PhD thesis work

A brief summary of my PhD thesis work is shared here.

Abstract

Coastal areas are highly vulnerable to frequent hazards, which can disrupt community life and have significant social and economic impacts. While vulnerability assessments using index-based approaches are common for coastal areas, studies focusing specifically on beach vulnerability are limited. This study addresses this gap by examining exposure, susceptibility, and recovery potential as key components of beach vulnerability. To assess beach vulnerability, a temporal and spatial beach vulnerability assessment tool called the Beach Vulnerability Index (BVI) is developed using Multiple Linear Regression (MLR) and Artificial Neural Networks (ANN). Additionally, a comprehensive analysis investigates beach susceptibility and resilience by considering beach morphology.

Data from the Hasaki Oceanographic Research Station (HORS) in Ibaraki prefecture, Japan, including hourly wave and water level observations, tidal predictions, and daily beach profile data from 1987 to 2010, were collected. The MLR model was initially used to predict storm-induced erosion, and the BVI was formulated based on MLR predictions, incorporating variables of wave energy flux, initial shoreline position, and maximum surge. Subsequently, an ANN model was introduced and compared to MLR in terms of predicting shoreline change during storms. The BVI is refined for both models, resulting in BVI_ANN for the ANN model and BVI_MRL for the MLR model.

The comprehensive analysis categorizes beach profiles into four groups: unbarred, inner zone sandbar, outer zone sandbar, and double sandbar. Statistical analyses and metric development were conducted to assess beach susceptibility using the Beach Erosion Susceptibility Number (BESN) and beach resilience using the Beach Resilience Number (BRN). The study successfully identified key morphometric factors that influence beach erosion and quantified the BESN accordingly. Additionally, 10-day post-storm beach recovery calculations were employed to quantify the BRN as a ratio relative to the erosion that occurred during the storm. Both the BESN and BRN are valuable tools for chronological analysis of beach characteristics and for identifying abrupt changes in the beach system resulting from unexpected events in coastal zones. To select the morphometric indicators, XGBoost regression models were fitted for both susceptibility and resilience quantification. The SHAP explanation method was then applied to globally and locally quantify the importance of each morphometric feature in influencing the outcomes of the models.

During the preliminary study period using data from 1993 to 2000, 48 storms were identified, and the first 38 storms were used to create two MLR models. These models achieved R² values of 0.58 and 0.52 for predicting shoreline change (dSL) and volume change (dV) during the training period, respectively. During testing, the corresponding R² values were 0.48 for shoreline change (dSL) and 0.52 for volume change (dV). The comparison of the BVI calculated from the MLR predictions with observed erosion values showed satisfactory model performance. However, when the complete storm data set from 1987 to 2010 was used, the MLR model proved less effective in capturing the dynamic behaviour of beach profiles under different storm conditions. In this expanded period, a total of 347 storm cases were identified, and 5% of the data (18 storms) were used to test each regression model. In contrast, the ANN model demonstrated superior performance, resulting in more accurate predictions of beach vulnerability. Specifically, when comparing the BVI predictions from each model, the Mean Absolute Errors (MAE) for MLR were 1.33, 0.83, 0.78, 0.90, and 1.07, while for BVI_ANN they were 1.00, 0.20, 0.69, 1.05, and 0.57 for indexes 1-5, respectively. The BVI_ANN model also achieved higher R² scores for both training (0.65) and testing (0.62) data in predicting dSL compared to the MLR model (0.26 for training and 0.35 for testing).

The comprehensive analysis revealed that the contributing morphometrics for beach susceptibility varied depending on the beach profile types. The initial shoreline position was found to have an impact only in the unbarred and double sandbar profiles. When comparing the predictions of the BESN with observed beach changes, a Pearson correlation coefficient (r) of 0.75 was obtained for unbarred profiles for average storm conditions, indicating satisfactory performance. However, the accuracy of the BESN was found to be lower under certain conditions, likely due to variations caused by storm characteristics. In terms of beach resilience analysis, 104 cases were were identified with no sequencing storms and initial shoreline erosion cases during the storm events. Among these cases, 79 incidents exhibited partial or full shoreline recovery, while 25 cases experienced further erosion.

