Visualizing Earthquakes

[rasagy]

For the final assignment, let’s use the USGS Earthquake Catalog to visualize earthquakes on a map.

For the assignment, try to narrow down on the three filters for your narrative: Minimum magnitude, Date range and Geographic Region and add that as a comment below.

Next, pick one of the following tools to visualize this data:

1. Tableau: Read about how to make proportional symbol maps (dots sized by data), and how to add a custom Mapbox style to Tableau.
2. Flourish: Try the Time Map template for a timeline view, or experiment with the 3D Marker map
3. Mapbox Studio + Mapbox GL: Check out the heatmap tutorial, timeline visualization tutorial and scrollytelling example. Also see Alli’s video tutorials on visualizing trees.
4. Excel: Use the geocode excel to geocode additional data, and pick relevant India-specific choropleth excel files from here.

For the final submission, make sure your visualization tells a story — choose an insightful heading, give it a description, use annotations & help the reader understand your story.

For inspiration, have a look at what your senior batch did: #38.

[prachitank]

Earthquake Ideas

Idea 1

The DYFI responses and people's reactions to earthquakes seemed like an interesting dataset to work with. A story titled 'How many earthquakes go unnoticed' could be fun. But I realised that the responses on the USGS portal are not anywhere close to being representative of the entire population. But then I realised that another (more rich) source for DYFI responses could be Twitter. So my first approach can be mapping the tweets or the number of tweets which were posted at at the time of the earthquake (+ ~2 hrs) with the hashtags earthquake, quake, etc. (the entire list is yet to be identified). This could also be mapped to the actual damage that was caused to see how the popularity of a topic relates to the actual impact.

But it was pointed out that the chances of that working out weren't so great. I decided to pivot. I still pursued the base idea of how many people notice an earthquake. This led me to explore two directions 1) How many earthquakes are recorded each day - the number for which was surprisingly high and 2) The number of earthquakes below a 2.5 scale

I plotted the data of the earthquakes below 2.5 for a month and used a slider to filter each day’s quakes. On plotting the 15000 odd points and got moderate results. I was hoping to maybe get a visualisation of the plates moving. But that didn’t happen like I wanted it to. You can find the experimentations here. This didn't seem to be going anywhere though.

Next I looked at the earthquakes that came before a large major earthquake or what the 'calm before the storm' looked like and decided to plot it. And the juxtaposition of three different major earthquakes' build up was interesting - in the case of Chile, the seismic activity followed the general trend, in the case of Japan, there was a spike in activity for 2 days before the major earthquake and for Haiti, there was an absence of seismic activity for almost 3 months before the major earthquake. It was interesting to see these unravel side by side.

So, I decided to make a worksheet which shows three maps next to each other. One of the features I wanted to achieve was to get all three maps to be controlled by the same filter. So (this was the workaround I figured out for it, there could also be other possible ways to go about it) I created another colomn with normalised dates that were applied to all three maps. I use a data set of 90 days at first but it seemed like excess amount of data that didn't really illustrate the point. So I changed it to 30 days. The maps were good at illustrating the geographical distribution and the sharp spike in the number of earthquakes, however, to drive the point home, I decided to add simple line graphs that would help the viewer know what to look for in the map visualisation.

Since Tableau also gives a lot of freedom of exploring the aesthetic, I meddled around with that a bit to make the viz look more like a story on a webpage rather than a presentation.

Here is the final visualisation.

[mayura7]

I read about the Foreshock, Mainshock and Aftershock phenomenon which explain the repetitive earthquakes happening in one place. I looked up earthquakes happening in the North East India as my husband has been working there for past 5 years and has experienced a few earthquakes. I found that there are places where earthquakes have happened multiple times over a time period. I'd like to follow the theme of repetitive earthquakes (searching a pattern of the three phenomenon mentioned above) in the NE India and look for a story through the data I get.

I played around with the data to see how I can isolate these phenomenon for a particular region but realized that it makes the data set extremely small. So I decided I'll make a story on how this beautiful calming region is internally devastated by quakes. Sort of melodramatic approach! Here is the story I created on Flourish using one of its timeline templates.

[rishivanukuru]

Earthquake Data Stories

Newer additions at the bottom - Last edit 14/9/2019

Current Idea- Bhuj 2001: Before the aftermath, after the aftershocks

This is something I'm developing on the Bhuj earthquake. While investigating the area, I found a mild earthquake that took place about a month before the main one struck on Republic day 2001. Together with a lack of similar incidents for nearly 50 years in the past, and the constant cyclical nature of minor quakes in the years since, I thought it would be interesting to present this case study of sorts through a data story.

There were a couple other interesting information pieces that came out from a cursory look at the data. A small proof of concept of what I hope to properly develop by later today is linked below -

Bhuj: Before the Aftermath, After the Aftershocks

Disclaimer: This just builds upon the scrolling text + flying map example that Rasagy mentioned in class. It will only work on laptop screens, and isn't anywhere close to completion.

9/9/2019: I made the mistake of pushing the updated website, so this no longer links to the state at which it was when I wrote the first part

There are some interesting insights that come from the data regarding the aftershocks and the years of seismic events since. I began reading into news articles on the process of rebuilding Bhuj itself, and another nice angle to look at would be how the city has become better at responding to the minor quakes that have struck since.

