Numerical modeling of coastal inundation in response to a tropical cyclone along the east coast of India

(1)

NUMERICAL MODELING OF COASTAL INUNDATION IN RESPONSE TO A TROPICAL CYCLONE ALONG THE EAST

COAST OF INDIA

SMITA PANDEY (NEE. MISHRA)

CENTRE FOR ATMOSPHERIC SCIENCES INDIAN INSTITUTE OF TECHNOLOGY DELHI

OCTOBER 2020

(2)

©Indian Institute of Technology Delhi (IITD), New Delhi, 2020

(3)

NUMERICAL MODELING OF COASTAL INUNDATION IN RESPONSE TO A TROPICAL CYCLONE ALONG THE EAST

COAST OF INDIA

by

SMITA PANDEY (NEE. MISHRA) Centre for Atmospheric Sciences

Submitted

in fulfilment of the requirements of the degree of DOCTOR OF PHILOSOPHY

to the

INDIAN INSTITUTE OF TECHNOLOGY DELHI

OCTOBER 2020

(4)

Dedicated to my family

(5)

Certificate

This is to certify that the thesis entitled "Numerical modeling of coastal inundation in response to a tropical cyclone along the east coast of India"

being submitted by Ms. Smita Pandey (nee Mishra) to the Indian Institute of Technology Delhi for the award of the degree of DOCTOR OF PHILOSOPHY is a record of original bonafide research carried out by her. Ms. Smita has worked under my guidance and supervision and has fulfilled the requirements for the submission of this thesis. The results contained in this thesis have not been submitted in part or full to any other University or Institute for the award of any degree or diploma.

(Dr A. D. Rao)

Professor, Centre for Atmospheric Sciences Indian Institute of Technology Delhi

New Delhi-110016.

(6)

Acknowledgements

Idea of doctoral degree and its conceptualization, prepares an individual for a new and different journey with all kind of uncertainties, challenges and excitement of a fresh work. After reaching this stage of completing my PhD thesis reminds me of all kinds of good enthralling times mixed with very shabby and insurmountable spots where we feel helpless. Then propelling force of some unknown enlightenment of energy creating moments of magic and showing bright paths, generated much more believe in almighty. I bow myself to all invisible positivity I experienced through my journey of the doctoral degree work.

It is an opportunity for me to express my feelings and sense of gratitude to my esteemed supervisor Prof A. D. Rao, for his insightful supervision, progressive outlook, consistent help and energy towards research. He motivated me at every moment of this period and without his sustained encouragements, numerous lively discussions and suggestions, it could have been a difficult journey to complete. His energy and enthusiasm towards work and life is un-matched and that inspired me at moments of despair to stand and think beyond horizon. I felt him a guardian and a mentor always motivating to thrive for bigger heights in academic life. I feel grateful of him for all his kind support and guidance.

At various moments (beginning from PhD course work), I have been supported with full help and guidance from Prof. Manju Mohan, Head, Centre for Atmospheric Sciences (CAS) and I would convey my deep gratitude to every faculty of the CAS IITD, particularly Prof. Krishna Mirle AchutaRao and Prof. Vimlesh Pant for their scientific inputs which helped me to shape my thesis work. I am also grateful to all supportive and friendly staff of the CAS, library, administration as well as HPC facility (for computational resources), which made way easy to work and focus towards the research goal. It is moment to feel great being a proud member of the IIT, Delhi and

(7)

would like to convey my heartfelt gratitude towards the Esteemed Institute. It is again feel exhilarating about the time I spent with my dear friends Sachiko, Tanuja, Ravi, Sandeep, Vidya, Pooja, Jismy, etc, who were the source of positivity, support, care, inspiration, co-operations and a feeling of family at office, which made me sail through all challenging times and moments of cheer are well etched in memory to last forever. It is my great pleasure to express thanks from deep of my heart to Raktim to provide me learning of the hydraulic model and science behind it.

The person I am today is the upbringing, values and care given by my parents, Mr. Rajendra Kumar Mishra and Mrs. Saroj Mishra, they have been always inspirational for me. Also I can never forget the unconditional love and support given to me by my in-laws Mr. Ashok Kumar Pandey and Mrs. Ramavati Pandey. I convey my love and seek their infinite blessing ever in my life, which gives me energy to move forward with positivity. It seems that my lovely daughter Advita too finished her PhD by knowing all kinds of intricate words (submitting research paper, working of model etc.) of the research work, as she grew up with my PhD. But she has been a fantastic kid and I had no trouble caring her and completing my work. At every evening her glowing face lead me to a more energetic person. And most of all, I convey my thanks from the deep of my heart to my husband Rakesh for patience, care and support throughout my PhD journey. After my parents he was the person who encouraged me a lot to join PhD. Finally, there would be numerous people through this time without their help and support (directly/indirectly), I could not complete my work and I convey my sincere thanks to all these friends and well-wishers.

