Earth Observation Plays a Vital Role in Global Supply Chain Monitoring
Disruptions in global supply chains caused by COVID-19 restrictions, labor challenges, raw material shortages, and the global political situation have raised awareness of the need for reliable and accurate information about the movement of goods from manufacturer to end-user. The satellite industry plays a critical role in monitoring assets, offering near real-time data for decision-makers about the global supply chain situation, helping to select optimal routes, anticipate congestion, and more.
Over the past two years, we have learned how fragile the global supply system is—and how little we know about it. After becoming aware of the inherent risks, many organizations started to invest in technology to mitigate them. However, it emerged that reliable monitoring of the transfer of goods from manufacturers, through ports, ships, terminals, routes, trucks, and warehouses, was much more challenging than many anticipated. For instance, while tracking the shipments on land can be done using mobile networks, it is not as easy at sea, where approximately 80% of global trade per volume (and over 70% by value) is transported.
Earth Observation data has proved to be an effective source of information, especially with regard to maritime transportation. A combination of AIS data with optical and SAR data is used to identify ships, track vessel movement, and monitor the conditions of maritime routes, as well as detecting signs of illegal activity, dark vessels, and maritime pollution. Individual technologies have distinct limitations but, when combined, can provide a far better view of what is happening at sea. All of the required data can be found via online platforms such as UP42.
The value of satellite data for mitigating maritime supply chain risk goes far beyond tracking shipments. It aids in better understanding marine traffic in areas of interest, and also in obtaining information on changes in port activity patterns — such as the change in the number of vessels entering the port, or the volume of commodities stored or shipped. Moreover, satellite-derived bathymetry data helps to monitor coastal regions for water-depth-related risks, which can be a major challenge for massive container ships. But perhaps the biggest and best example of the value of Earth Observation data in this arena was the blockage of the Suez Canal…
A deeper look into the Suez Canal case
On March 23, 2021, the Panama-flagged container ship Ever Given became wedged perpendicularly across the Suez Canal, blocking the path of dozens of vessels until its liberation on March 29. This unique crisis—unprecedented in the canal’s 150 years of history—severely interrupted a supply chain that was already under pressure due to the COVID-19 pandemic.
12% of total global trade moves through the canal that separates Africa from Asia. Ever Given was holding up an estimated $9.6bn of goods daily. In addition, while hundreds of ships stayed in their place in the ‘queue’, waiting for the blockage to be resolved, other shipping companies decided to take an alternate, far longer route around the Cape of Good Hope, requiring additional time and fuel, and consequently costs. The array of goods that were caught up in the jam due to the blockage of the canal varied from oil tankers to liquified gas and biodiesel, live animals, crops, cement, automobiles, furniture, and tea, among others.
An image from Planet, captured on March 23, clearly shows the powerful sandstorm that was said to be a major cause of the container ship’s deviation from its course, leading to the blocking of the canal. On March 24, an image from BlackSky shows other vessels waiting in the canal (before the canal had been completely evacuated of cargo ships). On March 25, a SAR image from Capella Space shows the ship and the tugboats working on it. On March 26, ICEYE imagery provides a wide-area view of the obstruction and the traffic at the entrance of the canal. On March 27 and March 28, images from Maxar show the progress of the dredging operations underway to release the ship. And on March 29, an image from European Space Imaging shows Ever Given finally freed, with tugboats surrounding it.
This crisis proved the value of Earth Observation data to all the organisations along the supply chain, as satellite images became a major source of information to update decision-makers and the public regularly. As a result, satellite data companies and marketplaces received a massive amount of tasking requests to provide fresh imagery as often as possible. Earth observation data were crucial to understanding the severity of the problem and making short- and long-term decisions by organizations along the supply chain. In the array of unproven data, satellite imagery provided physical evidence of activities to restore the channel’s capacity, as well as their progress.
For many organizations, access to such data turned to be of a high importance. There were 64,887 containers stuck on Ever Given and about 360 other ships waiting at the entrance to the Suez Channel. This translated to thousands of trucks or rail carridges to transport it, hundreds of warehouses that had to freeze their capacity and factories stopping the production process because of lack of raw materials. Understanding the situation from an independent reliable source influenced decisions worth millions of dollars. It was a situation where satellite imagery not only became part of our daily news, but also played a major role for organizations that might not have used satellite data in their decision-making process before.
The commercial satellite industry continues to grow steadily, with more than 3,371 satellites circling the earth. This progress is demonstrated by the diversity of services that remote sensing companies are offering across all sectors, including global supply chain monitoring. The blockage of the Suez Canal by Ever Given has been a significant example, proving the value of Earth Observation for decision-making, risk mitigation, and public awareness. It has been a major trigger in the adoption of Earth Observation for maritime supply chain monitoring. This well-documented event triggered researchers and technology companies to build new solutions that could preemptively avoid such problems. We can certainly expect to see many startups moving into this space in the near future.
