Skip to main content

Putting satellite data into the hands of farmers

By Amy Edwards

September 28, 2017

satellite image of river and irrigated area
The STARS Landscaping Study provides an overview of how remote sensing can support agricultural development and poverty alleviation. Pictured is the Cubango River on the boarder of Namibia and Angola with diverse forms of agricultural development. Image: Plant, Creative Commons 4.0.

Some of the poorest farmers in Sub-Saharan Africa are looking up at the stars with renewed hope thanks to advances in remote sensing technology.

And they are not alone …

Whether CSIRO is talking satellites with farmers on the ground in outback NSW or offering advice for using drones in remote parts of Africa, it is a new era in agriculture.

Technology advances in agriculture have included the STARS Landscaping Study, which outlines how remote sensing can help alleviate poverty by bringing the benefits of this technology to smallholder farmers in Africa and South Asia.

Meanwhile in Australia and the United States, farmers are embracing technology such as IrriSAT – remote sensing irrigation scheduling service that informs farmers how much water their crop has used and how much irrigation they need to apply. The service uses the Google Earth Engine to process imagery acquired from Sentinel-2 and Landsat satellites.

CSIRO Environmental Informatics team leader Matt Stenson believes the use of remote sensing in agriculture will significantly increase, especially when it comes to supporting ‘real-time decision making’.

“We are getting to a point where we will be able to monitor crops on a daily basis,” Stenson says.

“Previously, tools were too hard to use, internet connections were not good enough, satellite images were expensive and required huge amounts of processing. Now technology like IrriSAT can be used on a daily basis by farmers.”

Remote sensing data for African farmers

Supported by the Bill & Melinda Gates Foundation, the STARS Landscaping Study brought together some of the world’s leading minds in agricultural remote sensing including CSIRO’S soil, agriculture and food expert Dr Neil McKenzie and Dr Juan Guerschman from the Earth Observation and Data Synthesis team.

people in Africa in a field
Mali women farmers pointing at an aerial vehicle collecting imagery to calibrate relationships between satellite vegetation indices and plant growth variables sensitive to fertilization levels over a smallholder cotton field. Image: PCS Traore, ICRISAT.

“Sub-Saharan Africa and South Asia are the two regions where the poorest of the poor are located,” Guerschman says.

“The main type of agriculture in these areas is subsistence farming, including crop and livestock production. We identified ways to help farmers move out of subsistence agriculture into some form of ‘family farming’ that would generate a significant and marketable surplus leading to a long-term improvement in household well-being.”

The STARS Landscaping Study conducted literature reviews, workshops and individual consultations to identify a range of feasible remote sensing opportunities to help farmers. The opportunities included establishing online remote-sensing data services for Africa, improving rainfall data for famine early warning, improving soil and land assessment as well as providing timely information on agricultural production for national decision-making. In most, cases the opportunities identified had already been successful in developed countries.

The Sub-Saharan Africa and South Asia regions are more likely to use remote sensing technology and data from wealthier nations such as the United States or European countries than be able to establish their own systems.

“It is not expected nor desired that these countries invest in ‘space technology’,” Guerschman says.

“It makes a lot of sense for poorer countries to utilise already existing data streams and take maximum advantage of established operational systems because these have a higher likelihood of being maintained in the long run.”

The overarching and ongoing STARS research project is already using some remote sensing technology, such as satellites and drones (and the information they collect) to help farmers make better farming decisions. The project has led to farmers in Mali, West Africa, learning first-hand how satellite technology can be used to interpret crop performance and alter decision making.

drone over a field
Multirotor-based remote sensing system over farmers fields in Tanzania. Image: Roberto Quiroz, International Potato Center

Cotton planter Usman Sania Berthe from Sukumba, Mali was dismayed to see his cotton crop’s poor response to Sabunyuman fertilizer through drone and satellite images captured during the STARS project.

“… during later group discussions, I remembered that I was late in applying the recommended dose at the recommended date. That a satellite way out in the sky could lead me to this realisation is flabbergasting. For sure, next season I will do my best to be prepared, if fertilizer is available in due time …,” he told the STARS website.

While the STARS Landscaping Study and report were funded by the Bill and Melinda Gates Foundation (a non-government organisation), the outputs are freely available to any other organisation working in this space.

From satellite to paddock in real time

CSIRO senior software engineer and IrriSAT project lead Jamie Vleeshouwer believes low cost or free and easily accessible remote sensing technology is what has made programs like IrriSAT in Australia so successful.

IrriSAT is mostly targeted at cotton farmers in NSW and Queensland with the goal of improving water conservation. The service uses the Google Earth Engine to process imagery acquired from Sentinel 2 and Landsat satellites to monitor a paddock’s crop growth. This is paired with regional weather information; and local paddock information to track the daily soil moisture deficit in near-real time. The system is delivered to the users through a custom made interface developed using the Google App Engine. The system currently has more than 1300 active users in Australia and the US.

“Up until recently, satellite images required massive amounts of processing, were low resolution and could cost about $600 per individual image,” Vleeshouwer says.

“Today satellites like the European Sentinel 2 and NASA’s Landsat are providing data free of charge. We can tap into real time data and display images from the last year, week or even yesterday.”

satellite image of farm land
IrriSAT displays a Sentinel 2 image of Cubbie Station cotton farm in Queensland.

Once you log into IrriSAT, you mark out your own fields. Users can then access further details about crop growth and how much water the crop is using.

“It was well understood that farmers regard sitting at a computer as a waste of time and feel like they should be in the field. However this technology enables them to be out in the paddock using technology that is in their hands,” Vleeshouwer says.

The IrriSAT technology also provides data feeds to other agricultural service providers around Australia. These service providers build their own value added services around the IrriSAT information which they resell to their customers. One of the many examples of this is the ‘Go SAT’ product provided by Goanna Telemetry Systems.

IrriSAT was funded by the Cotton Research and Development Corporation and developed in partnership with the CSIRO, Deakin University and the NSW Department of Primary Industries.

According to Vleeshouwer and colleague Matt Stenson the next step in remote sensing technology in agriculture is to integrate all the data systems being used.

“Remote sensing data feeds can be accessed worldwide and we need some cohesion,” Stenson says.

If farmers world-wide were using the same data and applying similar agricultural methods, it would go a long way in closing the gap between cotton farmers in outback NSW and sub-Saharan Africa.

Leave a Reply

We love hearing from you, but we have a few guidelines. Your email address will not be published. Required fields are marked with *


We love hearing from you, but we have a few guidelines.

*
*