Wave Transformation along Southwest coast of India using MIKE 21

23 Wave Transformation along Southwest coast of India using MIKE 21 Parvathy K.G. 1, Deepthi I. Gopinath 2, Noujas V. 3 and K. V. Thomas 3 1 National Institute of Technology Karnataka, Surathkal, 575025, India, pkg130590@gmail.com 2 National Institute of Technology Karnataka, Surathkal, 575025, India 3 Centre for Earth Science Studies, Thiruvananthapuram, 695031, India Received: Feb. 28, 2013; Accepted: March 16, 2014 Abstract Nearshore wave transformation is a complex coastal process of shoaling, refraction, diffraction, reflection, and energy dissipation due to bed friction and breaking contributing variations in the wave height, period and direction. A well defined sediment cell of about 45 km extending from Kovalam headland to Varkala cliff which forms a part of Thiruvananthapuram coast along the southwest coast of India, is selected for the wave transformation studies. In the present study MIKE 21 Spectral Wave model (DHI, 2011) was used. The model simulates the growth, decay and transformation of wind generated waves and swells both in offshore and coastal areas. Providing MIKE 21 SW with a suitable bathymetry is essential for obtaining reliable results from the model. Usually the offshore bathymetry is derived from C-MAP, ETOPO, GEBCO etc. and the nearshore bathymetry is generated from close grid bathymetric surveys. In this study offshore bathymetry was generated from GEBCO-08 grid which is a freely available software with 30 arc (~ 1 km) resolution. In the nearshore zone, surveyed close grid bathymetric data were used. The other inputs such as wave measurements and wind data provided in the model were from observations in Lakshadweep Sea. Model result is calibrated with field observations along this sector. The model has efficiently simulated the process of shoaling and refraction along the coast. The percentage of observed shoaling is 12.7% at a distance of 24 km from the shoreline at a depth of 70m and it was seen to be increasing to 27.9% when it reached around 2.4 km from the shore at a depth of 10m. The model result also shows that the wave is almost aligned parallel to the coast as wave approaches the coast. This model result can be used for further applications in designing along this coast. Key words: Coastal process, Wave transformation, Shoaling, Refraction 1. INTRODUCTION Coastal zone is the triple interface of land, ocean and atmosphere. Any developmental activity along the coastal zone requires a clear understanding of the dynamic processes controlling its very existence. The definition of wave conditions in shallow water is an essential requirement in the design and operation of a wide variety of coastal facilities and for the safe performance of human activities in coastal areas. Port developments, marine terminals, coastal management strategies, downtime analysis of vessel activities, etc. are a few examples for which the assessment of wave conditions is important (Fassardi, 2004). Engineers are usually confronted with insufficient or no wave data when performing studies in remote or under developed coastal areas. Measurements at the site of interest would be the most adequate source of data, but these are often expensive. Considering the difficulties involved in field measurement, theoretical and modeling approaches present an alternative. Currently, various commercial and open source spectral wave models are available for wave hindcast and forecast studies. Volume 5 Number 1 2014

