Four pieces of advice for leveraging Earth observation data and artificial intelligence to create competitive edge

Four pieces of advice for leveraging Earth observation data and artificial intelligence to create competitive edge

I recently returned to a favorite fishing spot for the first time in a couple of years. Sitting at the tip of a peninsula, I’d caught a lot of fish there in the past. Imagine my surprise when I found it gone – not just the fishing spot, but the entire peninsula.

My experience is a microcosm of how the physical characteristics of Earth are changing at an unprecedented pace and scale. Climate change, deforestation, urbanization, infrastructure build out, energy transition, and a myriad of other factors are transforming our world faster than we expect.

Sometimes this change comes in something as small and personal as the loss of a favorite fishing spot. Increasingly, however, change plays out at large scales, such as extreme weather or shifting agricultural productivity, with cascading economic and social impacts.

For business leaders, the challenge is how to understand this accelerating geophysical change, keep pace with it, and predict its impact to make better decisions.

The key is to gain perspective. As we discuss in the EY Space Tech Lab report, space provides a strategic vantage point, offering both big picture views and pinpoint insights into change on Earth. Satellite-based Earth observation and sensing has reached a tipping point, enabled by falling launch costs, growing sensing capabilities, and a maturing set of enabling technologies, such as cloud and artificial intelligence (AI).

Ten years ago, most satellite data was only understood by the researchers and scientists who produced and studied it. Today, the large cloud providers host petabytes of satellite data streamed to Earth every day and make it accessible through applications available to anyone. For example, back home from my fishing trip, it took me only a few minutes to call up a time series of satellite imagery of my fishing spot, which revealed that a series of storms had washed the peninsula away.

AI augments human capabilities to draw deeper insights from this vast satellite data. Machine learning (ML) can be trained to identify a specific feature in an image or detect change in a series of images over time. This opens a universe of use cases such as detecting infrastructure at risk, flagging incipient forest fires, and tracking oilfield methane emissions.

Generative AI further augments our decision-making abilities by allowing us to interrogate giant data sets in our native language. Our EY alliance partner, Microsoft, applied generative AI to satellite data in an interesting way in “Queryable Earth”.

Soon, business leaders will be able to generate insights which integrate satellite data with other data sets using the right prompts.

This democratization of the massive stream of space data – much of it open source – gives individuals new powers to generate their own insights and innovations. EY teams are tapping into this augmented human capability with a series of open data sciences challenges which give participants an opportunity to use AI for good, working with satellite datasets to help address sustainability problems. Our latest challenge focuses on developing AI models for coastal resilience in data-poor environments and developing practical disaster response plans. ?

As a business leader, how should you start to leverage Earth observation and sensing data? This is the advice I frequently give clients:

  1. Understand the capabilities of satellite data. We can image the ground at high resolution (30-cm scale) daily with tasked commercial satellites. Medium-resolution ground images (10 to 30-meter scale) can be accessed every 2 or 3 days with free data from NASA and ESA. The combination of this optical and radar data can tell us about the existence of objects or land classes, their change over time, and their state. So, be thinking about the combinations of spatial detail (what you see) and temporal frequency (how often you look at it).
  2. Think "out of the box". The most common applications of satellite data focus on asset monitoring and land change. But, with growing technological advances, satellites are beginning to do so much more. For example, emerging use cases include: carbon accounting associated with nature-based assets, risk modeling associated with climate change, disaster response (fires, floods, storms), microclimate modeling, 3D surface mapping, small object detection, land deformation using radar, water quality, and vegetation disease detection using hyperspectral imaging. Imagine how capabilities like these could create use cases and value for your company.
  3. Consider the build vs. buy decision. While satellite data is becoming increasingly accessible, distilling business insight from Earth imagery still requires AI and data science skills. For business leaders, the question of whether to build in-house capabilities or work with business partners will depend on several factors: - Anticipated breadth of use of Earth observation and sensing data- Degree to which the data and capabilities will be strategic to the business- Opportunity for spill-over benefits into the organization’s innovation and technology strategy
  4. Find the right business partner. A growing cohort of companies provide value-added services by accessing and processing the satellite data into client products and integrating client data into those solutions. Businesses should choose a company that seeks to understand their needs, brings experience and subject matter expertise to provide the most effective and efficient solution, and utilizes the latest advances in computing and AI/ML to yield the best outcomes.