If the idea sounds good enough, I'll focus on completing the story, and then more on details like how the map looks, what data to show when, what the right POV for each plot point should be, and then packaging it as a small website in and of itself.

---

Other (unexplored) ideas

1. The correlation (real or imaginary) between real earthquakes and earthquakes in fiction (books, film and TV)

I've made a list of a few major earthquakes in fiction that have been based on real events, and few that have not. The idea moving forward seemed vaguely interesting, but given it was incorporating another dataset entirely, I decided not to go ahead with it for the moment.

2. With no faults but our own: Looking into man-made seismic events

After stumbling upon the Pokhran easter egg in the USGS dataset, I thought it would be nice to look for earthquakes that have happened far from natural faults, and see what might have caused them. Some immediate examples that came to mind were other Nuclear Tests (most recently in the Korean Peninsula) conducted by other countries. Over the last decade, more powerful mining and drilling equipment have also started having measurable seismic impact, and it might be possible to locate these sites purely by looking at the incident data from USGS.

I might also be able to find examples of football fans causing an earthquake by jumping too much.

---

Bhuj 2001: A Data Story

So the title links to the current version of the scrolly-telling article. For visual reference, it looks like -

Before getting here, I tried out a few ways of bringing in interactivity within the Map. This was attempted by getting the earthquake data directly from the USGS website through code, and generating the layers inside Mapbox GL. However this turned out to not be the best way of going about things, and after discussing with Rasagy, I decided to leave out map interactivity beyond the usual zoom + pan, as the scroll on the right would be engaging enough for a reader.

I built upon the basic scrollytelling example given on the Mapbox website. The article itself is broken into multiple chapters. When a chapter reaches the middle of the screen, it comes into focus. Now each chapter has two things attached to it:

• Location Data: What coordinate, zoom level, pitch and bearing correspond to the required view, and how long of an animation should be taken to get there (also the curve of animation itself)
• Layer Data: What layers of information should be visible for a given chapter

The story itself isn't as compelling as I'd like it to be. I limited it to 10 chapters because any more would feel too long without added visualisations/interactivity.

This required a bit of js and css to get working. It should do fine on most laptop/desktop screens, but is not good for mobile phones yet.

While the code part itself perhaps isn't the most complicated thing, I spent a fair amount of time fine-tuning parameters of the map itself in Mapbox Studio. Odds are most of those changes aren't going to register on a reader's mind, but perhaps that's the point. Examples include -

• Setting labels for the earthquake points, manually adding captions in some cases, and figuring out the right zoom level to show or not show them

• Figuring out a meaningful formula to map magnitude to radius and opacity (still not quite satisfied with the result)

• Redefining fonts across zoom levels and label hierarchy levels

As a bonus towards the end I was able to find GeoJSON files for the Gujarat State and Bhuj District, so I included those as layers as well (can be seen in the introduction of the page).

Things to work on moving forward -

• Fixing the narrative, coaxing meaning out of the post-Bhuj earthquake data
• Making a more suited visual style for the map, and adding better labels
• Experimenting with and adding interactivity, and deciding where it might add value
• Porting it to work on mobile devices

The code in it's most updated form can be found here.

That's all for now, Rishi

[aishaanam]

From my Volunteer experiences in BHUJ, Gujarat. As a volunteer in an NGO, situated in BHUJ, which was established just after the Earthquake in 2001. I have visited many areas of Kutchh, observing and hearing stories about the Earthquake that shook the region in 2001. Will be exploring the places, which have the highest impact and the ones that didn't and the probable reasons also throwing some light on the how vernacular architecture survived while most damages had occurred due to collapse of modern structures.

Major Events Indus Valley Civilization - River Indus changed its course Earthquake of Kutch - 1819 - ( Bhuj Earthquake - 2001

### Revised Intent One of my friends told me, despite the epicentre being at Bhachau, its devastating effect could be witnessed to far off cities like Lakhpat, Bhuj. So from the data available, I have tried to analyse the if depth & magnitude have any relation with it. But it eventually the main reason came out to be the Faults in our Land

**Revised concept of story telling -

Trying to understand the relation between the depth & magnitude of the earthquake and is there another factor which also affects the damage.**

[maulicule]

Rather a lot of ideas, but foggy roads ahead...

I've begin with a lot of enthusiasm, loads of ideas, but not much in terms of clarity... some of the ideas coming up are:

• Not One but Many: Looking at the distribution of earthquakes in a particular location across time, initial shock waves leading up to the 'main event' and what follows - possibly shifting epicentres in a close series of earthquakes, etc.
  • Direct and Open: Correlating magnitude, depth and impact of major earthquake events in record (filtered on magnitude)
  • Whispers and Yells: A map of recent global seismic activity, mapped to magnitude as a possible means of looking at areas of activity - which areas are quieter? which are loud? may be mapped through time to look at trends and movements in the 'loud' zones
  • Myth-busting: San Andreas / Pacific Ring / Bermuda Triangle - specific locations mixed with urban legend and glorified in pop culture - do they have any basis in earthquakes?
  • Poles Apart: A comparison between the two geographic poles in terms of seismic activity (I'm assuming there is some significant data available). Alternatively, could look at two maps which show the 'polar opposites' or antipodes of each other at the same time, mapping the places at opposite ends of the world to see if there is any correlation
  • A History of India: Looking at all the different major seismic events in India that have happened over time, their cause, magnitude and impact

A lot of possible visualisations, but not much in terms of a story. So let's think afresh.

