Supporting Soil Fertility From Space

Sentinel-2 pseudo-true colour composite from 2016 with a Kompsat-3 Normalized Difference Vegetation Index (NDVI) product from 2015 inset. Sentinel data courtesy of ESA/Copernicus.

Last Tuesday I was at the academic launch event for the Tru-Nject project at Cranfield University. Despite the event’s title, it was in fact an end of project meeting. Pixalytics has been involved in the project since July 2015, when we agreed to source and process high resolution satellite Earth Observation (EO) imagery for them.

The Tru-Nject project is funded via Innovate UK. It’s official title is ‘Tru-Nject: Proximal soil sensing based variable rate application of subsurface fertiliser injection in vegetable/ combinable crops’. The focus is on modelling soil fertility within fields, to enable fertiliser to be applied in varying amounts using point-source injection technology which reduces the nitrogen loss to the atmosphere when compared with spreading fertiliser on the soil surface.

To do this the project created soil fertility maps from a combination of EO products, physical sampling and proximal soil sensing – where approximately 15 000 georeferenced hyperspectral spectra are collected using an instrument connected to a tractor. These fertility maps are then interpreted by an agronomist, who decides on the relative application of fertiliser.

Initial results have shown that applying increased fertiliser to areas of low fertility improves overall yield when compared to applying an equal amount of fertiliser everywhere, or applying more fertiliser to high yield areas.

Pixalytics involvement in the work focussed on acquiring and processing, historical, and new, sub 5 metre optical satellite imagery for two fields, near Hull and York. We have primarily acquired data from the Kompsat satellites operated by the Korea Aerospace Research Institute (KARI), supplemented with WorldView data from DigitalGlobe. Once we’d acquired the imagery, we processed it to:

  • remove the effects of the atmosphere, termed atmospheric correction, and then
  • converted them to maps of vegetation greenness

The new imagery needed to coincide with a particular stage of crop growth, which meant the satellite data acquisition period was narrow. This led to a pleasant surprise for Dave George, Tru-Nject Project Manager, who said, “I never believed I’d get to tell a satellite what to do.’ To ensure that we collected data on specific days we did task the Kompsat satellites each year.

Whilst we were quite successful with the tasking the combination of this being the UK, and the fact that the fields were relatively small, meant that some of the images were partly affected by cloud. Where this occurred we gap-filled with Copernicus Sentinel-2 data, it has coarser spatial resolution (15m), but more regular acquisitions.

In addition, we also needed to undertake vicarious adjustment to ensure that we produced consistent products over time whilst the data came from different sensors with different specifications. As we cannot go to the satellite to measure its calibration, vicarious adjustment is a technique which uses ground measurements and algorithms to not only cross-calibrate the data, but also adjusts for errors in the atmospheric correction.

An example of the work is at the top, which shows a Sentinel-2 pseudo-true colour composite from 2016 with a Kompsat-3 Normalized Difference Vegetation Index (NDVI) product from 2015 inset. The greener the NDVI product the more green the vegetation is, although the two datasets were collected in different years so the planting within the field varies.

We’ve really enjoyed working with Stockbridge Technology Centre Ltd (STC), Manterra Ltd, and Cranfield University, who were the partners in the project. Up until last week all the work was done via telephone and email, and so it was great to finally meet them in-person, hear about the successful project and discuss ideas for the future.

AgriTech Seeds Start to Grow in Cornwall

On Monday I attended the Jump Start AgriTech event hosted by the South West Centre of Excellence in Satellite Applications at the Tremough Innovation Centre on the University of Exeter’s Penryn campus near Falmouth in Cornwall. As the name suggests the one day event covered innovations in AgriTech with a particular focus on what is, or could be, happening in the South West.

The day began with a series of short presentations and Paul Harris, Rothamsted Research, was up first on their Open Access Farm Platform. North Wyke Farm in Devon has been equipped with a variety of sensors and instruments to understand the effects of different farming practices. Of particular interest to me was their analysis of run-off, weather monitoring and soil moisture every 15 minutes; this is a great resource for satellite product validation.

