Benjamin Strick Profile picture
Dec 10, 2021 20 tweets 17 min read Read on X
Today we release a database showing WHERE, WHEN and HOW more than 38,000 buildings were destroyed in Myanmar's Rakhine State between 2017 & 2020.

Here is the data: bit.ly/ocellirakhine. Here is an explainer thread on the work by @ocelliproject @KayleighRuiter & @C4ADS👇🛰️
The data has been turned into a visualisation and interactive website by @C4ADS & @Jack_Mrgln. We wanted to make sure that this data is accessible to all levels of researchers, so please do click around and use it to further your investigations & research ocelli.c4ads.org/datainteractive
First, WHO are we?

We're a group of volunteers (yes we do this in our spare time) specialised in mapping and databases. You can see more of what we do at @ocelliproject.

We focus on making publicly available geodatabases to help journalists, investigators and researchers.
WHY did we do this? One look at Rakhine on @googleearth shows scores of villages wiped off the map.

We felt a need to build a freely available dataset documenting every single building with a transparent methodology that could be replicated by anyone.
There has been numerous independent sources of research done in the past, but there's still so much work to do in documenting the horrific story of what was referred to as "a textbook example of ethnic cleansing" ohchr.org/EN/NewsEvents/…
This work is important given the current activity in Myanmar's North where numerous villages and buildings have been burned to the ground.

Through work we do at @MyanmarWitness we're concerned about what happened in the past & #WhatsHappeningInMyanmar
HOW did we do this project? We followed the six steps seen in the graphic below and we've made this workflow replicable for other subjects that require this approach to large-scale documentation.

Here is a link to our full @ocelliproject methodology: drive.google.com/drive/folders/…
Our workflow is based on dividing imagery analysis over a team. How?

We conducted an initial sweep of the area, layered a grid and divided the area up amongst volunteers to go through each square kilometre. We were able to cover 3,283 square kilometres in that manner.
WHAT did we find by doing this? You can see all of our results in this spreadsheet: bit.ly/ocellirakhine & data interactive: ocelli.c4ads.org/datainteractive.

The data interactive shows satellite imagery of before and after the destruction occurred.
As a result of doing this documentation we identified more 38,000 destroyed buildings, more than 24,000 of which had clear burn marks as seen in @googleearth satellite imagery.
We cross-referenced all of our findings with @sentinel_hub imagery. We used freely available satellite imagery in this project for transparency & replicability of findings.

Where possible, we've added timelapses of the destruction such as this one from Myo Thu Gyi
Building off the back of amazing work by @Nrg8000 & @elisethoma5 @ASPI_org (pageflow.aspi.org.au/rakhine-state#…), we documented significant land clearing and when new buildings popped up on top of where villages were destroyed.

You can see that analysis here: ocelli.c4ads.org/investigations
For example, here is one interesting location where villages were destroyed and land was completely cleared. This location is here:
goo.gl/maps/n9hbRegRY… around the Anauk Myinhlut and Al Le Than Kyaw area.
Observing areas like this is important, as we have also identified changes along the coastline and small encampments where villages were destroyed and cleared.
We've documented a lot more of new constructions that you can find in the second tab of the spreadsheet here: bit.ly/ocellirakhine

If anyone is doing further work on investigating Rakhine State - we hope this will be helpful.
If you want to get your hands on the KML and see the annotations on @googleearth yourself, the KML is in this drive folder: drive.google.com/drive/folders/…
We have made all of the data accessible in this drive: drive.google.com/drive/folders/….

In there you will also find before and after imagery of every site - we hope that this type of database might assist anyone looking for large datasets for potential development fields
This project would not have been possible without the countless days/weeks/months of dedication from @KayleighRuiter - give her a follow as she is an incredible analyst and makes an epic spreadsheet (here's the proof: docs.google.com/spreadsheets/d…)

Also a massive kudos to the wider volunteer team that worked on the @ocelliproject documenting change in Rakhine State: @hrgwilson1, Gavin de Ruiter, @anant_saria, @TomTomjarvis, @bullock_milo, @JKPoupon, @jadolecek, Paul Gruet, Tineke Ellens
There's a lot of resources in this thread, all of which are on the @ocelliproject & @C4ADS website here: ocelli.c4ads.org.

We hope this work inspires future investigations, research and strengthens justice & accountability for what happened in Myanmar's Rakhine State.

