Satellite data has the potential to change an entire ecosystem »
Founder and CEO of start-up Blue Sky Analytics, Abhilasha Purwar discusses the importance of using satellite data and AI in the fight against climate change.
What is the core business of your company?
Blue Sky Analytics uses satellite data and artificial intelligence resources to develop environmental monitoring and climate risk assessment products. Ten years ago, the satellite resolutions were much lower, cloud computing infrastructure was limited, making cloud and satellite based environmental monitoring very difficult, and hence there was a much higher reliance on IoT sensors or manual monitoring. But the advancement in sensor technologies, resolutions, democratization of computing power through large cloud infrastructures have enabled young Indian engineers and data scientists at Blue Sky Analytics to access, process, and model such huge volumes of satellite data. When we founded our company in 2019, AI had evolved a lot for variety of other use cases in consumer markets, financial markets, but use of AI in Earth Observation for Environmental Monitoring was relatively nascent. Given Climate Change and Environmental degradation is the biggest challenge facing our planet, it seemed obvious to us at Blue Sky Analytics to dedicate our energies to that challenge. It then seemed obvious to us to use all this satellite data in the context of climate change by relying on the resources of AI.
How can earth observation be used as a tool in the fight against climate change?
Climate change is already a reality, as confirmed every day by extreme events such as droughts, floods and forest fires. Earth observation using satellite data not only allows us to observe these phenomena but also to provide mitigation and adaptation solutions.
For example, thanks to satellite data and artificial intelligence, we have managed to build a database on forest fires around the world, on the greenhouse gas emissions they generate and on various other parameters related to risk, forecasts, intensity and models.
And it’s no small thing: all of the Fortune 500 companies have set targets of zero emissions within 10 or 20 years. But a large forest fire can wipe out all of these attainments.
What are the current barriers to greater use of these AI resources for earth observation purposes?
Although progress has been made, climate change is still seen as a matter for non-governmental organisations, state agencies and activists. However, initiatives are also multiplying in the private sector, with, like us, start-ups that need funding to progress. Our business model is very effective: start-ups like ours require investments of $5-10 million, a drop in the bucket compared to the needs of other start-ups, and with that capital efficiency we can provide much higher degrees of environmental monitoring that can catalyze and enable entire climate action ecosystem.
How advanced is the technology?
Satellites, Big Data, and AI; I think you cover all the latest technical buzzwords of technology with Blue Sky Analytics. But if we fundamentally go back to basic engineering maths, we are using ground data and eyes in the sky to monitor what is happening with various resources on our planet like air quality, water quality, industrial emissions, and this data is critical to monitoring and evaluation of countless climate mitigation and adaptation projects around the planet. Our product is pretty developed, we have a common dashboard called SpaceTime for all clients and it can be expanded with 100x more datasets and an API gateway through which all our clients can access our datasets with just few lines of code and integrate on their excel sheets, apps, internal company dashboards, quarterly reports, etc. However, our goal really is to simply our “advanced technology” and make it very easy and accessible for users, and thus we would love active feedback of variety of users in the global climate space whether its insurance or banking industries or asset managers or policy makers, so we can improve both the quality and accuracy of our datasets and usability and accessibility of our platforms.