I was up next talking about Earth Observation (EO) Satellite Data for AgriTech. Having seen people overpromise and oversell EO data too many times, I began with getting people to think about what they were trying to achieve, before looking at the technology. The circle of starting questions, on the right, is how I begin with potential clients. If satellite EO is the right technology from these answers, then you can start considering the combinations of both optical/microwave data and free-to-access and commercial data. I went on to show the different types of satellite imagery and what the difference in spatial resolution looks like within an agriculture setting.

I was followed by Vladimir Stolikovic, Satellite Applications Catapult, who focused on the Internet of Things and how it’s important to have sensor network data collected and communicated, with satellite broadband being used in conjunction with mobile phones and WiFi coverage.

Our last talk was by Dr Karen Anderson, University of Exeter, who looked at how drones can capture more than imagery. I was particularly intrigued by the ‘structure from motion photogrammetry’ technique which allows heights to be determined from multiple images; such that for a much lower cost, you can create something similar to what is acquired from a Lidar or laser scanning instrument. Also, by focusing on extracting height, data can be collected in conditions where there’s variable amounts of light, such as under clouds, and it doesn’t requirement high accuracy radiometric calibration.

After coffee, case studies were presented on farming applications:

  • VirtualVet – Collecting data on animal health and drug use digitally, via mobile apps, so paper records don’t become out of data and data can be collated to gain greater insights.
  • Steve Chapman, SC Nutrition Ltd, talked about improving milk production by making sure dried food is optimally prepared – large pieces of dried sweetcorn are digested less well, and a lower nutritional value is extracted from them.
  • The delightfully named, Farm Crap App from FoAM Kernow, aims to encourage farmers to spread manure rather than use artificial fertilizer. Farmers tended to go for the latter as it is easier to calculate the effects, and so having advice, regulations and the important calculations in a phone app, rather than in paper tables, should help them use manure.
  • Caterina Santachiara, ABACO, describing their siti4FARMER solution which is a cloud-computing based platform that includes data which scales from the field to farm and large land areas, with individual customisation so that users can easily see what they need to know.
  • Finally, Glyn Jones from AVANTI, talked about how farmers can stay connected to the internet, and tech support, while out in their fields. This sounds straightforward, but none of the current technologies work well enough – mainly due to the fact that fields aren’t flat! So a new technological area of investigation is ‘white space’ – these are frequencies allocated to broadcasting services, but left unused in particular geographical locations as buffers. The availability varies from location to location, but it is available to lower-powered devices.

After lunch, there were some presentations on Agritech funding opportunities from Innovate UK, AgriTech Cornwall and the South West Centre of Excellence in Satellite Applications. The day concluded with a facilitated session where small groups explored a variety of different ideas in more detail.

It was a really good day, and shows that there is real potential for AgriTech to grow in the South West.

Remote Sensing and the DIKW Pyramid

DIKW PyramidSatellite remote sensing industry is evolving and anyone working in it needs to become familiar with the Data, information, Knowledge, Wisdom (DIKW) pyramid as this is one map, albeit simplistic, of the industry’s and our current journey.

Historically, satellite data was either sold as the original image or with a small amount of processing undertaken. If anyone wanted to do anything beyond basic processing, they had to do it themselves. However, things are changing.

According to a recent Euroconsult report, at least 3,600 small satellites will be launched over the next decade. The United Nations Office on Outer Space Affairs only lists 7,370 objects that have ever been launched into space, of which only 4,197 are still in orbit. We’re increasing the number of objects orbiting the Earth by 85% by smallsats alone, larger satellites will add even more.

The volume, variety and speed of this data collected by these satellites will present a step change not only in the type of applications companies will be able to offer, but, crucially, also in customer expectations – more and more they will be looking for added value.

One way of considering this is through the DIKW pyramid, which can be seen at the top of the blog, it’s credited to American organisational theorist Russell Ackoff in 1989, building on the ideas of Milan Zeleny two years earlier.

A simple summary of the pyramid starts with the collection of data which means nothing in its own right, it is simply data. Information is derived from data by asking the who, what, where, when and how questions. Knowledge is information to which expert skills and experience have been added to create more value – which is more profitable in a business context. Finally, wisdom is understanding what actions to take based on the knowledge you’ve gained.