Ideating 2.0

Tried to look at trends and then narrow down to specifics:

Location-based

Picking up one tectonic plate/area of seismic activity and tracing seismic activity through the ages. Possible Candidates:

• Juan de Fuca plate
• Chile
• Indian Plate
• Philippines Plate
• Cocos Plate
• Pacific Ring
• San Andreas Fault Image from here

Event-based

Picking up one major (recent) seismic event and tracing the activity leading up to it and the aftershocks. Telling a single story with data. Possible Candidates:

• Indian Ocean (2004)
• Tohoku (2011)
• Haiti (2010)
• China (2008)

Cause-based

Earthquakes caused by non-tectonic causes like due to fracking, dams, reservoirs, pumping gas, waste etc. nuclear tests, and other causes a.k.a. HiQuakes or human-induced quakes a.k.a. Induced Seismicity

Year/Date-based

Global/Large-scale Seismic Activity in a particular time period, especially early 2009 to late 2011

These are, again, a lot of ideas. But they now have structure, and an emerging pattern. I can see some especially interesting narratives possible, such as:

Indian Ocean (2004)

The story of the Indian Ocean earthquake (2004), with a chain of event leading up to it, including some older quakes in the region, minor 'premonitory' tremors, foreshock, main shock (principal event) and aftershock, possible fallout/major earthquakes in the region since.

The Seismic Frenzy

Across 2009 to late 2011, seismic activity displayed an increase in number of earthquakes worldwide. The death-toll was also high for these years, with notable earthquakes in the series such as Sumatra (Sep 30, 2009), Haiti (Jan 12, 2010), Chile (Feb 27, 2010), China (Apr 14, 2010), Tohoku (Mar 11, 2011) to name a few. A narrative through time around the world in this high-activity period could tell an interesting story

Table from here

Gathering Data

This was more of a filtering step than searching for sources as @rasagy and Prof. Venkatesh made it quite easier by providing this amazing resource. This is the USGS Earthquake search tool that lets us filter the data and download the result as a geoJSON, CSV and a bunch of other quirky formats like QuakeML. It even let me see my data plotted on a map (no frills): This was when I'd asked for high magnitude (6.5+ on the Richter Scale) shallow earthquakes (epicenter <60 km from earth's surface) across the world in the timeperiod of Sep 2009 to Dec 2011

Tried something quickly with Flourish... pretty easy to understand the tool and needs close to no coding 😅 The version liable to change is here.

The visualisation was imagined so as to resemble the old pen and paper type seismographs, with a faded parchment-like feel to it. Looks somewhat like this yet... not a whole lot done in terms of visual finesse and clarity, but decided to use the basemap orientation which would best suit the data (i.e. Pacific Ring of Fire in primary focus):

The Aha! Moment

It was when I was playing around with the slider that I noticed... there's some interesting activity in the Indonesia area, apart from the notable Japan, Haiti and Chile regions which had major events in this period. Just curious, decided to look at slightly longer term data from the Andaman/Indonesia hotspots, and bumped into the 2004 Indian Ocean Earthquake, which had its epicentre just off the west coast of Sumatra.

The intriguing bit about this data was that the historic 2004 Andaman-Sumatra earthquake, which has been one of the most devastating natural calamities of the past 20 years, seemed to have come as a real shock! (no puns intended) Digging a little deeper, the data for a month before the earthquake showed a deceptively calm seismicity, with nothing out of the ordinary. The principal event was sudden, unexpected and almost unprecedented - with a magnitude of 9.1!

The month following this main shock, however, was filled with aftershocks, with the rate of seismic activity levelling out at higher than what it was before. Not only that, I was sure I could develop some trends of the same geographical area over the following years, to get a picture of what the event had triggered. Or maybe even the rest of the world? Curiouser and curiouser. This could prove to be an interesting story to tell!

Taking up the initial dataset of Nov 26, 2004 - Jan 26, 2005, a month on either side of the principal event, I plotted it on Flourish.

The page on Flourish is here but might get updated in the due course of things.

It currently looks like this:

Dec 26, 2004 Sumatra, Indonesia After a month of light background seismic activity, what seemed to be a regular December morning brought along a wave of shock for Indonesia and its neighbouring countries. Almost as if from the blue, a magnitude 9.1 earthquake occurred, just off the west coast of Sumatra. It set off a volley of aftershocks along the fault line between the India and Burma plates. The immediate result of this was a huge tidal wave, while the seismic fallout continued for about a month.

Currently, I've tried a darker background and coordinated text so as to better contrast the earthquake categories (sorted as per magnitude classes). The magnitude has been rescaled and moderated to give a slightly more true-to-reality mapping of how the earthquakes can be compared:

To compare two earthquakes in terms of shaking, you subtract one magnitude from the other and raise 10 to that power: 10^(M1-M2).

Meaning to say that a magnitude 9.1 earthquake would shake 1000 times as violently as a magnitude 6.1 earthquake, which in itself is pretty scary and destructive!

The scale needed to be representative, but while giving due significance to smaller multitudes as well as the single large occurrence. I tried this by toning down the size through the square root method, which would keep the differences between the sizes of the plotted circles appreciable and logically correct, but all the plotted datapoints would be reasonably visible in the frame. That is to say, its the area of the circle which corresponds to the intensity of the quake, and not radii.

