clean. The value of 25000 MPN/100 ml was selected as mean usual coliform concentration in river Daya already been recorded by the Central Laboratory, State Pollution Control Board of Orissa, Bhubaneswar, India. The same output time interval of half an hour in Hydrodynamic, Water properties and Lagrangian models were calculated.
IV. Results and Discussions
Courant number of the model was 0.5653 due to decreasing the time step to 6 second and some default boundary changes according to the study area. It should be less than 1 for stability (Abbott and Basco, 1989), so the model is completely stable with this time step. The following figure shows the initial of the simaulation model(Figure 3).

Figure 3 : Initial of the Simulation model with the different scales for different parameters used

Hydrodynamic of the study area is represented on Figure 3 with the time interval of 1 hour. The initial time with the zero velocities is 13.00; the simulation is started with the beginning of the low tide. The low tide is finished with the almost motionless water at 19.00. High tide appears after 19.00 and lasts until 01.00 at the final figure. So during the modelled 72 hours there are 6 tidal cycles: low tide, high tide, then again low tide, high tide and so on. Velocity modulus in the lagoon varies between 0 and 0.1 m/s. They are represented by graduated colours (Figure 3). Arrows represent vectors of velocity and vector scaling is 1 m/s = 500 m, so vectors of 0.1 m/s are represented by arrows of 50 m. The higher velocities occur in the narrow channels, with the highest velocity in the entrance of the lagoon. The only one cycle of diurnal tide is shown on the Figure 4. The second one finishes at 13.00 of the next day. Comparing this tide cycle with the one in presence of the 2 freshwater discharges, the difference in velocity modulus can be concerned as negligible. Hence, the freshwater discharges don’t influence significantly the hydrodynamics of the system.
The change of salinity is shown in accordance with tides and freshwater discharge from the river (Figure 4). Salinity is shown for every three hour from 13.00 to 13.00 next day. The arrows show the velocities of water. It can be observed from the Figure 4, that the salinity in the river of Daya does not change with accordance to the tidal cycle in the lagoon system. In the river near to the lagoon the lowest salinity of the water is ca 20. This value is quite high probably due to the high tides pushing salty water upstream two times every day.

A) at 3rd hour, during the start of high tide

B) at 9th hour, during the peak of high tide

C) at 15th hour, during the peak of Low tide
Figure 4. Distribution and change of salinity levels due to the freshwater discharge into the lagoon during high and low tides
Figure 4 represents salinity at the end of modelled twenty-four hours (13.00 of the next day). At the right corner of the map, salinity in the lagoon is shown with the presence of the two freshwater discharges.
It can be noticed that water in the lagoon was slightly desalinated due to the freshwater discharge from these two channels of River Daya. It means that the two freshwater discharges influenced salinity but very little.
Figure 5 demonstrates the distribution of Coliform bacteria which were released in the wastewater by two discharges in the corners of the lagoon. The distribution is shown during the one tide cycle. The low tide comes firstly, then the high tide. The first picture is after one hour of the beginning of discharge of freshwater with coliform contamination; the next one is after 6 hour at the time of high tide, the rest one is at the 24th hour. The colour scale is from 20 MPN/100 ml (blue) to 25000 MPN/100 ml (red). According to CDA (2005), Pollution level A denotes 20 MPN/100 ml and the ‘C’ as 25000 MPN/100 ml.

Figure 5: Discharge of Coliform contaminated water from the channels of river Daya at 2nd (A), 6th (B) and 24th (C) hour.

Figure 6 shows the scenario of the lagoon at 48th hour and 72nd hours. It can be observed, that the bacteria occupancy is increasing from a small to bigger part of the lagoon area and are distributed within the fresh water influenced areas of the lagoon.

Figure 6. Tidal, Saline and Coliform distribution at the 48th (A) and 72nd (B) hours of the simulation