24 Wave Transformation along Southwest coast of India using MIKE 21 Nearshore wave transformation is a complex coastal process of shoaling, refraction, diffraction, reflection and energy dissipation due to bed friction and breaking, contributing to changes in the wave height, period and direction. Unlike in deep water, the wave climate in shallow water exhibits much spatial variation owing to the complex transformation processes (Kurian, 1989). Waves moving towards shoaling water at an angle to the bottom contours, change their direction of progress and become parallel or almost parallel to the shore before breaking, which is the wave refraction. Wave refraction is caused by the dependence of wave speed upon water depth in the regions where water depth is less than half the wave length. For planning constructions on the beach for development and tourism, for establishing a harbour and also for evolving coastal zone management strategies, the knowledge of wave transformation at the particular shoreline is very essential. Numerous studies have been carried out to understand the wave transformation process. Hubertz (1981) discussed two numerical models which allow a wave field to be transformed by the process of refraction and shoaling from offshore to nearshore. Fassardi (2004) used MIKE 21 Parabolic Mild Slope (PMS), developed by the Danish Hydraulic Institute (DHI) as wave transformation model and a simple methodology for the derivation of long-term hindcasts in shallow water including the particular features of the nearby bathymetry. Aboobacker (2010) attempted to study seasonal response of coastal waves and wave transformation along the Indian coast using measurement, modelling and remote sensing. This paper aims to study the nearshore wave transformation process along Thiruvananthapuram coast using MIKE 21, a third generation commercial model. In the present study numerical simulations of waves in the deep waters as well as shallow waters have been carried out using MIKE 21 SW models. 2. STUDY AREA The state of Kerala is situated along the south west coast of India. It stretches along the coast of the Lakshadweep Sea. The state lies between 8 18 and 12 48 north latitude and 74 52 and 77 22 east longitude. The total length of coastline is about 570 km of which about 569.7 km is affected by erosion (Sanil Kumar et.al., 2006). The coast of Kerala is subjected to coastal erosion which is perennial in nature. The magnitude of erosion is high during the south-west monsoon season (June September) causing irrepressible loss/damage to the property adjacent to the shoreline. Different types of coastal protective measures have been implemented along the Kerala coast during the last century. Of the various protective measures adopted so far, construction of seawalls is the most commonly used method. Various designs of seawall have been attempted along different stretches of the coast covering about 75% of Kerala s coastline. In spite of this, erosion is continuing even at many locations which have been protected by seawalls. The enormous cost of maintenances and frequent failure of seawalls is a major concern for the Government (Ajeesh 2011). It is understood that the coast has severe erosion problems due to enhanced erosion due to human interventions like breakwaters, groins, seawalls, etc. It calls for redesign in certain sectors and new interventions in other sectors for which better data and knowledge of the dynamic processes along the coast is a prerequisite. Thiruvananthapuram coast has a major sediment cell between Varkala cliff to Kovalam headland. Veli and Varkala could form a minor sediment cell within the Kovalam-Varkala sediment cell (Ajeesh 2011) The Kovalam (8 57 59 N, 75 37 21 E) to Varkala (8 04 49 N, 77 19 13 E) coastal stretch (Figure 1) which forms a part of Thiruvananthapuram coast along the southwest coast of India, was selected for the wave transformation studies. The Kovalam-Varkala coastline is oriented in NNE SSW direction. The coastal plains extending north from Panathura to Anchuthengu are extensively wide landwards. The tidal inlet at Panathura is about 3 km north of Kovalam headland. Seawalls protect the entire coast from Kovalam headland to Panathura inlet and two groins presents at Panathura having length 25 and 33m. The coastal stretch immediately north of the inlet is an open coast with an existing groin field at Poonthura about 1 km north of inlet. Further north, the Poonthura - Beemapally sector of the coast has seawalls without frontal beach. Valiyathura is located about 6 km north of Panathura. This sector of the beach has a fairly wide fair-weather beach seaward of seawalls. There is a pier at Valiyathura which stands on pillars and does not significantly affect the sediment transport. A coastal International Journal of Ocean and Climate Systems

Parvathy K.G., Deepthi I. Gopinath, Noujas V. and Thomas K. V 25 stretch of about 300m north of northern breakwater of Muthalapozhi harbour is an unprotected beach. North of the above stretch of the coastline is being protected by seawall. The Perumathura coast and Thazhampally-Anchuthengu coast constitute barrier beaches. Figure 1 Kovalam to Varkala coastal sector 3. DATA AND METHODOLOGY Winds, waves and bathymetry are the major oceanographic parameters required for wave modeling (Aboobacker, 2010). 3.1. Bathymetry The accuracy of all numerical simulations primarily depends on how accurately the bathymetry is generated. The bathymetry for the simulation of waves was generated from the General Bathymetric Chart of Oceans (GEBCO_08 Grid). It is a global terrain model for ocean and land at 30 arc-second intervals (Appendix 1). The data within the GEBCO_08 grid represent elevations in meters, with negative values for bathymetric depths and positive values for topographic heights. The regional bathymetry obtained from GEBCO_08 Grid is illustrated in Figure 2 Volume 5 Number 1 2014