The new 30cm Pléiades Neo constellation offers, the precision agriculture sector, data that bridges the gap between drones and open-source satellite imagery
Precision agriculture has been one of the big buzzwords in the world of agri-tech. It is also one of the most important applications for imagery data from both satellites and drones, which are a critical component of the precision agriculture process. Multispectral imagery data can be combined with other data (such as crop type, terrain models, soil type etc.), to calculate so-called ‘prescription maps’. Farmers use these maps to analyse the situation on the ground to make informed land management decisions. To trigger specific actions such as variable-rate spraying, the maps can be uploaded to automated farming equipment.
Typically, prescription maps are based on satellite data, which can be complemented with drone or airborne data for individual fields where cloud coverage makes satellite data unusable—particularly relevant at the peak of the growing season, when the time window to make the right decision may be as short as 1–2 weeks.
The new Pléiades Neo constellation from Airbus is filling the gap between satellite data from open-source constellations such as Sentinel 2 and drone data.
Temporal and spatial resolution
Pléiades Neo offers 30cm native resolution and after launching the complete constellation later this year, it will be able to revisit any location around the world several times a day.
Large crops, (cereals, rapeseed, corn etc.), are mostly monitored by satellites such as Sentinel, which offers a much lower resolution compared to Pléiades Neo, but a larger swath. This covers more fields over a given location in a single pass.
Pléiades Neo, despite offering a smaller swath compared to Sentinel, with four satellites in orbit, Pléiades Neo satellites can cover large areas within a single day and at much higher resolution. The level of detail obtained may be a little excessive for current precision agriculture equipment, but what it does do is provide very valuable information when it comes to monitoring crop growth to prevent potential diseases – 30cm resolution imagery enables you to see a much clearer picture, meaning early detection is possible.
This high-resolution data also plays an important role for specific use cases, such as high-value crops like vineyards and olive orchards for example. Across Europe, such plantations are now facing increasing challenges of drought, so farmers need to use water resources in a far more efficient way. 30cm resolution is important in specifically identifying the condition of each separate tree in order to make an informed decision about how to treat it.
Another use case is in crop trials performed by technical institutes in the context of research and development activities. Such institutes have objectives to continuously improve systems of crop management, to develop advanced decision-making criteria and to benchmark new cultivation methods. Every year, they perform numerous tests in frameworks of micro-plots planted in open fields.
One more instance where higher resolution is useful is in locations where fields are narrower than 10m, which there are many of, particularly in central Europe. When the field is 5-7m wide (and, as an example, 2km long) the error rate for a 10m pixel size is significant.
In precision agriculture, spatial resolution is very often less important than spectral resolution. Spectral bands allow us to see and measure things not visible to the human eye. The most popular measure coming out of various spectral bands is ‘normalised difference vegetation index’ (NDVI), which is based on red, plus the infrared bands – easy to measure for the majority of satellite or drone multispectral sensors. But in practice, NDVI is just scratching the surface of the possibilities that come with adding additional spectral bands, such as the new Pléiades Neo ‘Red Edge’ band, or visible, near-infrared and shortwave infrared in Sentinel 2, among others.
NDVI will not work well with developed crops. To create a full model of the canopy reflectance, you need to use different indices, such as Leaf Area Index, ‘normalised difference red-edge index’ (NDRI), or ‘modified soil-adjusted vegetation index’ (MSAVI), depending on crop type, growth stage and the purpose of the analysis. Using the Red-Edge band from Pléiades Neo satellites, allows better chlorophyll content measurement and as a therefore a more precise estimation of the Leaf Area index values on crops, at plain development. Indeed, the Red-Edge band helps to pinpoint subtle stress in crops in advance compared to traditional indices, as it corresponds to the region of electromagnetic spectrum where the reflectance of green vegetation changes rapidly.
The role of spectral bands goes beyond vegetation indices. Spectral bands are of no interest to farmers of course, but they do care about the quality of agronomic information derived from that data. Satellite sensors need to pass through the atmosphere and then adjusted for the atmospheric distortions. Therefore having a set of bands helps to characterise vegetation as well as enabling the best and most accurate data correction. This is why the ‘Deep Blue’ band is very important in helping Pléiades Neo obtain the most accurate data: it significantly aids the atmospheric corrections.
Satellite vs. drone data
In fact, the accuracy and reliability of the data related to radiometric corrections is the biggest challenge for drone sensors. Satellite data typically uses a large number of bands, with a very stable sensor and radiometry procedure for regular calibration, providing trusted results. By contrast, the calibration of a multispectral drone sensor is not an easy task.
To obtain reliable drone data, the sensor calibration has to be in line with the specific weather and light conditions (e.g., cloud coverage and time of day). If calibration is made for clear skies, a single cloud passing by will influence the result. There are techniques to manage this, but it adds a certain level of uncertainty and has a significant impact on the prescription map produced from of this drone data.
Pléiades Neo offers a unique combination of both spectral bands and high spatial resolution, smartly completing lower resolution satellite constellations, such as Sentinel, ensuring precision and timeliness of the information. Moreover, it can substitute drones or aerial data, especially over larger areas, as it is much more economically viable. With its very high spatial resolution, excellent spectral parameters and more frequent revisits increasing the chances of cloud-free data, it is an excellent asset.