The results indicate that the MLR models provide moderate predictions of beach erosion and adequate predictions of beach vulnerability when tested with a small sample of storms during the preliminary study period. However, when the larger and more diverse storm data set from 1987 to 2010 was utilized, the MLR model showed limitations in capturing the dynamic behavior of beach profiles under different storm conditions. In contrast, the ANN model demonstrated superior performance, delivering more accurate predictions of beach vulnerability. The Mean Absolute Errors for the BVI predictions were lower for BVI_ANN compared to BVI_MRL, indicating the improved accuracy of the BVI_ANN model. Additionally, the BVI_ANN model achieved higher R² scores for both training and testing data in predicting shoreline change compared to the MLR model.

The comprehensive analysis revealed that beach susceptibility varies depending on the beach profile types, with the initial shoreline position playing a significant role in unbarred and double sandbar profiles. The BESN demonstrated satisfactory performance, particularly for inner zone sandbar profiles, indicating its potential as a tool for assessing beach susceptibility. However, it also highlighted the need for further improvements to account for variations in storm conditions that can affect the accuracy of the predictions. The beach resilience analysis identified cases of partial or full shoreline recovery, as well as instances of further erosion, providing insights into the post-storm behavior of beach profiles.

These findings further underscore the potential of employing machine learning-based algorithms such as ANN and XGBoost to enhance the accuracy of beach vulnerability studies, particularly in capturing the dynamic nature of beach morphology changes under diverse storm conditions. The study also highlights the importance of considering sandbar formations and sediment volume as crucial factors in determining the processes of erosion and recovery associated with beach vulnerability.

Future work should focus on testing the BESN and BRN under various beach conditions and conducting numerical simulations to explore beach morphology changes in similar wave climates. Furthermore, the direct applicability of the BVI, trained using the ANN model, represents a key advantage that can be leveraged for assessing beach vulnerability in other coastal areas. This study contributes to the growing body of research on beach vulnerability by utilizing Machine Learning algorithms to predict coastal morphology changes and assess beach vulnerability, highlighting their potential for future applications in coastal engineering.

Introduction

Citations are then used in the article body with the <d-cite> tag. The key attribute is a reference to the id provided in the bibliography. The key attribute can take multiple ids, separated by commas.

The citation is presented inline like this: (a number that displays more information on hover). If you have an appendix, a bibliography is automatically created and populated in it.

Distill chose a numerical inline citation style to improve readability of citation dense articles and because many of the benefits of longer citations are obviated by displaying more information on hover. However, we consider it good style to mention author last names if you discuss something at length and it fits into the flow well — the authors are human and it’s nice for them to have the community associate them with their work.

Other literatures

Just wrap the text you would like to show up in a footnote in a <d-footnote> tag. The number of the footnote will be automatically generated.This will become a hoverable footnote.

Beach Vulnerability Index

Syntax highlighting is provided within <d-code> tags. An example of inline code snippets: <d-code language="html">let x = 10;</d-code>. For larger blocks of code, add a block attribute:

var x = 25; function(x) { return x * x; }

Note: <d-code> blocks do not look good in the dark mode. You can always use the default code-highlight using the highlight liquid tag:

var x = 25;
function(x) {
  return x * x;
}

Beach Susceptibility

You can add interative plots using plotly + iframes

The plot must be generated separately and saved into an HTML file. To generate the plot that you see above, you can use the following code snippet:

import pandas as pd
import plotly.express as px
df = pd.read_csv(
  'https://raw.githubusercontent.com/plotly/datasets/master/earthquakes-23k.csv'
)
fig = px.density_mapbox(
  df,
  lat='Latitude',
  lon='Longitude',
  z='Magnitude',
  radius=10,
  center=dict(lat=0, lon=180),
  zoom=0,
  mapbox_style="stamen-terrain",
)
fig.show()
fig.write_html('assets/plotly/demo.html')

Beach Resilience

Details boxes are collapsible boxes which hide additional information from the user. They can be added with the details liquid tag:

Click here to know more

Additional details, where math $2x - 1$ and code is rendered correctly.

format guide

The main text column is referred to as the body. It is the assumed layout of any direct descendants of the d-article element.