Smita Pandey New Delhi

(8)

i

Abstract

The present thesis uses ADvanced CIRCulation (ADCIRC) model to compute coastal inundation due to combined effect of storm surges and tides along the east coast of India.

Simulations are made to compute coastal inundation due to storm tides at every 10 km along the coast by using synthetic cyclonic tracks constructed based on the past cyclone data. Thecyclonic winds are computed using the maximum pressure-drop of the cyclone based on 100-year return period. The coast is mapped for the maximum possible extent of inland inundation with water levels at district level. Peak water levels of about 10m are found along the north of Odisha coast.

The most vulnerable regionin terms of coastal inundation is found in the districts of West Bengal with a maximum extent of about 130km interior. In Andhra Pradesh, the maximum inundated districts are Nellore, Prakasam, Guntur, Krishna and East Godavari. Though the water levels in the Ramanathapuram district in Tamil Nadu reaches more than 8 m, the region is unaffected by the coastal inundation due to high local topography. By examining the inundated area of different water levels, it is seen that more than 75% of the total area is inundated with greater than 2 m water levels in the northern districts of Odisha and Ramanathapuram district in Tamil Nadu.

In recent times, the east coast of India is threatened by cyclones making landfall from different directions. Hence, a detailed investigation is made on the effect of cyclone’s approach angle on generation of storm surges and its non-linear interaction with tides and wind waves using a standalone ADCIRC and a coupled ADCIRC+SWAN model. Numerical experiments are executed using 17 idealized straight cyclone tracks of same intensity moving at constant forward speed. The study domain considers an idealized bathymetry with a straight coastline and uniform shelf width. All cyclones make landfall at the same location with angles approaching from 10° to 170° with an increment of 10°. The simulations show that the maximum storm surge is computed

(9)

ii

for tracks land falling with 60ô-90ô. Increase in the peak storm surge is seen about 15% from 10ô- 60ô, while it decreases by about 13% from 90ô-170ô. Surge-wave interaction modifies the maximum water levels by about 21-26%. The maximum wave contribution is seen for a track with 90ô followed by 160ô and 20ô. During different phases of tide, surge-tide interaction modulates the peak surge, its occurrence and location as cyclone makes landfall at different angles. Extent of affected coastal stretch is maximum on either side of the landfall for the tracks moving close to the coast, while it is minimum for the perpendicular track (90ô), confined only to the right of the landfall. The peak surge-tide-wind wave interaction along the coast at both high and low-tide is seen about 2-4h after the landfall. The interaction along the coast depends on approach angles of the cyclone, however the total water elevation (TWE) is mainly modified by both tidal phase and approach angle.

The rise of TWE at the coast is caused primarily by three factors that encompass storm surges, tides and wind waves. The accuracy of TWE forecast depends not only on the cyclonic track and its intensity, but also on the spatial distribution of winds, which include its speed and direction. In the thesis, the cyclonic winds are validated using buoy winds for the recent cyclones formed in the Bay of Bengal since 2010 using Jelesnianski wind scheme. It is found that the cyclonic winds computed from the scheme show an underestimate in the magnitude and also a mismatch in its direction. Hence, the wind scheme is suitably modified based on the buoy observations available at different locations using a power law, which reduces the exponential decay of winds by about 30%. Moreover, the cyclonic wind direction is also corrected by suitably modifying its inflow angle. The significance of modified exponential factor and inflow angle in the computation of cyclonic winds is highlighted using statistical analysis. The ADCIRC model is used to compute TWE as a response to combined effect of cyclonic winds and astronomical tides.

(10)

iii

As contribution of wave setup plays an important role near the coast, a coupled ADCIRC + SWAN is used to perceive the contribution of wind waves on the TWE. The experiments are performed to validate computed surge residuals with available tide gauge data. On comparison of observed surge residuals with the simulations using modified winds from the uncoupled and coupled models, it is found that the simulated surge residuals are better compared, especially with the inclusion of wave effect through the coupled model.

The low-lying Mahanadi river deltaic region along the Odisha coast is highly prone to inland flooding during the cyclone period. In particular, upstream river discharge, cyclone induced precipitation and storm tides modify the coastal inundation in the delta region. Experiments are performed using the 1999 Super cyclone and the 2013 Phailin cyclone with a standalone ADCIRC and a coupled hydraulic HEC-RAS and ADCIRC model. The Mahanadi, Brahmani and Baitarani Rivers are included in the computational domain with representative depths. The model simulations infer that coastal inundated area is enhanced by 64% after representing the river delta in the domain while, the river discharge from the upstream contributes additional 14%. The effect of volume of discharge and role of LULC information on computation of coastal inundation during the 1999 Super cyclone is also investigated. The coupled model system is used to quantify the contribution of precipitation on inland flooding during the cyclone period. The results signify that the inundated area becomes almost double after including rainfall data in both the cyclonic cases.