26 Wave Transformation along Southwest coast of India using MIKE 21 Figure 2 Regional Bathymetry obtained from GEBCO Since GEBCO has concentrated upon the interpretation of bathymetry in deep water, standard contours have been fixed at intervals of 500m with 200m and 100m in shallow zones. Therefore in the near shore zone from St. Andrews to Nedunganda (around 18 km) the bathymetric data collected by Centre for Earth Science Studies (CESS) was adopted. The survey was carried out along transects normal to the shore extending up to 20 m depth in a closed interval. The echo sounder was integrated with GPS (Global Positioning System) for accurate positions. The data were collected and interpolated using MIKE 21 Mesh Generator for obtaining the continuous bathymetry of the area (Figure 3). Figure 3. Bathymetry along study region International Journal of Ocean and Climate Systems

Parvathy K.G., Deepthi I. Gopinath, Noujas V. and Thomas K. V 27 3.2. Wave data In the present study, a continuous wave data of eight months collected in the Off-Mangalore region using Directional wave rider buoy was used which was made available from NIOT Chennai. It is important to ensure that reliable data are used to initialize the numerical models. The wave data used is represented as time series plot (Figure 4 a, b and c). Figure 4.(a) Significant wave height off-mangalore Figure 4 (b) Peak wave period off-mangalore Figure 4 (c) Mean wave direction off-mangalore 3.3. Wind Data Wind data used in modeling was also obtained from the data collected by NIOT Chennai. The wind data used is represented as time series plot (Figure 5 (a) and (b)). Volume 5 Number 1 2014

28 Wave Transformation along Southwest coast of India using MIKE 21 Figure 5(a) Wind speed off-mangalore v Figure 5(b) Wind Direction off-mangalore 3.4. Numerical Modeling Numerical simulations have been carried out to study wave features prevailing along the coast. Wave parameters obtained from the simulations are analyzed to understand the wave transformation phenomena. For simulating the wave transformation from the deep water to the nearshore region, the Spectral Wave (SW) model of MIKE 21 was used. The model is capable of simulating the growth, decay, and transformation of wind generated waves and swell in both offshore and coastal region. MIKE 21 SW is a new generation spectral wind wave model based on unstructured mesh which takes into account various phenomena like wave growth by influence of wind, nonlinear wave-wave interaction, dissipation due to white capping, bottom friction and depth induced breaking (DHI,2011). 3.4.1 Model Domain With a view to understand the wave transformation off the Thiruvananthapuram Coast along the West coast of India, a 45km stretch of coast between Kovalam and Varkala is considered. The coast line is oriented along NNW-SSE direction. Providing MIKE21 with a suitable mesh is essential for obtaining reliable results from the model. Setting up the mesh includes selection of appropriate area to be modeled, adequate resolution of the bathymetry, wave, wind and flow field under consideration and definition of codes for open and land boundaries. MIKE 21 is based on flexible mesh approach. In order to study wave transformation in Thiruvananthapuram coast, a large domain, having a water depth around 700m is selected. An unstructured triangulated mesh is generated with varying sizes of triangles (elements) with finer triangles on nearshore area and coarser triangles on offshore area (Figure 6). International Journal of Ocean and Climate Systems

Parvathy K.G., Deepthi I. Gopinath, Noujas V. and Thomas K. V 29 The mesh file is generated in the MIKE Zero Mesh Generator which is a tool for the generation and handling of unstructured meshes including the definition and editing of the boundaries. The mesh file is an ASCII file including information of the geographical position and bathymetry for each node point in the mesh. The file also includes information of the node connectivity in the triangular element mesh. The mesh file generated in the present study has 2781 nodes and 4515 elements. Figure 6 Model domain used for wave transformation In the present case the model was set up with three boundaries north, south and west, along with a land boundary on the eastern side. For defining the western boundary offshore wave data consisting of significant wave height, mean wave height, mean wave direction, peak wave period and directional spectral index were given as input parameters. The other two open boundaries (northern and southern) were defined as lateral boundaries. 4. RESULTS AND DISCUSSIONS Several locations of Thiruvananthapuram coast are under severe erosion. In order to address this problem various coastal engineering measures have to be adopted. But a lack of proper data, particularly wave data, required for design purposes is badly felt. Many coastal protection designs now in place have not given the expected results. Hence the present numerical model has been set to study the wave transformation along the coast. The results from the numerical simulation studies were analyzed to bring out the wave transformation along Thiruvananthapuram coast. From the statistical mean (Figure 7) the value of the significant wave height for the study area was found to range between 0.9m and 1.5 m. From the statistical mean (Figure 8) the value of the mean wave direction for the study area was found to range between 190 to 285. The waves are generally high from May to October while they were low during the rest of the months. It is observed that significant wave height is less than 1.5 m during January to April. During June to September it is usually above 1.5 and 2m.The observation agree with earlier studies [Thomas et al 2007, Thomas et al 1983, Swamy 1979]. Volume 5 Number 1 2014