I might change this late if it feels too arbitrary, but I believe that such a representation brings out the differences more clearly than expecting the viewer to compare areas logarithmically!

Went back to the parchment paper and seismogram visual language to incorporate more information with lesser visual noise. Cleaned up the data and iterated through the story a number of times before finalising on this.

Created with love and data, Maulashree

[kaishwary08]

Aftershocks of the Aftershocks

March 2011, Japan was shaked by the Great Tohuku Earthquake, measuring 9.1 on the Richter scale. Earthquake triggered Tsunami of the Pacific coast of Tohuku, causing damage to lives and property. Tsunami, also caused the Fukushima reactors to melt down, exposing several to the radioactive heavy metals and contaminating soil and water.

2011, could be called Year of Earthquakes for Japan, where they witnessed more than 19000 earthquakes in that year, most of them measuring less than 6 on Richter Scale. The country received about 60 aftershocks more than magnitude 6, and 3 aftershocks more than magnitude 7. As per USGS Seismologists, a rule of thumb for the magnitude of aftershock would be 1 magnitude lower to the Magnitude of main quake. Though the number of aftershocks get lower the magnitude, their frequency of occurrence decreases (given there is no larger aftershock than the previous), and may be experienced even months after the Main quake.

Time period: year of 2011. Area: Japan Richter Scale: 5+ Medium: Please suggest.

unfiltered ideas b: could be to link it with Pacific Ring of Fire and showcase the relatedness for earthquake occurring on the Pacific Plate. And how California could be affected. Reference.

unfiltered idea c: What if we could energy released in a tremor could be visualised?

Idea 2 Mapping tremors caused due to Volcanic Eruptions. Comparing Seismic activity for Iceland, Indonesia and South America across years.

Time Period: 2010 (Eyjafjallajökull, Iceland Erpution) to 2019 (Mt. Sinnabung, Indonesia)

Execution Data set for year 2011 was selected and was plotted for a geo-visualisation. The color corresponds to the range of the magnitude of earthquake, whereas the size corresponds to the magnitude of earthquakes.

Thinking that some interesting insights could emerge from the mapping depth corresponding to the magnitude and dates.

Work in Progress could be found here

Both these visualizations were made against a time line giving a sense of occurrence of the earthquake and bounded them in a dashboard for visualizing them together.

To bind them to story, an combination could be look at earthquakes above mag 5, which had hit Japan since there last Major earthquake in 1998. Idea was to show the sudden outbreak of earthquakes right after the Tohoku 2011, and the time it took for earthquakes to diminish.

Whilst limitations of use and share in Tableau, I decided to switch to Flourish to curate the story. Several iterations and restructuring of parts enabled me to deliver the output..

Thanks Aishwary

[sargamprakash]

Does prior experience affect the perception of an earthquake?

I was specifically looking at the data of the 2004 Indian Ocean Tsunami and the 2001 Bhuj earthquake when I noticed something strange. The Andaman earthquake had a magnitude of 9.2 on the Richter scale while Bhuj had 7.6, but on the Mercalli scale (which measures the perceived shake of the earthquake), Andaman had a IX, while Bhuj had a X (maximum).

Then I took the data of the past 10 years of earthquakes leading up to 2004 for Andaman and 2001 for Bhuj and found that Andaman has low-intensity earthquakes very frequently and while Gujarat does not.

This leads to the question of do people get used to minor earthquakes such that in case of a major one, they don't perceive it to be as bad?

P.S. I have to look for more places to test out my hypothesis, but in any case, the Mercalli scale depends on the people in that particular area and their past experiences so it's very subjective. It would still be interesting to see how people rate their earthquakes across the globe and how battle-hardened they are.

Tableau link

https://public.tableau.com/profile/sargam.prakash#!/vizhome/Earthquakes_15680063511390/Story1?publish=yes

[avyayrkashyap]

IDEA 1

Looking at the correlation between earthquakes, hypothesising that an earthquake caused on one side of the earth results in an earthquake on the other side as well. Maybe also saying that the earth is doing a Mexican wave of earthquakes...

IDEA 2

Look at how manmade activities are causing earthquakes, and the frequency and magnitude of these earthquakes in non seismic regions... Look at a before and after kinda situation...

Proceeded with IDEA 2...

Decided to focus on the Sichuan Earthquake of 2008. The Sichuan region is located quite close to the Longmen Shan fault but is a generally a peaceful region. This was until the construction of the Zipingpu Dam. The construction which began in 2001 is widely considered to be the reason for the large number of earthquakes that have recently occurred in the region, including the massive one that occurred in 2008, claiming the life of over 65000 people.

Through a scrollytelling approach, I hope to illustrate the point that the dam has placed loads of stress on the landmass, leading to the earthquake. I will first introduce the reader to the event that took place, and then trace back to how peaceful the region in general was to illustrate the point.

I have a work in progress uploaded here.

I am in the process of figuring how I can call layers and maybe even overlay other visualisations to help illustrate my point.

---

Narrowed down the dam to the Zipingpu Dam in the Sichuan region of China. The construction of the dam is widely believed to have caused an increase in seismic activity in the region.

I have used a scrollytelling format to demonstrate this rise in the number of earthquakes in the region.

The story can be found here.