Now the bacteria occupy only the mouth of the river and are propagated to the entrance of the lagoon during the low tides. The most of the area of the lagoon is clean.
Chilika receives fresh water through 52 rivers and rivulets, amongst which Daya is the most impacting river which accounts for the maximum sediment pumped into the Chilika Lake. It has been estimated that around 3.5 lakh tons of sediment enter into the Lake every year posing continuous threat to the shrinkage of water spread area of the Lake (CDA, 2005).
A 32 km long narrow channel is connecting the main Lake to the Bay of Bengal, near the village Arakhakuda. The mouth connecting the channel to the sea is close to the north eastern end. The above figures show that high tides near this inlet mouth uses to drive in salt water through the channel during this 3 days simulation period. With the onset of the rains, the rivers falling into the northern zone are in spate, causing fresh water currents which gradually push the sea water out. As a result of these dynamics, the inlet mouth would constantly change position. Previous studies in this area (Mohapatra et al, 2005) reveal that due to shifting of the mouth, proper exchange of water is not taking place and due to choking of outer channel and its mouth opening into the Sea, the exchange of water between the Sea and the Lagoon is decreasing and thereby the salinity level of the Lake is gradually decreasing affecting the biodiversity of the Lagoon system.
The above figures clearly indicate the sediment load along with contamination with faecal coliform within the lagoon system. It is supported by Mohapatra et al (2005) who showed that the Mahanadi river system contributes major sediment flow to the lagoon through river Daya, mainly. It has a direct influence from the Hirakund dam, as the sediment flow to the lagoon is increasing considerably.
Regarding coliform load, it is clear that the river daya has an influencial role in the lagoonal system since it carries a substantial load of coliform. As In absence of sewerage system, people are using septic tanks and soak pits. In most of the places, where it drains (mainly Bhubaneswar city), sewage is discharged in to open drains without any treatment, which joins to form Gangua Nallah, near Bhubaneswar city upstream, and ultimately discharges to river Daya (Panigrahi, 1998). The Gangua Nallah joins river Daya near Birimula Village about one kilometer down Stream of Kukuria Bridge. Gangua nallah which finally meets River Daya serves as the ultimate for the wastewater discharges of Bhubaneswar city (CDA, 2005).
V. Conclusion
The MOHID Water model of the Chilika Lagoon was created for showing the influence and discharge of freshwater contaminated with the intrusion of faecal coliform from the river of Daya. Lagrangian observational method was used for propagation and transport of the faecal coliforms by the tide currents and Eulerian method for the distribution of salinity.
Hydrologically, Chilika is influenced by three subsystems; the Mahanadi distributaries, the rivers/ streams draining in to the Lagoon from the western catchment (rivers i.e., Godavari mainly) and the Bay of Bengal. Ecologically, Chilika is an assemblage of marine, brackish and freshwater eco-systems which supports a diverse and dynamic assemblage of plants and animal. Salinity is the most dominant factor determining the Lagoon’s ecology and the salinity dynamics are controlled jointly by the nature of the connection to the sea, associated tidal fluctuations, and the volume and timing of freshwater inflows to the Lake from the delta distributaries and western catchments.
The influence of the discharges on salinity, tide cycle and distribution of coliform bacteria was analysed. Related to the tide cycle, the wastewater discharges don’t affect velocities of water movement, but slightly decrease salinity. The freshwater discharges have the most significant impact on the distribution of the coliform bacteria. The bacteria occupy gradually a significant part of the lagoon though the tides don’t have a big influence on the distribution of the coliforms in the lagoon.

As it can be observed from the figures, the faecal coliforms propagate not that far, how it can be expected in the real situation. The most probable reason is a small period of coliform inactivation. It is necessary to validate coliform parameters such as time of doubling of particle volume, time needed to inactivate 90 % of coliforms and volume factor for particle deletion. Instead of simple time parameters there can be a model considering water quality because the bacteria are sensible to pH level of the water.
In the maps some changes in velocities and salinity close to the land-sea boundary can be noticed. One way to solve this boundary problem is to expand the grid further to the water body.
The MOHID Model simulates hydrodynamics and different kinds of water properties rather well by the both Lagrangian and Eulerian methods. Even in such large scale it is possible to have a stable model with very small time step, but in this case the time of calculation greatly increases. This comparative analysis can be repeated more efficiently with real data of sediment data, river flow, discharges flows and coliform concentrations. MOHID simulations can be used for deeper understanding of the processes and for sustainable management. MOHID Models can be presented to decision makers to better understand the influence of the measures they are going to take to increase the sustainability of this major ecosystem. Constant monitoring and detailed studies can only mitigate the issues, regarding the existing water pollution levels, poor sanitation, lack of sewerage system, polluted drains and river, overflowing sewage, which are highly deterrent to the tourism activity and for the health of the local people and the biodiversity of Chilika lagoon.

VI. References
Abbott, M. and D. Basco. 1989. Computational fluid dynamics: an introduction for engineers. Longman Scientific & Technical. London.
Braunschweig, F. and L. Fernandes. 2005. MOHID Graphical User Interfaces. User Manual. Instituto Superior Técnico, Technical University of Lisbon.

CDA Brochure, 2005. Chilika, the living resource. State Pollution Control Board of Orissa, Bhubaneswar, India.
Council Directive of 8 December 1975 concerning the quality of bathing water (76/160/EEC), Official Journal of European Communities, L 31/1.
Mohapatra, M., Mohanty, U.C. 2005. Some characteristics of very heavy rainfall over Orissa during summer monsoon season. J.Earth Syst. Sci.,, 114(1): 17-36.
Panigrahi, A.K. 1998, Chilika Lake – An Overview. Proc of the International Workshop on Sustainable Development of Chilika Lagoon, 1998, Bhubaneswar, India.
Taguchi, K., Nakata K. 1998. Analysis of water quality in Lake Hamana using a coupled physical and biochemical model, Journal of Marine Systems 16: 107–132.
Trimétrica – Engenharia, Lda. 2006. URL