When Satellite Data Improves - What Happens in Clean Technology?

 In June this year, we had a lively discussion and online workshop on remote sensing data and how monitoring processes occurring on the Earth was why the Landsat satellite program was launched in the 1970s - a program that’s still running today. 


But here’s an interesting question that came up in our conversation - since water, agriculture, energy and other clean tech sectors have been using remote sensing data for such a long time - what is so different now? 


To answer that question, let’s first talk about how satellite data is used in clean technology. The sectors where satellite data, and data science in general, are widely used both commercially and in research and development are agriculture, energy, water, climate and disaster management. So, what are the different uses of satellite data in each of these sectors?


Let’s take agriculture first. Researchers and scientists have been using satellite data since the 1970s in the agricultural sector. The first product from Landsat data was the agriculture and crop data set - where satellite data was used to identify agricultural areas, the type of crop that was grown in the fields and the yield of the crops. Initially, this dataset was used by agricultural agencies in states and governments to better estimate the contribution of the agricultural sector to the economy, to identify farms and areas that were distressed and needed assistance, and to determine if the type of crop was suitable for the soil and environmental conditions found in that area or if incentives were needed to shift to a better use of the land. 


Over the years, the spatial and temporal resolution of the satellite data has improved dramatically - to the point where we now have satellite data at the centimeter scale as opposed to data that’s available in meters or kilometers. The availability of high quality data, that is also free in many cases, has meant that several additional uses were found commercially. Several companies in the agricultural sector now routinely use satellite data to optimize farm operations, predict crop yields for a wide range of crops early in the season - which in turn leads to better estimation of crop prices when harvest occurs, and identify areas at the farm and field scale where there are issues related to water or pests so that farm management can be improved. All these lead to better management of agricultural land and thus a significant increase in economic productivity.


Next, let’s take a look at the water sector. Satellite data has been used for a long time to monitor the state of water bodies - both the extent of pollution and the size of the water body in question. This has meant that the availability of water resources for different uses, including agriculture, can be monitored and modeled more accurately. Further, areas where pollution or illegal activities like over-fishing are occurring can also be monitored more frequently and solutions devised. The improvement in satellite data both spatially and temporally has meant that scientists and organizations are better able to estimate water availability, identify possible shortfalls resulting from decreases in surface water bodies, explore groundwater over-reach, pollution near the ocean coasts, lakes and streams - and in general - better manage water resources. 


Commercially, this sector is just beginning to take off with satellite imagery being used in a few different ways by different players in the sector. For example, we’re seeing water treatment plants use satellite data to identify water conditions prior to treatment. We’re also looking at organizations that are monitoring ocean pollution and the state of global fisheries, and companies that are using satellite data to track the movement of ships and thus better manage logistics and transportation issues. Of course, in Covid times, we’re seeing startups and organizations that are monitoring wastewater for the virus and using satellite imagery in combination with pooled testing of wastewater samples to identify areas where there may be a flare-up - which in turn helps target preventive measures that can minimize the spread of the disease.


Now, how about the energy sector? Probably the largest use of satellite data in the energy sector has been in monitoring the state of the grid and identifying issues related to the grid. An additional use case has been in using satellite imagery to evaluate sites for siting wind farms, solar plants and other renewable energy sources. And of course, to identify places in countries where electricity is intermittently available and improve energy access to those areas. 


As satellite data has become more accessible and higher quality, the models and monitoring associated with these use cases have improved significantly - leading to better estimation of the grid and greater ease in establishing wind, solar and other renewable energy power plants. Another use case where there’s significant commercial interest is in using satellite weather data to build models that better manage the energy used by buildings and hence minimize greenhouse gas emissions. 


Finally, let’s take a look at the climate and disaster management sector. Satellite datasets have been invaluable here - both in modeling climate change impacts and in improving the response to disasters that result from a changing climate. Weather data, wildlife monitoring, changes in season lengths, precipitation and drought conditions - all these have been essential in building climate models and in exploring how much variability is being seen as the climate changes. 


But probably the biggest impact that we can see in our lives is in how disasters are managed. As satellite data quality and availability have improved, countries and organizations around the world have been using this data to figure out the optimum use of resources in a natural disaster - what’s required, how efficiently it needs to be transported, which areas are most impacted and how to alert people so that they can move to safety faster.


So, in the last 5-10 years, we’ve been seeing an explosion in the amount and quality of satellite data - and a corresponding increase in how it’s being used in different clean tech sectors. Of course, one of the challenges as the amount of data increases is how to access the data efficiently and build useful models - machine learning, physical or statistical models. And that’s where data engineering comes in…..


If you’re curious about how to access satellite imagery using Python and what data engineering is, join us on Sunday, August 9th for a live, virtual, hands-on workshop!

What our community are reading

Moonshots, Models, IoT and Machine Learning in Agriculture

Our online community space is now live!

How much water should an email consume? Data centers and water use