[SaiAnjan]

~~1. How many mobile calls are made at locations in a particular year where the earthquake happened

2. Taking the locations with high richter scale If we map the types of houses in those ares in the country of china
3. Compare satellite maps with earthquakes which shows any topographical changes~~

The above ideas needed additional data and these ideas are for another time when I get time to work on. For the current exercise I'm sticking to the following idea as discussed in the workshop:

What are the most recorded earthquakes in the last month and which network locations have recorded the most?

I started with USGS earthquake data for the last month. Then I filtered the data for earthquakes with the same location source. These location sources are coded with the area/state name. Each location source contains several contributors in the network area.

For ex: CI-California Integrated Seismic Network

The individual contributors in this seismic network are:

• Berkeley Seismological Lab
• California Geological Survey
• Caltech Seismological Laboratory
• California Integrated Seismic Network
• USGS Menlo Park
• USGS Pasadena

So I picked up the location sources(seismic network contributors) in the last month from the data sheet. These are the top 5 location sources which have recorded the most number of earthquakes in the last month with a magnitude ranging from 2.5 to 7 on richter scale:

Now I opened Mapbox to visualise the data. I used the monochrome style(Any style would have worked. I haven't thought about which style is better) to show the contrast of the circles. I added the earthquake data layer above the map layer in the stylesheet. Then I created a new tileset where I picked the location points for each location source in this tileset and imported this dataset into the stylesheet. Then I gave a colour coding for the location source layer and the earthquake layer. The radius of the location source is chosen more for easy identification. Then I duplicated the location source layer and added an SVG icon from the maki icon set to show the location sources. Link to the visualisation

Next steps:

• [ ] Adding text label to each location source
• [ ] Making a scrollytelling web experience using Mapbox
• [ ] Showing the metadata at each location source(highest magnitude, lowest magnitude, no. of recordings)

[GauriTillu]

The country that never falls The visualization story that I came up with for this exercise was about Japan. Japan undergoes shocks of multiple earthquakes every year. However, the country gets itself together and moves forward on the path of progress. The visualization that I imagined for this story was a bar graph along with the map. The bar graph would progress with time, and the columns would seem like a cityscape that keeps on growing with time. This would be a metaphor of Japan developing and progressing irrespective of the setbacks it faces Area: Japan Richter Scale:5+

Collecting data: I collected the dataset from www.usgc.gov (U.S. Geological Survey) for Japan for the year 2018 to 2019. The magnitude of the earthquakes considered for the dataset was equal to or greater than 4.5 on the Richter scale.

Visualization Tool used: Tableau

Final story: Japan, the country that never falls...

Furthering on Japan's story, I looked up the strongest earthquakes on the Richter scale In the last two decades, there have been 21 extremely strong earthquakes measuring more than 7.5 on the Richter scale. A few of them have also caused tsunamis to be formed as they occurred on the outskirts near the sea.

The final visualization was made in two applications:- datawrapper and flourish

Data viz in flourish: This is a timeline of all the earthquakes that occurred in Japan in the last to decades with a magnitude of more than 7.5 on the Richter scale

Data viz in datawrapper: This data viz shows all the places where these earthquakes occurred.

The visualization story can be found here

[dhirajdethe]

Is there any impact of earthquakes in neighboring country Nepal on India?

When I tried to look at the earthquake data of India and the region around India of the last five years, I found Nepal is affected most and being close to Nepal, north-east part of India have also been affected. So I am curious if the earthquakes in Nepal have affected in some ways, the parts of India.

Region: India & Nepal Magnitude: 2.5+ Span: Last 5 years

I tried to find and extract data related to different events happened in India related to Nepal after the major earthquake of 2015 to see whether there is any impact of earthquakes of Nepal on India. but the attempt went miserably wrong and I didn't get any useful data and ended up wasting a lot of time. therefore I decided to look upon the dataset in Nepal only and built a story on earthquakes occurred in Nepal in the year 2015.

The Scary year for Nepal: 2015

Tool used: Tableau Process: I collected the earthquake data of the last five years of Nepal(magnitude 4.5+) and tried to analyze using Tableau tool. I found major earthquakes occurred in the year 2015 that damaged Nepal very badly. Therefore I decided to make a story on the most tragedic earthquakes occurred in 2015 in Nepal.

The story progresses in 4 main steps of visualization,

1. Visualization of earthquakes of Nepal in the last five years

2. Visualization of earthquakes in the year 2015(Distributed along months in a year)

3. Visualization of earthquakes in April 2015(Distributed along days in a month with magnitudes)

4. Visualization of earthquakes in May 2015(Distributed along days in a month with magnitudes)

Link for the story: https://eu-west-1a.online.tableau.com/#/site/dhirajdetheidc/workbooks/163205?:origin=card_share_link

[harshitsatija]

Rethink : Settling in Popular Countries

I was looking at the data of different popular countries and noticed that some countries had continuous earthquakes for a decade and then the number reduces significantly. So, the idea was to collect the data for popular countries to present their earthquake histories and show people whether they are earthquake-prone countries or they are currently going through a phase, etc.

Since the data required is of a longer period, I choose the time period to be 1950 - 2019 I wanted to show the existence of earthquakes. Thus, I choose a very low value, i.e. 2.5.

Collecting Data

First I took the data of major countries and tried to find some pattern.