30 Wave Transformation along Southwest coast of India using MIKE 21 Figure 7 Statistical mean of significant wave height Figure 8 Statistical mean of mean wave direction 4.1. Shoaling To study the shoaling process, the spectral output obtained for significant wave height is analyzed and variation of significant wave height with respect to water depth is plotted. From the graphs (Figure 9) it was observed that the wave transformation occurs approximately at a water depth of 100m.The theoretical relations were used to validate this data. The mean significant wave height is observed to be 1.58m. International Journal of Ocean and Climate Systems

Parvathy K.G., Deepthi I. Gopinath, Noujas V. and Thomas K. V 31 Figure 9 Significant wave height at 300m, 200m, 100 m and 10m water depths Table 1 Statistics of Significant wave height Wa ter depth, d (m) Mean Significant height (m) Rati o % shoaling 300 1.48 93.6 6.4 200 1.48 93.6 6.4 100 1.45 91.7 8.3 70 1.38 87.3 12.7 60 1.35 85.4 14.6 50 1.32 83.5 16.5 40 1.25 79.1 20.9 30 1.23 77.8 22.2 20 1.2 75.9 24.1 10 1.14 72.1 27.9 The statistics of significant wave height (Table 1) clearly shows that the percentage of shoaling is 12.7% at a depth of 70m (offshore) and it increases to 27.9% at a depth of 10m i.e. 2.4 km away from the shore. 4.2 Wave Refraction The spectral output of mean wave direction obtained from the model is analyzed for refraction studies. From the graphs plotted it was found that the wave direction changes from deep water to shallow water and it tries to get aligned with the shoreline as these waves approach the shore. Figure 10 shows variation of mean wave direction. Volume 5 Number 1 2014

32 Wave Transformation along Southwest coast of India using MIKE 21 Figure 10 Mean wave direction at 100m,10m, 1m and 0.5 m water depths The Kovalam - Varkala shoreline is oriented at an angle of approximately 232 degrees. The wave direction observed at some points nearshore is also around 232 degrees. Hence it is observed that refraction is taking place along the coastal stretch as the wave try to align with the orientation of the coast. Table 2 clearly shows that as wave approaches to nearshore it tries to get aligned with the shore. Table 2 Mean wave direction at different water depth Water Depth(m) Mean Wave 400 281.96 300 282.13 200 283.81 100 285.25 70 285.07 60 284.21 50 283.03 40 278.38 30 278.20 20 273.91 10 265.18 1 232.70 Shoreline ~ 232.00 International Journal of Ocean and Climate Systems