Also the model generated inundated area during Phailin cyclone is in good match with the satellite image, demonstrating the coupled system can simulate a reliable inland flooding in the delta region. It concludes that it is essential to resolve the river systems and incorporate hydrological components like river discharge and precipitation for precise computation of inundation.

(11)

iv

साराांश

वर्तमान शोध प्रबांध (थीससस), भारर् के पूवी र्ट पर र्ूफान वृसि और ज्वार के सांयुक्त प्रभाव के कारण आने वाली र्टीय बाढ़ की गणना करने के सलए एडवाांस्ड सकुतलेशन (एडसकत या ऐ. डी. सीआईआरसी. ) मॉडल का उपयोग करर्ी है। सपछले

चक्रवार्ी आांकड़ों के आधार पर सनसमतर् कृसिम चक्रवार्ी पथ़ों का उपयोग करके र्ट के साथ हर १०सकमी पर र्ूफान की वजह से

र्टीय बाढ़ की गणना करने के सलए ससमुलेशन सकये गये है। चक्रवार्ी हवाओां की गणना १०० वर्त की वापसी अवसध के आधार पर चक्रवार् के असधकर्म दबाव में कमी (प्रेशर–ड्रॉप) का उपयोग करके सकया गया है। सजला स्र्र पर जल स्र्र के साथ अांर्ःस्थलीय बाढ़ की असधकर्म सांभव सीमा के सलए र्ट की मैसपांग की गई है। ओसडशा र्ट के उत्तर में लगभग १० मीटर का जल स्र्र पाया

गया है। र्टीय जलभराव के मामले में सबसे सांवेदनशील क्षेि पसिम बांगाल के सजल़ों में पाया गया है, सजसमें असधकर्म जलभराव लगभग १३० सकमी भीर्र (इांटीररयर) है। आांध्र प्रदेश में, सबसे असधक बाढ़ग्रस्र् सजले नेल्लोर, प्रकाशम, गुांटूर, कृष्णा और पूवी

गोदावरी हैं। हालाांसक र्समलनाडु में रामनाथपुरम सजले में जल स्र्र ८ मीटर से असधक र्क पह ांच जार्ा है, लेसकन स्थानीय स्थलाकृसर्

के कारण र्टीय जलभराव से क्षेि अप्रभासवर् रहर्ा है। सवसभन्न जल स्र्ऱों के बाढ़ क्षेि की जाांच करने पर, यह देखा गया है सक २ मीटर से असधक जल स्र्र वाले जलभराव के कुल क्षेिफल का ७५% से असधक ओसडशा के उत्तरी सजल़ों और र्समलनाडु के

रामनाथपुरम सजले में है।

हाल के सदऩों में, भारर् के पूवी र्ट को चक्रवाऱ्ों द्वारा सवसभन्न सदशाओां से टकराने का खर्रा है। इससलए, चक्रवार्ीय जलभराव की उत्त्पसत्त पर चक्रवाऱ्ों के उपगमन कोण (एप्रोच एांगल) के प्रभाव और उनका ज्वार-भाटे (टाइड) और हवाओ से

उत्त्पन्न लहऱों (सवांड वेव्स) के साथ होने वाले सवर्म परस्पर प्रभाव (नॉनलीसनअर इांटरेक्शन) की सवस्र्ृर् जाांच एडसकत और

एडसकत+स्वान युसममर् मॉडल्स के उपयोग से की गयी है। समान र्ीव्रर्ा और सनयर् वेग से आगे बढ़र्े ह ए सिह सीधे आइडीलाइज़्ड

चक्रवार्ी पथ़ों का उपयोग करके सांख्यात्मक परीक्षण़ों का सनष्पादन सकया गया है। अध्ययन पररक्षेि (स्टडी डोमेन ) में एक

(12)

v

आइडीलाइज़्ड बाथीमेट्री के साथ सीधी र्टरेखा और एक समान (यूसनफामत) शेल्फ चौडाई (सवथत) मानी गयी है। सभी चक्रवार् एक ही स्थान पर १०° से १७०° के कोण के साथ १०° की क्रसमक वृसि के साथ लैंडफॉल करर्े हैं। ससम्युलेशन्स दशातर्े है की ६०- ९० सडग्री कोण पर टकराने वाले चक्रवार्ी पथ़ों के सलए असधकर्म र्ूफानी जलभराव स्र्र का पररकलन सकया गया है। १०° से

६०° र्क १५ प्रसर्शर् की वृसि जबसक ९०° से १७०° र्क १३ प्रसर्शर् की कमी शीर्त र्ूफान जलभराव स्र्र (पीक स्टॉमत सजत) में देखी गयी है। सजत-वेव इांटरेक्शन असधकर्म जल स्र्र को २१-२६% पररवसर्तर् करर्ा है। र्रांग का असधकर्म योगदान ९०° उसके