• • •

Missing some Tweet in this thread? You can try to force a refresh
 

Keep Current with Benjamin Strick

Benjamin Strick Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @BenDoBrown

Dec 18, 2024
I have received a number of messages asking about the location of this IL-76 (RA-76502) from Russian Aviacon Zitotrans (usually responsible for arms cargo).

It is not landing in Sudan (alleged in the comments), but is landing in Faya-Largeau in northern Chad. More in thread 👇 Image
I identified an original and more clear version on TikTok (not hard if you want to find it). In it, we can see the registration number of the plane much more clearly, as well as the dune and cliffs in the background.
The registration, RA-76502 that had been owned by JSC Aviacon Zitotrans, named by US as "a Russian cargo airline that has handled cargo shipments for sanctioned Russian Federation defense entities". Image
Image
Read 4 tweets
Dec 1, 2024
This festive season I’m sharing a video every day for the next 24 days showing useful OSINT tools & techniques. Creating this OSINT Advent Series has been a lot of fun and I hope it’s helpful for the ever-growing OSINT community! 🎄👇
1. Searching Facebook with WhoPostedWhat.
2. Using AI to identify a car model in an image with Carnet.
Read 13 tweets
Oct 10, 2024
Today is #WorldMentalHealthDay. As digital investigators we're often not experiencing what we see online, in real life, but it can still affect many.

So to keep doing the important documentation and investigative work online, here's a few practical steps you can take.👇
1. When sharing graphic content with colleagues and friends, remove those previews and give a little graphic warning indicator. There's always a little 'X' in the corner to remove preview. Image
2. Consider changing the settings on your social media platforms so you're not absorbing horrific content while doomscrolling. You don't have to filter it out, but at least stop the autoplay. Image
Read 10 tweets
Sep 18, 2024
I was recently on a flight across Australia when I spotted this massive figure on the ground. It led me on a digital journey to find out what it was, how it got there and who made it.

I'm going to explain a bit more about what it is in this thread, and how I found out. 👇🧵
Image
Image
Without internet on the plane, I made a screenshot on my phone of the location (yes it was on flight mode) and later used flight tracking to pinpoint exactly where I saw it.

Using @flightradar24, I traced the path my plane took and found the spot!
Image
Image
Next stop: @googleearth 🗺️

I zoomed into the area and there it was — this huge humanoid figure etched in the Australian outback. Measuring 2.7km tall and covering an area of more than 1.7 square kilometres, it's a sight to behold.

Here's the location: -29.527084, 137.465400
Image
Image
Read 8 tweets
Sep 2, 2024
Despite our reporting last week on the fake network of pro-Trump MAGA accounts, there appears to be many more accounts actively posting the exact same content. This one, @brenda_otto_ with 18k+ followers, is stealing photos from an Australian Instagram model in Queensland.

Image
Image
Here's another fake MAGA account, @Tracy_Miller044, stealing images from a fashion blogger (a popular one too).

@Tracy_Miller044 - if you want to reach out and talk about your work, why you're running these campaigns etc, I'm open.
Image
Image
Here is @Sarah_Hickey__

Sarah is stealing the photos of a Czech Instagram influencer to create a persona claiming to be a MAGA republican conservative with 40k+ followers and a blue tick (which means it's apparently not misleading).

@Safety @GlobalAffairs @elonmusk
Image
Image
Read 4 tweets
Aug 28, 2024
A network of fake accounts are posing as young American women, posting pro-Trump content and disinformation, but they’re hiding behind, and manipulating, the images of European fashion influencers.

Our latest investigation at @Cen4infoRes. Details in this thread 🧵👇 Image
Our full analysis can be seen at @Cen4infoRes here: . We also collaborated with @CNN to dig out the human stories behind those who had their photos stolen

Let's look at the some details below 👇info-res.org/post/unmasking…
edition.cnn.com/2024/08/28/eur…
One of the accounts is Eva. She lives in the US, and likes hanging out at the beach and posting to her 5000+ followers on X. Eva posts strong opinions against LGBTQ people and the US Democratic party and is a loyal supporter of former US president Donald Trump. Image
Read 21 tweets

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Don't want to be a Premium member but still want to support us?

Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal

Or Donate anonymously using crypto!

Ethereum

0xfe58350B80634f60Fa6Dc149a72b4DFbc17D341E copy

Bitcoin

3ATGMxNzCUFzxpMCHL5sWSt4DVtS8UqXpi copy

Thank you for your support!

Follow Us!

:(