After a short while, I noticed a pattern in France and Spain. The number of earthquakes is reducing significantly after 2009.

This leads me to collect data from the 1950s and I again noticed a pattern.

This leads me to find out about the whole of Europe. I collected the data of earthquakes only above 4 magnitudes. The pattern here was missing.

Then I searched for total earthquakes, but a shorter period and noticed the same pattern. ‘

The conclusion I drew was that there was a clear reduction in the total number of earthquakes. However, the reduction was mainly due to the low number of earthquakes in the range of magnitude of 2.5 to 3.5.

Building a Narrative

I wanted to go with something catchy and something that represented the data. I didn’t want the viewer to draw any factual conclusion from the data and just wanted to present this anomaly in the data.

The headline I decided was “Has Europe Solved Earthquakes?”. I chose this as it was clear from the headline that the data was only about Europe. And since it was absurd, as a country cannot solve earthquake, it worked as one wouldn’t draw any conclusion from the story. Thus it fits in perfectly. To build up the story I started with the rise in earthquakes in the 1980s in the whole of Europe. To build up more, I also put in the data of France and Spain. I Ended with showing the sudden stoppage in the Earthquakes and also pointing out the absurdity of the situation.

The story is posted here

Now the plan is to find more anomalies in the data like this. Thank You.

[rohanjhunja]

DAMMED QUAKES

The closest earthquake prone area to us is Koyna. Earthquakes in the area began to increase after the construction of the Koyna dam.

A devastating 6.3 magnitude earthquake in Koyna, where a grand, huge dam was built in 1962, caused much misery. But, more importantly, it shattered a long-held belief that the Deccan Traps were solid rock and not prone to earthquakes.

The needle of suspicion pointed to the reservoir. Sections of earth scientists firmly believed it had triggered the trembler. A raging debate ensued on whether reservoir-induced seismicity (RIS) is a cause for earthquakes. https://www.thehindubusinessline.com/news/50-years-after-koyna-lessons-from-the-mega-earthquake/article9988376.ece

Globally, about 120 cases of triggered earthquakes have been identified. At four places they were above 6 magnitude. These include Kariba on the Zambia-Zimbabwe border (1963), Hsingfenking in China (1961), Kematsa in Greece (1966) and Koyna, which was the biggest, in 1967. The total area impacted was 20 km by 30 km, with no other earthquakes around 100 km. These earthquakes originate much closer to the surface (<40km) than tectonic quakes.

Directions

• Earthquakes caused by Dams
• Risks of Dam bursts by Earthquake

Shaking Myths

‘The earthquake was a good sign’ because it occurred during full moon: Bali governor Having visited Bali around the occurrence of a major 7.5 earthquake, volcanic activity and tsunami at neighbouring Sulawesi, I was surprised to see how normal this was to the locals. Moon-quakes are thought to be caused by stresses on the plates by tidal changes. The direction would be to investigate myths associated with earthquakes across Balinese history or across the world.

Alternate Direction Shake Map Paired Earthquakes

Taking the quakes to Q-GIS

Dams of the world A list of all the dams in the world was visualised in QGiz. This process was fairly simple as QGiz can handle a wide variety of filetypes.

Buffer Zones of Dams >20,000 Maximum Capacity Dams with a maximum capacity similar to or larger than Koyna at 20,000 were selected.

Earthquakes in India from 1900-2019 USGS data on earthquakes over this 1000 year period were visualised and intersected with the buffer zones of 20km around the large dams. Importing new datasets only required matching of the projection types to ensure that the maps aligned.

Buffer Zones of Fault Lines Cases where earthquakes in zones of major fault lines were separated as we are interested in cases with a clear connection to Reservoir Induced Seismicity.

Koyna Region (100 km radius) As the first example, the Koyna dam region was used to filter seismic activty between 1900 and 2019.

This was used to identify two sets of activity, before the dam was constructed (1957-77) and after (1983-2003). A wider range of earthquake magnitudes in these periods were then selected.

A set of identified RIS related activity was later found and visualised on the map. The final visualisation allows you to explore the seismic data before and after the construction of these dams.

The Visualisation of RIS data is hosted here

[eeshani]

IDEA 1:

Earthquake & Volcanoes:

Vicinity of earthquakes and volcanic eruptions. This will explain the magnitude of the earthquake and expected volcanoes in future. This will help in understanding the relation better and take precautions.

Below are the datasets for volcanic eruption.

1. https://www.kaggle.com/texasdave/volcano-eruptions#volcano_data_2010.csv
2. https://www.ngdc.noaa.gov/nndc/struts/results?type_0=Like&query_0=&op_8=eq&v_8=&type_10=EXACT&query_10=None+Selected&le_2=&ge_3=&le_3=&ge_2=&op_5=eq&v_5=&op_6=eq&v_6=&op_7=eq&v_7=&t=102557&s=5&d=5

Final design: https://eeshani.github.io/Earthquakes_Volcanoes/

IDEA 2:

Continents shifting and mountains forming

Earthquake occurence study across various areas and the plate movement studied. The dataset has pictures showing the movement and shifting of plates and this can be correlated with the earthquakes happening there because of these tectonic movements.

https://sos.noaa.gov/datasets/plate-movement-200-million-years-ago-to-today/ ftp://public.sos.noaa.gov/land/plate_movement/

[gyanl]

Where do the worst earthquakes in the world happen? How do people prepare for them?