बाद १६०° और २०° वाले पथ़ों के सलए देखा गया है। ज्वार-भाटे की सवसभन्न अवस्थााओां के दौरान, सजत-टाइड

इांटरेक्शन

शीर्त जलभराव स्र्र को, उसके होने के समय और स्थान को पररवसर्तर् करर्ा है क्य़ोंसक चक्रवार् सभन्न कोण़ों पर टकरार्े (लैंडफॉल) है। प्रभासवर् र्टीय खांड की सीमा, र्ट के समीप जाने वाले पथ़ों के सलए लैंडफॉल के दोऩों ओर असधकर्म है, जबसक यह लांबवर् पथ़ों (९०°) के सलए न्यूनर्म है, और केवल लैंडफॉल के दासहनी र्रफ ही सीसमर् है। लैंडफॉल के २-४घांटे बाद, उच्च और सनम्न ज्वार-भाटे के दौरान शीर्त सजत-ज्वारभाटे -सवांड वेव्स इांटरैक्शन को र्ट के समानाांर्र देखा गया है। र्ट के साथ,

इांटरेक्शनचक्रवार्

के उपगमन कोण़ों पर सनभतर करर्ा

है,

हालाांसक कुल जल ऊांचाई (टी डब्लू ई) मुख्य रूप से ज्वारीय अवस्था और उपगमन कोण दोऩों पर सनभतर करर्ी है।

र्ट पर टी डब्लू ई की बढ़ोर्री मुख्यर्ा र्ीन कारक़ों से होर्ी है सजसमे र्ूफान जलभराव स्र्र, ज्वार और हवा की लहरे

ससम्मसलर् हैं। टी डब्ल्यू ई पूवातनुमान की यथाथतर्ा न केवल चक्रवार्ी ट्रैक और इसकी र्ीव्रर्ा पर सनभतर करर्ा है, बसल्क हवाओां के

स्थासनक प्रसार पर भी सनभतर करर्ा है, सजसमें इसकी गसर् और सदशा ससम्मसलर् हैं। थीससस में, २०१० के बाद से बांगाल की खाडी में

बनने वाले हासलया चक्रवाऱ्ों के सलए चक्रवार्ी हवाओां की गणना जेलससनएन्स्की पवन प्रणाली का उपयोग करके की गयी है सजनको

बॉय हवाओां के सापेक्ष प्रमासणर् सकया गया है। यह पाया गया है सक इस प्रणाली से पररकसलर् चक्रवार्ी हवाएँ पररमाण में कम है और सदशा में भी एक असाम्यर्ा है। इससलए, पवन प्रणाली को एक पावर लॉ का उपयोग करके सवसभन्न स्थाऩों पर उपलब्ध बॉय पयतवेक्षण आांकड़ों के आधार पर उपयुक्त रूप से सांशोसधर् सकया गया है, जो हवाओां के घार्ीय क्षय को लगभग ३०% र्क कम कर देर्ा है । इसके

अलावा, चक्रवार्ी हवा की सदशा को भी उपयुक्त रूप से सांशोसधर् करके उनके प्रवाह (इनफ्लो) कोण को ठीक करर्ा है। चक्रवार्ी

(13)

vi

हवाओां की गणना में सांशोसधर् घार्ीय कारक और इनफ्लो कोण का महत्व साांसख्यकीय सवश्लेर्ण का उपयोग करके उजागर सकया गया

है। एडसकत मॉडल का उपयोग चक्रवार्ी हवाओां और खगोलीय ज्वार के सांयुक्त प्रभाव की प्रसर्सक्रया के रूप में टी डब्लू ई की गणना के

सलए सकया गया है। र्ट के पास लहर सेटअप का योगदान एक महत्वपूणत भूसमका सनभार्ा है, एक युसममर् एडसकत+स्वान मॉडल का

उपयोग टी डब्लू ई पर पवन र्रांग़ों के योगदान को समझने के सलए सकया गया है। उपलब्ध टाइड गेज आांकड़ों के सापेक्ष पररकसलर् शेर्

जलभराव स्र्र को प्रमासणर् करने के सलए प्रयोग सकए गए हैं। पयतवेक्षण आांकड़ों के सापेक्ष युसममर् और अयुसममर् मॉडल में सांशोसधर्

हवाओां का उपयोग करके गसणर् शेर् जलभराव स्र्र (सजत रेससडुयल) की र्ुलना यह बर्ार्ी है सक युसममर् मॉडल के माध्यम से र्रांग प्रभाव़ों को ससम्मसलर् करने के बाद ससम्युलेटेड रेससडुयल की र्ुलना बेहर्र होर्ी है।

