We're sorry but this page doesn't work properly without JavaScript enabled. Please enable it to continue.
Feedback

Building the Bloomberg for Climate Data with FOSS

00:00

Formal Metadata

Title
Building the Bloomberg for Climate Data with FOSS
Title of Series
Number of Parts
351
Author
License
CC Attribution 3.0 Unported:
You are free to use, adapt and copy, distribute and transmit the work or content in adapted or unchanged form for any legal purpose as long as the work is attributed to the author in the manner specified by the author or licensor.
Identifiers
Publisher
Release Date
Language
Production Year2022

Content Metadata

Subject Area
Genre
Abstract
I am the Founder and CTO of Blue Sky Analytics, a Climate-Tech Startup using satellite-derived climate intelligence to power financial decisions. We provide datasets through API spanning flood, drought, wildfire, heat risk for monitoring, measuring and mitigating climate risk which can be leveraged for various use-cases. In 2 years, we have analyzed TBs of data, delivered 5 datasets & built platforms for data visualisation and distribution from scratch using FOSS technology. This has been a rocket-ship of a journey, chasing our mission of building a ‘Bloomberg for environmental data’. However, the not-so-secret sauce to achieving these milestones has been FOSS. We are often asked how we procure raw geospatial data and how much we spend on it. Thanks to the abundance of open data, our data acquisition cost has been 0. Due to the generous open data policy of amazing organisations like NASA & ESA, we have been able to build a business collecting TBs of data daily & crunching them into useful insights. This helped us scale our vision to build a global environmental data stack for tracking climate change in real-time. Moreover, before this data can be applied to climate mitigation, it needs to be analysed. This is true for any big data and today, less than 1% of global data is analysed. Given that satellite data is the most significant source of tracking climate variables, it became imperative to tap this source. We discovered that the path to providing environmental datasets was by building a powerful geospatial data refinery along with SpaceTime™ (https://spacetime.blueskyhq.in/) and our dev portal (http://developer.blueskyhq.in). There was limited infrastructure available to support the delivery of geospatial datasets so we built it, leveraging open-source tools like Postgres, QGis, GDAL, k8s etc. While we have proprietary layers to our models, as a team of young developers, data scientists and designers, many self-taught, our cultural ethos stands firmly with FOSS and we plan to be a leading contributor to FOSS for climate action. The most significant step for us in that direction has been providing annual, country-wise data on biomass emissions to the Al Gore-led Climate TRACE (https://www.climatetrace.org/) inventory that can be used by the public via CC-BY-4.0 framework. As our business expands, we aim to open-source tools & innovations we have internally developed while building our infrastructure and have started that journey with the Raster Playground. (play.blueskyhq.in) Climate change is the most pressing challenge of our times, throwing at us various questions that need to be answered. This is not possible without data. Data helps us understand the problem and quantify the risk to various assets. Open source tools and data have made it possible for 20-year-old data scientists to access sophisticated satellite data to understand the changing planet and answer these questions. BSA is a testament to the fact that fighting climate change is not possible without FOSS.
Keywords
202
Thumbnail
1:16:05
226
242
BuildingOpen sourceForcing (mathematics)Goodness of fitAnalytic setComputer animation
Enterprise architectureOperations researchEndliche ModelltheorieComputing platformVisualization (computer graphics)Decision theoryPersonal digital assistantTemporal logicImage resolutionFrequencyScalabilityMeasurementDevice driverIntegrated development environmentMachine learningOpen sourceComputing platformParameter (computer programming)Office suiteDatabaseQuicksortSlide ruleSet (mathematics)Repository (publishing)PredictabilityRemote procedure callInsertion lossInternet service providerProcess (computing)Multiplication signPerspective (visual)Covering spaceCompilation albumScaling (geometry)Computer animation
SatelliteElectronic program guidePersonal digital assistantFormal verificationOpen sourceEndliche ModelltheorieNeuroinformatikLibrary catalogoutputData modelIntegrated development environmentRepresentation (politics)Temporal logicContext awarenessVector spaceComputer fontVisualization (computer graphics)Dependent and independent variablesSoftware developerTrigonometric functionsPlastikkarteUniqueness quantificationPoint (geometry)WebsiteBlogLink (knot theory)Parameter (computer programming)MathematicsWeb pageMotion capturePolygon meshVotingType theorySatelliteSoftware developerSpacetimeFrequencyOpen sourceBasis <Mathematik>Graph coloringQuicksortOpen setPurchasingMathematicsSet (mathematics)Computing platformCASE <Informatik>Data managementInternet der DingeExtreme programmingParameter (computer programming)Macro (computer science)Proper mapEndliche ModelltheorieArchaeological field surveyDenial-of-service attackBit rateLevel (video gaming)Computer animation
PredictionPower (physics)Level (video gaming)Water vaporParameter (computer programming)QuantificationSurfaceSatelliteQuantificationFrequencySoftware testingPredictabilityMereologyComputer animation
Open sourceCoalitionGame theoryFrequencySoftware frameworkCore dumpMereologyCoalitionOpen sourceCASE <Informatik>Slide ruleComputer animation
SpacetimeInternet forumWater vaporGroup actionData integrityGeometrySicInteractive televisionComputer programPrime idealTheory of relativitySpacetimeMereologyIntegrated development environmentComputer animation
Computer animation
Transcript: English(auto-generated)