.l-body

For images you want to display a little larger, try .l-page:

.l-page

All of these have an outset variant if you want to poke out from the body text a little bit. For instance:

.l-body-outset

.l-page-outset

Occasionally you’ll want to use the full browser width. For this, use .l-screen. You can also inset the element a little from the edge of the browser by using the inset variant.

.l-screen

.l-screen-inset

The final layout is for marginalia, asides, and footnotes. It does not interrupt the normal flow of .l-body sized text except on mobile screen sizes.

.l-gutter

Other Typography?

Emphasis, aka italics, with asterisks (*asterisks*) or underscores (_underscores_).

Strong emphasis, aka bold, with asterisks or underscores.

Combined emphasis with asterisks and underscores.

Strikethrough uses two tildes. ~~Scratch this.~~

First ordered list item
Another item ⋅⋅* Unordered sub-list.
Actual numbers don’t matter, just that it’s a number ⋅⋅1. Ordered sub-list
And another item.

⋅⋅⋅You can have properly indented paragraphs within list items. Notice the blank line above, and the leading spaces (at least one, but we’ll use three here to also align the raw Markdown).

⋅⋅⋅To have a line break without a paragraph, you will need to use two trailing spaces.⋅⋅ ⋅⋅⋅Note that this line is separate, but within the same paragraph.⋅⋅ ⋅⋅⋅(This is contrary to the typical GFM line break behaviour, where trailing spaces are not required.)

Unordered list can use asterisks
Or minuses
Or pluses

I’m an inline-style link

I’m an inline-style link with title

I’m a reference-style link

I’m a relative reference to a repository file

You can use numbers for reference-style link definitions

Or leave it empty and use the link text itself.

URLs and URLs in angle brackets will automatically get turned into links. http://www.example.com or http://www.example.com and sometimes example.com (but not on Github, for example).

Some text to show that the reference links can follow later.

Here’s our logo (hover to see the title text):

Inline-style: alt text

Reference-style: alt text

Inline code has back-ticks around it.

var s = "JavaScript syntax highlighting";
alert(s);

s = "Python syntax highlighting"
print s

No language indicated, so no syntax highlighting.
But let's throw in a <b>tag</b>.

Colons can be used to align columns.

Tables	Are	Cool
col 3 is	right-aligned	$1600
col 2 is	centered	$12
zebra stripes	are neat	$1

There must be at least 3 dashes separating each header cell. The outer pipes (|) are optional, and you don’t need to make the raw Markdown line up prettily. You can also use inline Markdown.

Markdown	Less	Pretty
Still	`renders`	nicely
1	2	3

Blockquotes are very handy in email to emulate reply text. This line is part of the same quote.

Quote break.

This is a very long line that will still be quoted properly when it wraps. Oh boy let’s keep writing to make sure this is long enough to actually wrap for everyone. Oh, you can put Markdown into a blockquote.

Here’s a line for us to start with.

This line is separated from the one above by two newlines, so it will be a separate paragraph.

This line is also a separate paragraph, but…

This theme supports rendering beautiful math in inline and display modes using MathJax 3 engine. You just need to surround your math expression with $$, like $$ E = mc^2 $$. If you leave it inside a paragraph, it will produce an inline expression, just like $E = mc^2$.

To use display mode, again surround your expression with $$ and place it as a separate paragraph. Here is an example:

\[\left( \sum_{k=1}^n a_k b_k \right)^2 \leq \left( \sum_{k=1}^n a_k^2 \right) \left( \sum_{k=1}^n b_k^2 \right)\]

Note that MathJax 3 is a major re-write of MathJax that brought a significant improvement to the loading and rendering speed, which is now on par with KaTeX.

This line is only separated by a single newline, so it’s a separate line in the same paragraph.