ओसडशा र्ट पर महानदी के सनचले डेल्टाइक क्षेि में चक्रवार् की अवसध के दौरान अांर्ःस्थलीय बाढ़ का अत्यसधक खर्रा

है । सवशेर् रूप से, अपस्ट्रीम नदी सनवतहन, चक्रवार् प्रेररर् वर्ात और र्ूफान ज्वार डेल्टा क्षेि में र्टीय जलभराव को सांशोसधर् करर्े

हैं । १९९९ सुपर साइक्लोन और २०१३ फेसलन चक्रवार् का उपयोग करके स्टैंड-एलोन एडसकत और युसममर् हाइड्रोसलक हेक-रास और एडसकत मॉडल के साथ प्रयोग सकया गया है। महानदी, ब्राह्मणी और बैर्रणी नसदयाँ कम््यूटेशनल डोमेन में प्रसर्रूप गहराई के

साथ ससम्मसलर् सकया गया है। मॉडल ससमुलेशन से अनुमान लगाया गया है सक समुद्र में नदी के डेल्टा का प्रसर्रूपण करने के बाद र्टीय बाढ़ क्षेि में ६४% की वृसि ह ई है, जबसक नदी का अपस्ट्रीम नदी सनवतहन असर्ररक्त १४% का योगदान देर्ा है। १९९९ के

सुपर साइक्लोन के दौरान, र्टीय बाढ़ की गणना पर नदी सनवतहन के आयर्न/प्रसार के प्रभाव और एल यू एल सी जानकारी की

भूसमका की जाांच भी की गई है। चक्रवार् अवसध के दौरान अांर्ःस्थलीय बाढ़ पर वर्ात के योगदान को सनधातररर् करने के सलए युसममर्

मॉडल प्रणाली का उपयोग सकया गया है। पररणाम दशातर्े हैं सक दोऩों चक्रवार्ी मामल़ों में वर्ात के आांकड़ों को शासमल करने के बाद

बाढ़ग्रस्र् क्षेि लगभग दोगुना हो गया है। इसके असर्ररक्त फेसलन चक्रवार् के दौरान उत्पन्न होने वाला क्षेि उपग्रह छसव (सेटलाइट

सपक्चर) के साथ अच्छा मेल खार्ा है, यह दशातर्ा है सक युसममर् प्रणाली डेल्टा क्षेि में एक सवश्वसनीय अांर्ःस्थलीय बाढ़ का

(14)

vii

अनुरूपर्ा (ससमुलेशन) कर सकर्ी है। यह सनष्कर्त सनकलर्ा है सक नदी प्रणासलय़ों को हल करना आवश्यक है और बाढ़ की सटीक

गणना के सलए नदी के सनवतहन और वर्ात जैसे हाइड्रोलॉसजकल घटक़ों को शासमल करना आवश्यक है।

(15)

viii

List of Figures

Figure No. Title Page

No.

Fig. 1.1 Vertical cross section of a tropical cyclone. 5

Fig. 1.2 Cyclones, Hurricanes and Typhoon with their peaks of activity

worldwide. 6

Fig. 1.3 Cyclonic storm and severe cyclonic storms developed during post- monsoon season (1988-2018) in NIO.

8 Fig. 1.4 Surge generated due to barometric effect and circulating cyclonic winds. 10 Fig. 1.5 Tidal cycles in different regions in the world. 12 Fig. 2.1 Computational domain along with tide gauge stations. 47

Fig. 2.2 Tidal range (m) along east coast of India. 49

Fig. 2.3 Comparison of modelled tide against the observed tide at (a) Ennore (b) Krishnapatnam (c) Kakinada (d) Visakhapatnam (e) Paradeep (f) Garden Reach tide gauge locations.

50-51 Fig. 3.1 Frequency of district-wise landfalling cyclonic storms during 1891-2018. 56 Fig. 3.2 Model mesh covering the east coast of India along with synthetic cyclone

tracks.

57 Fig. 3.3 Bathymetry and onshore topography for the model domain. The detailed

river systems are shown in the insets. 58

Fig. 3.4 Depiction of land topography upto 15m for all maritime states along the east coast of India: (a) Odisha, (b) West Bengal, (c) Andhra Pradesh and (d) Tamil Nadu.

59

Fig. 3.5 Computed synthetic tracks for districts (a) Jagatsinghpur and (b) Kendrapara in Odisha. Distance between marked stars gives length of coastal stretch of the district.

61

Fig. 3.6 Composite depiction of probable maximum extreme water elevation and probable maximum extreme water level associated with inundation due to all possible cyclones crossing (a) Balasore district (b) Odisha coast.

63

Fig. 3.7 Composite depiction of probable maximum extreme water elevations and probable maximum extreme water level associated with inundation due to all possible cyclones crossing West Bengal coast.

