Eyes and ears in the sky
How a revolution in satellites drives unique insights about our planet
Think of a satellite. Imagine what it looks like, its size and how much it costs.
Most likely you were thinking of an automobile-sized, multi-ton object that costs hundreds of millions of dollars to develop and another hundred million dollars to launch into orbit on top of a giant rocket. And you are right, this is what satellites were like. Until now.
The satellite industry has fundamentally changed with space becoming more accessible. Satellites have become smaller, more powerful and much cheaper to launch into orbit.
Due to lower equipment and launch cost it is now possible to build a constellation with dozens of satellites. Satellites can be frequently replaced allowing a progressive stream of innovation.
Three distinct drivers are responsible for this trend.
Firstly, Moore’s law has radically driven down the cost and size of electronics. Similar to how mobile phones have shrunk and become more powerful, satellite electronics that used to take up large amounts of space in traditional satellites now fits into a shoebox. Launching something smaller and lighter into space is inherently less costly. Figure 1 shows a modern cubesat – 10x10cm standardized satellite unit typically sitting in low earth orbit and capable of taking high-definition images.
Secondly, private companies such as SpaceX have entered the launch vehicle market and have radically driving down cost of launch services by aiming for re-usability of rockets. Traditionally, launch vehicles are lost after a successful launch. Imagine if every time a UPS truck delivers a parcel to your house it blows up afterwards. The cost of shipment would increase significantly. Through reusability and a radical approach to cost SpaceX has managed to offer a launch for $61.2 million dollar on the Falcon 9 today compared to an average cost of $225 on a ULA rocket, a traditional launch provider[1]. With SpaceX perfecting the reusability of their rockets costs are bound to come down even further.
Thirdly, because traditionally space projects have been incredible uncertainty and high-risk only governments or multi-billion dollar communications conglomerates had the ability to invest in them. With smaller satellites being deployed into low-earth orbit the cost of a loss is significantly reduced. Also, due to the lower manufacturing cost satellites can be launched rather quickly and business models become revenue generating within months.
Google’s acquisition of the satellite company Skybox Imaging for $500m has sparked great interest in the venture capital community. Skybox Imaging’s constellation of low-earth orbit satellites will allow Google to take a picture of any place on earth twice a day with possible applications in Google Maps. Leading venture capital funds such as Bessemer Ventures or Canaan Partners are actively looking for investment opportunities in the satellite market.[2]
Access to space essentially allows us to capture unique data that drives insights in all kinds of industries. For example, if you could count the number of cars on a Wall-Mart parking lot as a hedge fund you could predict earnings more accurately. Or imagine you have hourly pictures of forests around the world that allow you to identify fires early on.
While satellites only collect data, the key to success is processing and analyzing this data to generate actionable insights.
One startup successfully doing this is satellite company Spire. The four-year old startup was founded by Peter Platzer (HBS’03) and as of today has launched 12 satellites into orbit. By the end of 2016 it plans to have 40 in orbit in total.
Spire’s satellite constellation ‘listens’ to the planet. Based on the collected data Spire has developed two innovative products.
Spire Sense picks up Automatic Identification System (AIS) signals sent by ships to identify themselves and their position. After only 50 nautical miles from the shore the curvature of earth blocks the AIS signal from reaching a port. Spire’s satellites, however, can pick up these signals anywhere and thereby track ships globally. One interesting application of this technology is combining the AIS signals with terrestrial imaging which allows you to identify piracy or illegal fishing activities.
Spire Stratos utilizes GPS radio occultation to collect weather and climate data. As shown in figure 2 satellites pick up a signal from a GPS signal which passes through the atmosphere. The changes introduced to the signal by the atmosphere allow Spire to infer on weather and climate conditions in particular locations.
Recently Spire has signed a landmark contract with the National Oceanic and Atmospheric Administration (NOAA) to provide weather and atmospheric data, the first time NOAA purchased weather data from a private company. Spire’s data will feed into weather and climate models, helping us predict natural disasters und better understand climate change.