Every business leader should evaluate the opportunity and business case for utilizing satellite data in their operations. Not all businesses will conclude they should invest today. But many will. The longer some wait to commit, the more they are at risk of a competitor using it to create a new advantage in productivity or sustainability. The pace of change is only increasing – don’t get left behind.

#EarthObservation #AI #CompetitiveEdge #DataDriveninsights


Established in 2020, the EY Space Tech Lab is an innovation hub that includes business leaders, remote sensing professionals, data scientists, and AI professionals. Built on our deep sector knowledge, our traditional strengths in consulting, assurance and tax, and ongoing investment in data and technology, EY teams are helping clients who want to observe the Earth virtually through geospatial imaging, analytics, and trustworthy artificial intelligence.

The views expressed in this article are the views of the author, not Ernst & Young. This article provides general information, does not constitute advice and should not be relied on as such. Professional advice should be sought prior to any action being taken in reliance on any of the information. Liability limited by a scheme approved under Professional Standards Legislation.

Dan Klein

Independent Real Estate Consultant

1 年

Great thinking! Pay attention people, before it's too late.

回复
Jason Errey

Is your ground modelling work flow integrated into your BIM? ask me how.

1 年

Interesting post, and I think for the most part accurate. I do note this part: "Soon, business leaders will be able to generate insights which integrate satellite data with other data sets using the right prompts. This democratization of the massive stream of space data – much of it open source – gives individuals new powers to generate their own insights and innovations." To some extent this is already the case. For example, a project comes across my desk and the first thing I do is open Google Earth (Yes, I have high end GIS, but Google Earth is so much quicker in the first instance). But insights to support decision making (where, what and how), that is something different. First we must establish the rules and logic behind data provenance and data surety. How can the user be sure of the data they are ingesting? How does high data surety vs low data surety impact the probability of success? We see early examples of poor data surety already having negative impacts on decision making in divers fields such as the news media on one hand, and Carbon Credits on the other. It is critical we "experts" address these questions before "non-experts" gain the access to these tools. And any logic or rules must be transparent.

回复
Jennifer Oldfield

Senior Director Of External Relations, Global Partnership for Sustainable Development Data

1 年

Great post Brian!

回复
John Metzger

CaaS / Earth Monitoring (EM) and Geomatics / New Business Program Development

1 年

#EO - Earth Observation domain = #Science, #Military, #Commodities_markets, #Academics, #Itinerant_Business - snapshots of time affirm, discern, inform, or associate Earth surface status or condition at the time of the acquisition, only. Financial transactions supporting these - #emergency_related, #project_or_term_budgets, #misc_below_tender_resources, #IDIQ_type, #VC_funds_chits -- by nature extremely #volatile and potentially irreverent. Federally sponsored public money is as fleeting, as it is promised. It can evaporate overnight. A very difficult business regimen -- as noted in the current, and recent reports and balance sheets of "leaders", zombie corps, and startups. #EM - Earth Monitoring offers a very different engagement - long term recurring income - derived from ongoing intra-monthly/monthly data flows (acquisition, processing, assessment, dissemination). It does however, require engagement with a resource of data with an extensive archive, ongoing global collect, and an open facility for location specific, no fixed minimum results. These budgets are safety, evidentiary, or business production and operations related -- the data serving as a regular decision-making and asset assurance tool. #ItsaJungleoutthere

回复
James Cutler

NED and strategic commercial and digital advisor

1 年

Summary right for those prepared to make the effort and are aware that there is an outcome that makes that investment effort worthwhile. The flip side for the EO industry is that it really shouldn’t require the erstwhile customer to become the EO user (as point 2 implies). Rather that nascent market, generally seeking value through insights to inform decision choices, (not point solutions or projects) is ill served by the EO applications layer that often sustains itself via projects. The reasons why are well rehearsed ;-). #locationintelligence

要查看或添加评论,请登录

Brian Killough的更多文章

  • The Space Tech Opportunity

    The Space Tech Opportunity

    The Space Tech Opportunity The potential of Space Tech to address real-world business challenges on Earth is a…

    3 条评论

社区洞察

其他会员也浏览了