64

(20)

xiii

No.

Fig. 3.8 Composite depiction of probable maximum extreme water elevations and probable maximum extreme water level associated with inundation due to all possible cyclones crossing Andhra Pradesh coast (a) zoomed near Krishna district (b) Andhra Pradesh coast.

66

Fig. 3.9 Composite depiction of probable maximum extreme water elevations and probable maximum extreme water level associated with inundation due to all possible cyclone crossing Tamil Nadu coast (a) zoomed near Nagapattinam district (b) Tamil Nadu coast.

66

Fig. 3.10 (((a) Probable maximum extreme water elevations(m) and (b) Probable a maximum extreme water levels(m) for all districts along the east coast

o of India

68

Fig. 3.11 (a) District-wise inundated area along the east coast of India (b)

Inundated area at different water levels. 69

Fig. 4.1 Some cyclone tracks of the recent past in the Bay of Bengal having

unusual track directions. 73

Fig. 4.2 (a) Idealized computational domain along with cyclone tracks at different approach angles starting from north at 10° intervals. Yellow dot indicates the location of peak surge for tracks 2, 9 & 16, (b) Idealized local bathymetry along the tracks 1-9.

76

Fig. 4.3 Computed peak storm surge for different approach angles of a cyclone. 79

Fig. 4.4 (a) Representation of actual bathymetry of the western Bay of Bengal over the domain along with cyclone tracks approaching at different angles starting from north at 10° intervals, (b) Computed peak storm surge for different approach angles of a cyclone.

79

Fig. 4.5 Maximum storm surge envelope along the coast (a) tracks 2-9, (b) tracks 10-16. Positive values lie to the right hand side of the landfall location.

The black star represents the landfall location.

80

Fig. 4.6 Hovmöller diagram of zonal wind (a-c), meridional wind (d-f) and wind speed in contour and its direction by vector (g-i) along the coast for tracks 2, 9 & 16. The black dot indicates the landfall time.

83

Fig. 4.7 Temporal depiction of net water flux averaged over the box1 for (a)

tracks 2-9, (b) tracks 10-16. The black dot represents the landfall time. 85

(21)

xiv

No.

Fig. 4.8 Temporal depiction of net water flux averaged over box2 for (a) tracks

2-9, (b) tracks 10-16. The black dot represents the landfall time. 85 Fig. 4.9 Distribution of maximum significant wave height (SWH) over the

domain for tracks 2, 5, 9, 13 & 16 (a-e) and distribution of maximum radiation stress gradient (RSG) over the domain for tracks 2, 5, 9, 13 &

16(f- j).

88

Fig. 4.10 Effect of the wind wave on the maximum water elevation for (a) track2,

(b) track 5, (c) track 9, (d) track 13, (e) track 16. 88 Fig. 4.11 The (a) amplitude and (b) phase of the surface elevation for the M2 tidal

constituent.

91 Fig. 4.12 Computed maximum storm tide for track 2 (20^o), track 5(50^o), track 9

(90ô), track 13(130ô) and track 16 (160ô) at mid-flood, high-tide, mid-ebb and low-tide.

91

Fig. 4.13 Fig. 4.14 Fig. 4.15 Fig. 4.16

Fig. 4.17

Fig. 4.18

Maximum storm tide envelope along the coast for track 2, track 9 and track 16 at (a) high-tide and (b) low-tide.

Temporal depiction of total water elevation of S, W, SW, Tide, ST, STW for tracks 2, 9 & 16 at high-tide (a-c) and low-tide (d-f).

Temporal depiction of non-linear interaction of surge-tide-wave (NSTW) for tracks 2, 9 &16 at (a) high-tide, (b) low-tide.

Hovmöller diagram for STW (a, b & c) and NSTW (d, e & f) along the coast for tracks 2, 9 & 16 respectively, at high-tide. The black dot indicates the landfall time.

Hovmöller diagram of STW (a, b & c) and NSTW (d, e & f) along the coast for tracks 2, 9 & 16 respectively, at low-tide. The black dot indicates the landfall time.

Correlation between STW and NSTW along the coast for tracks 2, 9 &

16 at high-tide (a, b & c) and low-tide (d, e & f).

92 93 95 97

98 99 Fig. 5.1 Model domain of the Bay of Bengal along with cyclone tracks. Buoy

locations marked by thick black dot. 106

Fig. 5.2 Spatial distribution of cyclonic winds for Phailin (a) unmodified (b) modified.

109

Fig. 5.3 Comparison of unmodified and modified wind speed with the observed for 2010 Jal cyclone at different buoy locations a BD06, b BD13, c BD14. Black dot represents the landfall time.

111

(22)

xv

No.