A scrollytelling style visualization of the top 8 most devastating Earthquakes recorded, combined with some additional editorial content about what sort of impact it had and the disaster mitigation steps were taken before and after the event.

To start with, I looked only at the Earthquakes that were 7.5+ on the Richter scale over the past 10 years.

There were some patterns emerging, but I felt there were too few points to generalize, so I changed to magnitude 8+ over the past 20 years.

Eventually I settled on 8.6+ Richter scale earthquakes over the entirety of the USGS database, which gave me only 8 earthquakes to work with. I thought this was a good number to try and weave into a story.

My final visualisation is a curated scrollytelling list of the 8 highest ranking earthquakes on the Richter scale, which can be viewed here: www.gyan.com/earthquake

[seemskk]

Reservoir induced seismicity in Idukki reservoir

Background study

There has been several reports over the years on Idukki dam induced micro tremors in the western ghats region of Kerala. National registry of dams, 2018 shows a list of 62 dams in Kerala. According to a study, of which 43 dams and reservoirs are located in the highly fractured Western Ghats, 21 are now highly prone to dam induced seismicity after the high intensity 2018 rains. This calls for better management of dams in Kerala.

Scientists at the Centre for Earth Science Studies (CESS), Thiruvananthapuram, who have been conducting studies in the Idukki region since 2001, have identified “an active fault zone” near Idukki. In the four months from July 26 to November 26, 2011, there were about 26 low-intensity tremors in this area. But they said such minor tremors also prevent a major build-up of pressure within that could cause bigger earthquakes (above 4.0 on the Richter scale).

News reports:

An earlier report on 2011 tremors from Frontline and recent developments on 21 dams is reported in livemint. Some of the research papers pointing to the seismic activity can be found here and here.

Idea: To build the context, the idea was to map the 43 dams in Kerala. It was until the last flood, many in Kerala actually never knew the high number of dams in the sensitive western ghats region. The 21 dams prone to dam induced seismicity could then be filtered from the main map. It was also to show the rain fall during the high point of monsoon 2018 in these regions.

Data collection: One of the first attempts to find the micro tremors at USGS website failed since none of the tremors were visible in their record. 43 dam locations (wikipedia source) and rain fall for aug 15 -17, 2018 were collected and merged to one excel sheet.

Rain intensity and Dam locations

Attempt 1: The first attempt was using Geojason and Mapbox studio, the dams were mapped and layers were created in Mapbox. The micro tremors near Idukki dam were also mapped with great precision. But what I could not complete is the rain intensity layer and interactivity to the map.

Attempt 2: Second attempt was using Tableau, better control and tool tip information were achieved. But still not happy with storytelling part and layout.

[akshayrpatil]

I would like to visualize the earthquake happened on Tectonic Plate's edge and the earthquake whose center is far from tectonic teeth. this will help to visualize the comparison of certain attributes

1. magnitude of the earthquakes.
2. depth of earthquakes (still thinking)

I started by exploring and playing USGS earthquake data for all the time. Also searched on ESRI ArcGIS open data based for earthquake data. I got CSV file for tectonic plates which I superimposed on based map layer in Mapbox. I tried to visualized earthquake data with Tectonic plates on base map.

after that, I shifted to ArcGIS software which is similar to QGIS. -P.S I have been using ArcGIS from 2012. for analyzing and visualizing Geographical Information database on the maps.

Now I opened ArcGIS online web version to visualize the data. I added the earthquake data layer above the basemap layer in the Content. added tectonic plate world file. Then I applied filters to segregate data.

after appling filters i got clear ourline on tectonic plates and filtered earthquakes spots.

After applying filters and setting map data, on click interactions, I made the final version of Web-based App for data story.

Link to Data story app click here

[SONIBHAWNA]

IDEA Visualising Indian Earthquakes for past one year across various regions to show the most affected areas and how much they are affected. I wanted to show how depth and magnitude differs and are linearly unrelated to each other. The information is more useful for people who are well in geography but also show difference between the two to normal people who are looking at the picture as whole. 1.) I visualised on map first then showing the bar graph for each and every place earthquake happened.

2.) Explored a software a bit to show both can be made interactive to show connection between all of them together.

3.) One other thing which I found amusing how people can explore with the magnitude and depth to find out which state or city it occured.

[shraddhadhodi]

Nihon no Jishin

Earthquakes in Japan is not a new story. Japan has a long history of earthquakes, sometimes with shock and sometimes with tsunamis. The earthquake-like mild tremors hit japan on a daily basis and so the country has always shown the resilience of fighting it back, on daily basis now. The geographical location of Japan is in such a way that it sits on top of 4 large tectonic plates that seems to move and grind together making the earth's crust vibrate and hence give rise to Earthquakes.

With this visualization, it could be seen that which region of japan is more prone to earthquakes. Also, the maximum no. of impact could be seen as well as the depth of the magnitude too.

Here is the link of the visualization: https://public.flourish.studio/story/88706/

[AshmiK]

Idea 1

The earthquakes that happened but didn't happen. This could also help in explaining the Richter scale. How on average, Magnitude 2 and smaller earthquakes occur several hundred times a day world wide. Major earthquakes, greater than magnitude 7, happen more than once per month. "Great earthquakes", magnitude 8 and higher, occur about once a year. Also if I could also somehow relay the fact for each step up in magnitude an earthquake releases 30 times more energy.