Applying this to satellite remote sensing for agriculture, one example might be: data is the satellite data/image of the field. Information is knowing when the image was taken leading to where in the growing cycle the crop was. Knowledge is applying scientific algorithms to know the soil moisture, how much nutrients are in the soil or how much vegetation is present in various parts of the field. Wisdom is knowing what nutrients and fertilizers to apply, based on the knowledge gained, to improve crop yields.

A lot of Earth observation products are at the data or information level, with a few at the knowledge level, and even fewer at the wisdom level. Customers more and more want wisdom products, and they aren’t that interested in what was required to create them. When you add to this the additional types of geospatial information, e.g., optical and radar used together alongside airborne and in-field ground based measurements, the variety of open datasets and the new science and technological breakthroughs, things are going to look very different, very quickly.

We’d accept that the DIKW isn’t a perfect tool, nor a perfect representation of our industry, but it is simple, indicative and worth thinking about. We wrote about our intention to create products in an earlier blog. We’re a long way from the wisdom sector, but are hoping to be firmly within the knowledge sector and collaborating to create wisdom. It’s not easy and some companies will find it harder to do than others, but is going to be the future. How are you preparing?

Remote Sensing and Agriculture in Italy

Poster from the Game Changing Technologies in Agriculture Event in Milan on 1st October 2015

Poster from the Game Changing Technologies in Agriculture Event in Milan on 1st October 2015

Last week I was in Italy talking all things remote sensing and agriculture. At the start of the week I was in Rome with the European Space Agency (ESA) discussing the Sentinel-2 performance, before catching the train north to Milan on the Wednesday evening for a series of UK Trade and Industry events focused on technology in the agricultural industry (AgriTech).

Thursday’s event was titled ‘Game Changing Technologies in Agriculture’, and was held in what looked like a large greenhouse in the grounds of Villa Necchi. We began the day with a welcome from the UK’s Ambassador to Italy, which was followed by talks from those working most directly in the agriculture industry. It was fascinating to hear some of the facts and see how much of a technological revolution has been occurring within this field. This is being driven by both the world population’s increased need for food – a 60% increase in demand by 2020 – and the corresponding need for businesses to increase their productivity. An overriding theme was the need to be more robust to, or better understand, the environment, including protection food production from both the weather and pests to reduce wastage.

After coffee we moved onto the provision of technological solutions, and there were a couple talks about how both drone and satellite remote sensing could benefit agriculture. My favourite other talks included the fitting of accelerometer collars on cows to collect data about their move movements more effectively, and the use of robot mechanical hands to perform repetitive tasks.

The afternoon expanded into synthetic biology, nanotechnology and technologies to reduce energy requirements or produce it more sustainably. Refrigeration is an important technology for the developing world, allowing a reduction in the current 40% post-harvest food wastage, but needs to be undertaken efficiently; the engines powering the refrigeration units on lorries produce much more pollution than the lorry engines themselves. This was followed by an interactive session where UK innovation centres had ‘stands’ that were used as discussions points on issues such as crops, horticulture, livestock, aquaculture, satellite technologies and big data. The day concluded with talks by Williams Advanced Engineering (associated with the Williams Formula One team) and IBM on how technologies are crossing from one sector to another.

On Friday, whilst it rained heavily in Milan, I spent the morning at the first Sainsbury’s Italian supplier conference. It was interesting to see how a large company is defining, and following, its strategies that include a focus on simplification; both for the supply chain and what the customer experiences. In the afternoon we had an escorted visit around the Milan Expo 2015. This is a six month exhibition which began in May and runs to the end of October and has the theme of Feeding the planet, energy for life’; it has exhibititors from over 140 countries and an exhibition area of 1.1 million square metres; although I didn’t explore all it! The UK exhibit was a beehive structure and wildflower meadow that was connected back to a real beehive in UK.

It was an interesting week and gave me lots of food for thought on how we can further develop the AgriTech services Pixalytics offers.