Fig. 5.4 Comparison of unmodified and modified wind direction with the observed for 2010 Jal cyclone at different buoy locations a BD06, b BD13, c BD14. Black dot represents the landfall time.

112

Fig. 5.5 Comparison of unmodified and modified wind speed with the observed for Thane cyclone at different buoy locations a BD11, b BD13. Black dot represents the landfall time.

114

Fig. 5.6 Comparison of unmodified and modified wind direction with the observed for Thane cyclone at different buoy locations a BD11, b BD13.

Black dot represents the landfall time.

115

Fig. 5.7 Comparison of unmodified and modified wind speed with the observed for Phailin cyclone at different buoy locations a BD08, b BD09. Black dot represents the landfall time.

117

Fig. 5.8 Comparison of unmodified and modified wind direction with the observed for Phailin cyclone at different buoy locations a BD08, b BD09. Black dot represents the landfall time.

118

Fig. 5.9 Comparison of unmodified and modified wind speed with the observed for Hudhud cyclone at different buoy locations a BD09, b BD10, c BD11. Black dot represents the landfall time.

120

Fig. 5.10 Comparison of unmodified and modified wind direction with the observed for Hudhud cyclone at different buoy locations a BD09, b BD10 c BD11. Black dot represents the landfall time

121

Fig. 5.11 Validation of surge residuals calculated from different simulations with the observed for a Thane cyclone, b Phailin cyclone and c Hudhud cyclone. Black dot represents the landfall time.

124

Fig. 6.1 Depiction of (a) Mahanadi deltaic region along with river cross-sections and discharge observed locations (b) Digitized Mahanadi, Brahmani and Baitarani Rivers along with their tributaries.

137

Fig. 6.2 Comparison of observed and computed river cross-sectional depth at (a)

Pubansa (b) Alipingal (c) Marshaghai (d) Jenapur (e) Indupur. 139 Fig. 6.3 Bathymetry and onshore topography of the Mahanadi deltaic region for

the model domain (a) modified STRM data (b) unmodified SRTM data. 140

Fig. 6.4 Computational domain for HEC-RAS model along with cross-sections.

Label 1-5 shows the river coast boundary for Baitarani, Brahmani and the Mahanadi Rivers, respectively.

142

(23)

xvi

No.

Fig. 6.5 Computational domain (DT-1) along with cyclonic tracks for 1999 Super

cyclone and 2013 Phailin cyclone. 145

Fig. 6.6 Spatial coverage of coastal inundation and maximum water level for the 1999 Super cyclone (a) Domain with modified SRTM data (DT-1) (b) Domain with unmodified SRTM data (DT-2).

145

Fig. 6.7 Daily-discharge hydrograph during the 1999 Super cyclone at

(a)Mahanadi (b)Brahmani (c)Baitarani. 147

Fig. 6.8 (a) Spatial coverage of coastal inundation and maximum water level with daily river discharge for the 1999 Super cyclone (b) composite depiction of coastal inundation for 1999 Super cyclone as an aggregate response of without and with discharge.

147

Fig. 6.9 A composite depiction of coastal inundation as an aggregate response of

with no-discharge and with discharge only into Brahmani River. 148 Fig. 6.10 Depiction of LULC data over the Mahanadi delta. 149 Fig. 6.11

Fig. 6.12 Fig. 6.13 Fig. 6.14 Fig. 6.15

Fig. 6.16 Fig. 6.17

Spatial coverage of coastal inundation and maximum water level without including LULC information for 1999 Super cyclone.

A composite depiction of coastal inundation as an aggregate response of with and without LULC information for 1999 Super cyclone.

Inundated maximum water level with and without LULC for each classification.

Daily-discharge hydrograph during the 2013 Phailin cyclone at (a)Mahanadi (b)Brahmani (c) Baitarani.

(a)Comparison of modelled tide against the observed tide at Paradeep, (b) Comparison of model simulated storm tide with observations for Phailin cyclone. Black dot shows the landfall time.

Spatial coverage of coastal inundation and maximum water level for Phailin cyclone.

Temporal depiction of storm tide during the Phailin cyclone at river coastal boundary (a)Baitarani (b) Brahmani (c) Mahanadi opening1 (d) Mahanadi opening2 (e) Mahanadi opening3. Black dot represents the landfall time.

151 151 152 153 154

155 159

(24)

xvii

No.

Fig. 6.18

Fig. 6.19

Fig. 6.20

Fig. 6.21

Depiction of maximum water level and coastal inundation during Phailin cyclone (a) simulation with storm tide and river discharge, (b) simulation with storm tide, river discharge and precipitation. Black oval highlights the major inundated area.

(a) Landsat-8 satellite image of 26^th April 2013 when there was no weather system (b) Landsat-8 satellite image of 19^th October 2013 just after the cyclone. Black oval highlights the major inundated area.