Parvathy K.G., Deepthi I. Gopinath, Noujas V. and Thomas K. V 33 5. CONCLUSIONS The nearshore wave transformation along Thiruvananthapuram coast has been studied using the MIKE 21 SW. The spectral output obtained from MIKE 21 SW was analyzed for shoaling and refraction phenomena taking place along the coast. The percentage of shoaling is observed to be 12.7% at a distance of 24 km from the shoreline at a depth of 70m and it was seen to be increases to 27.9% when it reached around 2.4 km from the shore at a depth of 10m. The model result also shows that the wave is almost aligned parallel to the coast as wave approaches the coast. For most of the coastal engineering applications clear knowledge on wave transformation is a prerequisite. So the present model can be further used for structural designing along the coast. ACKNOWLEDGEMENT The authors are thankful to Dr. N.P. Kurian, Director, Centre for Earth Science Studies, Thiruvananthapuram for his encouragement and support. The NIOT, Chennai and ICMAM Project Directorate, Chennai are acknowledged for the data on waves and wind. APPENDIX 1 Gebco 08 Grid GEBCO_08 Grid is a global terrain model for ocean and land at 30 arc-second intervals. It was released in May 2009 and updated in November 2009 and September 2010. The grid was generated by combining quality-controlled ship depth soundings with interpolation between sounding points guided by satellite-derived gravity data. The data values within the GEBCO_08 Grid represent elevations in metres, with negative values for bathymetric depths and positive values for topographic heights. Getting started with GEBCO the display of the splash screen and information check box, the chart Definition Dialog Box is displayed. From here, the user can select the geographic area required, the required map projection, the colour palette to be used to display the data. We have the latitude and longitude of the domain. From the Data source tab of the Chart Definition Dialog Box or from the Display Gridded Data menu option on the main toolbar. The name of the gridded dataset is displayed in the caption bar of the main menu toolbar. To export the data for the area drawn to screen, use the menu options from the main toolbar, File Export Data Gridded Data. The limitations of GEBCO_08 Grid are as follows. GEBCO has concentrated upon the interpretation of bathymetry in deep water. Standard contours have been fixed at intervals of 500m with a 200m and 100m in shallow zones. Plotting the grid depths as a histogram reveals numerous peaks each of which occurs at a multiple of 500m. This terracing effect is a well-known problem of constructing grids from contours. Compared to total-coverage, high-resolution multi-beam bathymetric surveys it will likely to prove unjustifiable to plot the GEBCO global grid in fine contour interval. No thorough error-checking was performed for land areas. The very high density of globe land elevation data was taken as accurate. The grid is not intended to be used for navigational purposes. REFERENCES Aboobacker V.M. (2010) Wave transformation at select locations along the Indian coast through measurements, modelling and remote sensing, Ph D Thesis, National Institute of Oceanography (CSIR), Dona Paula, Goa 403004, India. Ajeesh N.R. (2011) Numerical Modeling Studies on Coastline Evolution off PanathuraCoast, M.Tech. Thesis, National Institute of Technology, Surathkal. Danish Hydraulic Institute (DHI) (2011) MIKE 21 Spectral wave Model FM User Guide (MIKE 21), DHI, Holsholm, Denmark. Volume 5 Number 1 2014

34 Wave Transformation along Southwest coast of India using MIKE 21 Fassardi C. (2004) The transformation of deep water wave hindcasts to shallow water,proceedings of 8th International Workshop on Wave Hindcasting and Forecasting-2004, North Shore, Hawaii,(2004). Hubertz J. M. (1981) Prediction of Wave Refraction and Shoaling using two Numerical Models, In. Coastal Engineering Technical Aid No. 81-12, U. S. Army Corps of Engineers, Coastal Engineering Research Centre, Kingman Builiding, Fort Belvoir. Kurian N. P. (1989) Shallow water wave transformation In: Ocean Wave Studies and Applications (Ed.M.Baba and T.S.Shahul Hameed), pp.15 32. Centre for Earth Science Studies, Thiruvananthapuram. Parvathy.K.G, Deepthi I Gopinath & Noujas V. (2013) Wave Transformation along Thiruvananthapuram Coast using MIKE 21, in Proc. National Conference of Ocean Society of India (OSICON 13), Role of Ocans in Earth System, Baba M & Jayachandran K V (Eds), pp 193-197. Pranav N. Desai & K.V. Radhakrishnan (2003) Science, Technology, Coastal zone Management and Policy Allied Publishers Pvt. Ltd., A-104 Mayapuri, Phase II, New Delhi-110 064 Sanil Kumar V, Pathak K C, Pednekar P, Raju N S N, Gowthaman R. 2006. Coastal processes along the Indian coastline, Current Science, Vol.91, No.4, pp. 530-536. Swamy G. N., Varma P. U., Abraham Pylee, Rama Raju V.S., Chandra Mohan P. (1979), Wave Climate off Trivandrum (Kerala), Indian Journal of Marine Science 12(3), 127-133 Thomas K.V. and Baba M. (1983), Wave Climate off Valiathura,Trivandrum, Mahasagar-Bulletin of National Institute of Oceanography, Vol 16, pp 415-421. Thomas K.V., Shahul Hameed T.S., Baba.M and Kurian N.P. (2007) Long Term Wave Climatology of the South West Coast of India, Proceedings of Second Indian National Conference on Harbour and Ocean Engineering, Thiruvananthapuram, 1301-1310. International Journal of Ocean and Climate Systems