OK, let's start. Good attendance. People woke up finally. I'm the CTO and co-founder of Blue Sky Analytics. We are building Bloomberg for climate data with forced technology. Climate change is one of the biggest crises that humanity is facing. And I think open source is a very crucial tool
to solving this. Without open source, I don't think there is a solution out there. Coming to the next slide. We are a three-year-old company. We were founded in India. Now we are expanding. We have our office in US. I'm based out of Netherlands. And we are hiring a team in Netherlands. So if you're sort of looking for a job, please bring me.
Then we have a million in revenue. We are sort of scaling that up. We have 300 plus users. We have a team of 30 people in India. And we have built all this platform in-house using Phosphogee. You might have attended my colleague's talk yesterday about building a spatiotemporal database using Postgres, PostGIS,
and TimescaleDB. My friend, Michael, just gave me another talk about how we are using open source tools to predict fire emissions all over the world. And that's the sort of stuff that we do. Coming next, we all know that climate emergency is a problem. And it is the only problem with a deadline.
And we have something around next 30, 40 years to solve this. And it's a very pressing issue. And I think whoever is from Italy might have realized that it's getting hotter every day, every year. And that's a problem that we have to start looking. And from a financial perspective, from an insurance perspective, we have to start
taking this climate data and this changing climate into the risk and underwriting that we have. So just like pandemic happened, we were not prepared for pandemic. And we are not prepared for climate change as well. But we should be, because if we are not, then we will incur losses. And there will be sort of a recession that we are sort of looking at maybe.
We are a data set provider, by the way. So building Bloomberg, just like Bloomberg is sort of the repository for financial data, we are the repository for climate data on environmental parameters. I'll come on to the data sets that we are working on. We are all global, so we have a global coverage. So all the data sets that we build
are built using remote sensing, AI, and machine learning. And they have a global coverage. Coming next. So we not only use satellite data, but also IoT sensors, a lot of data from surveys, ground data, government data, all the data that we can gobble up, we gobble up the name of the ingestion platform that we have. It's called Yum Yum, so that it can eat
every kind of data that we have out there. We have built this data refinery that we have. Like, you know, the oil refinery was the, oil refinery is the thing. We have this climate change problem, and we are building this data refinery. We're just sort of antithesis to that oil refinery thing. We have this aggregation, we aggregate a lot of data. We have built proprietary AI models,
which use all this data. We have pipelines, and we build environmental insights using them. A whole lot of it is actually built on open source. Oh, damn. Then we have the data sets that we are producing is floods, emissions, extreme heat, wildfires, carbon capture, and ESG ratings. You can read what sort of use cases that we are serving.
So underwriting risk is very big. Risk models, asset pricing, and asset management is something really big. We have two platforms, one is called Space Time. Space Time is a sort of a, not open source, we are sort of thinking of open sourcing it, but we don't know how. We're a very small team. And this is the platform that we use
to visualize the data sets that we have. And here you can go to spacetime.blue-sky.hq.io, and then you can see what sort of data sets that we have. You can visualize on various parameters, change colors and everything. There's a lot of fun that you can do. A lot of open data that we build, we open source it, that is it is available on Space Time, publicly, on a macro basis. On a more detailed level,
you have to sort of purchase it from us. Coming to the next, we have all the data that we have is also available through APIs and through our developer portal, and this is all the details that we have. What is the frequency of everything? I've got it. All right, so I'll come to the testing part. This is the thing that we build. So we start began with the air quality data set
from India, and now we are working on water quality monitoring, surface water quantification. I think it's very relevant to what's happening in Europe, like all the lakes and rivers drying up, and we can track from satellites on a 15-day frequency what's happening on the ground so that people can better manage. We can't prevent these things from happening, but we can better manage them. And power plant emissions is something really big,
and this is, we can now monitor emissions and catch the offenders from the satellite. That's truly amazing, actually. And fire protection is something that we are working on as well. Greenhouse fire emissions, that's something. And greenhouse fire emissions, we were part of, we were founding members of Climate Taste. It's a global coalition led by ex-U.S. vice president Al Gore.
And I think you will switch it off. Can I? Okay, yeah, just closing, just closing. Just last slide. So we are the Climate Taste. We build the, they are building the global inventory for greenhouse gas emissions.
We were the contributors for the biomass burning part. All the data is open source and CC by four, so you can have a look at there. We are a fairly young company, and still we have won a bunch of awards. We merged our Dutch prime minister in April, I suppose. We're part of AWSA SpaceX later, and we are the Bloomberg new economy catalyst for this year and we are meeting Michael Bloomberg.
So that's a very fun thing that Bloomberg for environmental data, no relation there. Yeah, that's about it. Thank you. And have a good day. Thank you. Thank you.