Spire estimates that weather impacts $26 trillion of the world economy and that 10% of that can be mitigated by improved weather models.[3]
Today many space startups generate a large amount of imagery and data. The big challenge, however, is to turn all this data into actionable recommendations for businesses back on earth. Spire is leading the way with highly innovative models to provide weather data and track ships. Other space companies should follow their lead and focus heavily on data analytics.
[1] http://breakingdefense.com/2014/05/ula-fires-back-at-spacex-at-space-symposium-details-launch-costs/
[2] https://techcrunch.com/2016/03/08/google-renames-its-satellite-startup-skybox-imaging-to-terra-bella-and-adds-focus-on-image-analysis/
[3] http://www.satellitetoday.com/nextspace/2016/10/26/spire-ceo-launching-satellites-every-month/
Stefan, great post! The ability to track AIS signals beyond the reach of shore receivers could be a powerful tool for commercial shipping firms and maritime insurers. Inexpensive nanosatellite constellations could also replace coastal radar systems that are beyond the budget of many developing country security forces.
Great post. Combining cheap satellites that gather massive amounts of data, with machine-learning tools that can analyze this data and provide insights has the potential to change many industries. To add to the industries already mentioned in the post, agriculture can be dramatically changed using this method. Analysis of crop field photographs taken by satellites is actually already taking place by Descartes Labs (http://www.descarteslabs.com/). The company claims to forecast global commodity crop production more accurately than the best government or commercial sources. If indeed the forecasts are better, the implications are huge: from estimating future commodity prices, supply & demand, to changing irrigation methods and improving yields using this data.
Thanks Stefan! The first thing thought that came to my mind when reading this was: is there enough space for everyone?
As this, and many other developments move forward, I’m curious as to how orbits are allocated. Is it a first come, first served system? Or do countries/companies have designated slots in space? I know space is huge, but the “best” slots for satellite orbit are likely few in comparison. Wikipedia does not shed a lot of light into the topic… https://en.wikipedia.org/wiki/Space_law#Geostationary_orbit_allocation
And second, what happens to the satellites after their useful life is done? You mention about SpaceX’s rocket recovery technology. Not only does it save cost, but also reduces the debris left in space. I looked a bit more into this topic, and was SHOCKED of the amount of debris in space. Nasa is actually tracking half a million pieces of space junk! And what’s worse, “They all travel at speeds up to 17,500 mph, fast enough for a relatively small piece of orbital debris to damage a satellite or a spacecraft.”
https://www.nasa.gov/mission_pages/station/news/orbital_debris.html
I’d be curious to hear how satellite technology firms are thinking about these threats to their business going forward.
I couldn’t edit my post, so I’ll leave it here – an interesting development in terms of recovering space junk. There appears to be a lot of work on the topic, but this shows just how complex and expensive it is to recover space junk!
http://www.space.com/23049-space-junk-satellite-swiss-space-plane.html
Very interesting post Stefan!
Applications here certainly seem endless, as Aviad started to dig into in his response. Another fascinating company I ran across was SilviaTerra, which uses satellite imagery to assess forests for potential lumber extraction (using color and density to determine type, quantity, and concentration of trees in minutes, vs. a traditional labor heavy process which could take months).
It would be very interesting to plot the cost curve and see what other applications may make economic sense as the price of satellite generated data decreases over time. If the cost curve drops fast and far enough, I wonder whether we may see a return of satellite based communication (a la Iridium) at a larger scale, particularly in more rural areas devoid of cell tower/ broadband infrastructure.
Stefan, this is fascinating, and, based on our offline conversations, I know this is something which is one of the next big things in the coming years!
I had a few points and would love to hear your thoughts on the same:
1) How do you feel about the improvement in accessibility of “SpaceTech” to young entrepreneurs. My thought behind the point is that, as compared to IT related products/technologies (which are immensely accessible to the greater community), space technologies are very niche and hence presents a barrier to entry for even well-educated individuals. Do you see something like, “Make your own satellite” happen in the near future?
2) I would also like to understand more mass scale uses of satellite technologies – outside communication and imaging. Would love to hear your thoughts on the same.