Idea 2

Earthquakes of certain magnitude ( 7 or more), the no. of seconds it lasted and the damage it caused.

Going ahead with the first idea, decided to just visualize the earthquakes happening over a period of 48 hours (to grasp the sheer number of earthquakes happening over such a small time frame). The start and end time of the data set were from 2004-12-25 (00:00:00) to 2004-12-26 (23:59:59). This specific time was chosen to get a data set which had earthquakes of all magnitudes (9.1 magnitude earthquake took place in Sumatra on 26th December, 2004). Maps for various range of magnitudes were created using Tableau (would have been better to use Mapbox Studio) to visualize the frequency of earthquakes happening.

The National Earthquake Information Center (NEIC) records an average of 20,000 earthquakes every year (about 50 a day) around the world. There are, however, millions of earthquakes estimated to occur every year that are too weak to be recorded. Here is a series of images showing the frequency of various types of earthquakes.

Micro earthquakes(0 - 2.9)

Minor earthquakes (3 - 3.9)

Light earthquakes (4 - 4.9)

Moderate earthquakes (5 - 5.9)

Strong earthquakes (6 - 6.9)

Major earthquakes (7 - 7.9)

The Great one (8 and above)

Here is a simple looped animation of it.

[harsuyash]

Some ideas to begin with:

1. Earthquakes that never happened Visualizing the errata from USGS, of the erroneous notifications and uploads about earthquakes that never happened Rejected: A bit of a shallow story maybe just for fun

2. Rendering different earthquake maps with the art style of the era in that region Rejected: A cool project I would like to pursue but more of an Info graphic than a Geo-viz and seemed unsuitable for the brief

3. 5 element quakes Categorizing seismic activities that are not just related with earth but also fire (volcanic eruptions), Water (Tsunamis) and Wind Rejected: Volcanoes and tsunamis seams feasible but the effect of wind on seismic activity is debatable. "Very large low-pressure changes associated with major storm systems (typhoons, hurricanes, etc) are known to trigger episodes of fault slip (slow earthquakes) in the Earth’s crust and may also play a role in triggering some damaging earthquakes. However, the numbers are small and are not statistically significant." says an article from USGS

4. Desperate times... Correlating earthquakes with contemporary events in the field of seismology Did a particular earthquake lead to a discovery or invention that would help us understand earthquakes better? How does a series of earthquakes and their frequency in a region effect seismologic research in that region? Did certain great research or products follow a great earthquake or were influenced by them?

This topic intrigued me and I tried to get info about certain studies and which earthquakes were contemporary to it. I was fortunate to found one such event - The great 1906 San Francisco earthquake on USGS that fueled a great amount of research and analysis in this field. Its importance comes more from the wealth of scientific knowledge derived from it than from its sheer size. It had led to the realization of "theory of elastic rebound". I also read about the San Andreas Fault and the ample amount of research and experiments that have been done over it. The San Andreas Fault is important for the following reasons: ~It has a high frequency of seismic activity of magnitude 3 or greater. ~It is a part of the ring of fire and at the edge of a tectonic plate ~It hosts important cities like San Francisco, Los Angeles ~It is surrounded by eminent research institutes including UC Berkeley ~We even have a movie to scare everyone about a possibility of a mega quake in this region

I finally decided to focus on the

San Andreas Fault and the research and development around it

I found plenty of story material around it and relevant data from USGS to back it up. After making a list of significant earthquakes, I further narrowed down to Earthquake activities in -

Parkfield - the most closely observed earthquake zone in the world.

I found an interesting story about a pattern of earthquakes occurring here and the efforts to successfully predict earthquakes.

The Data chosen

Magnitude: 4.5+

Duration: 1850 - 2019

Location: USA

(location for Parkfield, Central California):

Latitude: 30N - 37 N

Longitude: 115W - 125W

This much revealed a story that was accompanied by some bare facts

The final Flourish story:

The Story of San Andreas

[PGPagar]

A large crack, stretching several kilometres, made a sudden appearance recently in south-western Kenya. The tear, which continues to grow, caused part of the Nairobi-Narok highway to collapse and was accompanied by seismic activity in the area. The Earth is an ever-changing planet, even though in some respects change might be almost unnoticeable to us. Plate tectonics is a good example of this. But every now and again something dramatic happens and leads to renewed questions about the African continent splitting in two. So I started gathering data for earthquakes in Africa continent and the results were showing that it might be a split of the continent

Video of the incident https://youtu.be/wO7s5zIhX6k

my flourish project: https://public.flourish.studio/visualisation/932552/

[akshathyanki]

My flourish project on earthquakes in California, USA. https://public.flourish.studio/story/124313/

[ameyanikose]

An attempt at visualizing the seismic activity across the planet with Richter magnitude above 6 and across a time period of a century using Mapbox Studio's Heatmap GeoViz. The Link to the project- https://api.mapbox.com/styles/v1/ameyanikose/ck30ymcfa0w861cqjf8f3b3jh.html?fresh=true&title=view&access_token=pk.eyJ1IjoiYW1leWFuaWtvc2UiLCJhIjoiY2syeWU4Z3p1MDZwMjNvbzR0YXpxYXA5dyJ9.ddTWyWDu0IqAb9UWt7I2oA#0.9/0.000000/-177.956520/0