Temporal depiction of storm tide during the 1999 Super cyclone at river coastal boundary (a) Baitarani (b) Brahmani (c) Mahanadi opening1 (d) Mahanadi opening2 (e) Mahanadi opening3. Black dot represents the landfall time.

Depiction of maximum water level and coastal inundation during 1999 Super cyclone (a)simulation with storm tide and river discharge

(b)simulation with storm tide, river discharge and precipitation.

160

161

162

(25)

xviii Table No.

List of Tables

Title Page

No.

Table 1.1 Classification of cyclonic disturbance according to IMD. 4 Table 2.1 Correlation coefficient and RMSE at different tide gauge locations. 52 Table 3.1 Observed maximum intensity of the cyclone for each maritime state. 56 Table 4.1 Change of water level in % between surge and SW/ST/STW at high and

low-tide.

94

Table 5.1 Details of the buoy locations. 107

Table 5.2 List of cyclones along with corresponding maximum pressure drop and the radius of maximum winds.

110 Table 5.3 Mean bias error (MBE) and root mean square error (RMSE) at different

buoy locations for 2010 Jal Cyclone.

113 Table 5.4 Mean bias error (MBE) and root mean square error (RMSE) at different

buoy locations for 2011 Thane Cyclone.

.

116 Table 5.5

Table 5.6

Table 5.7 Table 6.1

Mean bias error (MBE) and root mean square error (RMSE) at different buoy locations for 2013 Phailin Cyclone.

Mean bias error (MBE) and root mean square error (RMSE) at different buoy locations for 2014 Hudhud Cyclone.

Root mean square error (RMSE) for SIM1, SIM2, SIM3 and SIM4 for Thane cyclone and Hudhud cyclone.

Daily area-averaged precipitation over the Mahanadi deltaic region during 1999 Super cyclone and Phailin cyclone.

119 122

125 143

Numerical modeling of coastal inundation in response to a tropical cyclone along the east coast of India

NUMERICAL MODELING OF COASTAL INUNDATION IN RESPONSE TO A TROPICAL CYCLONE ALONG THE EAST

COAST OF INDIA

SMITA PANDEY (NEE. MISHRA)

CENTRE FOR ATMOSPHERIC SCIENCES INDIAN INSTITUTE OF TECHNOLOGY DELHI

OCTOBER 2020

NUMERICAL MODELING OF COASTAL INUNDATION IN RESPONSE TO A TROPICAL CYCLONE ALONG THE EAST

COAST OF INDIA

by

SMITA PANDEY (NEE. MISHRA) Centre for Atmospheric Sciences

Submitted

in fulfilment of the requirements of the degree of DOCTOR OF PHILOSOPHY

to the

INDIAN INSTITUTE OF TECHNOLOGY DELHI

OCTOBER 2020

Dedicated to my family

Certificate

This is to certify that the thesis entitled "Numerical modeling of coastal inundation in response to a tropical cyclone along the east coast of India"

(Dr A. D. Rao)

Professor, Centre for Atmospheric Sciences Indian Institute of Technology Delhi

New Delhi-110016.

Abstract

साराांश

रामनाथपुरम सजले में है।

बाद १६०° और २०° वाले पथ़ों के सलए देखा गया है। ज्वार-भाटे की सवसभन्न अवस्थााओां के दौरान, सजत-टाइड

के उपगमन कोण़ों पर सनभतर करर्ा

हालाांसक कुल जल ऊांचाई (टी डब्लू ई) मुख्य रूप से ज्वारीय अवस्था और उपगमन कोण दोऩों पर सनभतर करर्ी है।

र्ट पर टी डब्लू ई की बढ़ोर्री मुख्यर्ा र्ीन कारक़ों से होर्ी है सजसमे र्ूफान जलभराव स्र्र, ज्वार और हवा की लहरे

सलए सकया गया है। र्ट के पास लहर सेटअप का योगदान एक महत्वपूणत भूसमका सनभार्ा है, एक युसममर् एडसकत+स्वान मॉडल का

ओसडशा र्ट पर महानदी के सनचले डेल्टाइक क्षेि में चक्रवार् की अवसध के दौरान अांर्ःस्थलीय बाढ़ का अत्यसधक खर्रा

सुपर साइक्लोन के दौरान, र्टीय बाढ़ की गणना पर नदी सनवतहन के आयर्न/प्रसार के प्रभाव और एल यू एल सी जानकारी की

अनुरूपर्ा (ससमुलेशन) कर सकर्ी है। यह सनष्कर्त सनकलर्ा है सक नदी प्रणासलय़ों को हल करना आवश्यक है और बाढ़ की सटीक

गणना के सलए नदी के सनवतहन और वर्ात जैसे हाइड्रोलॉसजकल घटक़ों को शासमल करना आवश्यक है।

Table of Contents

List of Figures

List of Tables