How do you respond to markets if you can’t see what’s on the farm?
That was the problem Women in Business Development Inc (W.I.B.D.I) in Samoa was trying to solve.
That problem led the non-governmental organisation to investigate different pathways. Should they look at satellite maps and make deductions from there? What were their other options? Drones? Some type of agriculture mobile app? What were the costs and benefits?
Was there the technical expertise in Samoa?
The answer came in a meeting with Skyeye in Samoa. Skyeye, owned by Nomeneta Saili and Chris Saili, specialises in vehicle tracking. Their service promised to reduce fuel costs through better monitoring and staff driving to a preset speed limit.
But they also had the capacity and desire to do much more. W.I.B.D.I executive director, Adimaimalaga Tafuna’i says she asked the combined team to “think forward” and to create a system that would hold the organisation in good stead for the next decade.
Through using drones, Skyeye has now mapped 80 per cent of W.I.B.D.I farms (15,000 hectares) by drone and counted 800,000 coconut trees.
As part of the process, Skyeye trained W.I.B.D.I extension staff to use the tablets and database officers to manage the data. Recognising the need for a server, another Saili family member, Sam, facilitated this donation from SysCentral Ltd in Melbourne.
Christopher Saili says the W.I.B.D.I project really allowed Skyeye to learn a lot about a whole range of things and they were very grateful for the opportunity. “To be able to work with W.I.B.D.I and demonstrate Samoan ingenuity and sustainable development is something we are very proud of.”
Tafuna’i says Skyeye’s commitment has been crucial.
“They understood our funding limitations as an N.G.O. We couldn’t afford to pay for everything up front and Skyeye have gone ahead with the work, even when we shared with them the fact that we were still negotiating possible funding avenues. This only happens when a business operates with values.”
W.I.B.D.I were also able to redirect funds with permission from its main funders, NZAid and Oxfam N.Z., from its fuel budget lines due to savings to the drone and database project.
An aspect of the project that W.I.B.D.I felt was essential was the contracting of a local company that understood the limitations of the mobile infrastructure, the capacity of its officers, the need to tweak products to fit requirements and most of all, robust testing in the field.
These were lessons from W.I.B.D.I’s first foray in trying to create mobile apps using an offshore developer.
The problem they experienced that time was that although the tests worked remotely, locally the app was not functioning. Another lesson was the first problem that our archaic and unwieldy database needed to be upgraded.
W.I.B.D.I had all its farmer database information in Excel. And, though there was lots of data, it was not disaggregated and difficult to analyse.
Previously, the organization had secured funding from the Pacific Media Assistance Programme (P.A.C.M.A.S) Innovation Fund that was funded by AusAid. This funding supported the first mobile phone and app project and really started the organisation on its technology and communication journey.
A second set of funding came from C.T.A – The Technical Centre of Agricultural and Rural Co-operation to develop a series of mobile apps that unfortunately stalled in the trial phase. These apps were essentially a standalone product that used data that had to be manually updated.
This time W.I.B.D.I was looking for an integrated solution for data collection and dissemination. They also had a large market asking how many coconut trees did W.I.B.D.I have on their 588 certified organic farms.
The fixes were largely systemic. The database is currently being overhauled to accommodate electronic forms that extension officers fill out on tablets. The forms are uploaded via mobile phone networks or the office wifi.
The farm gates all have G.P.S points so they are now easier to find, which are handy considering that very few roads in Samoa have street names. The next layer of information was the Geographic Information System (G.I.S) maps of individual farms and the counting of individual trees.
Skyeye’s G.I.S technician Ephraim Reynolds says since satellite imagery does not offer high enough resolution imagery needed to digitize key features on these farms, and manned imagery is too expensive, drone technology was the perfect answer.
“We use a fixed-wing professional mapping U.A.V that allows us to cover large areas — areas so large that we are sometimes able to cover multiple farms in a single flight. “Our U.A.V is an intelligent drone that has handled Samoa's difficult terrain and unpredictable weather very well. The drone allows us to capture images of farms that are not easily accessible, and gives us the flexibility to fly when we want (weather permitting).”
Reynolds says the biggest challenge encountered has been locating suitable landing areas, as the drone requires an open area free of vegetation to safely land after completing a mission.
“Google’s satellite imagery in Samoa is a few years old, and we found that areas that looked suitable on these outdated images were now covered with banana trees, pineapple plantations and shrubs.
Sometimes, we’ve found that the best solution is to ask the locals in the village where we can find suitably clear land.”
Maintaining a strong radio link to the drone was also problematic due to tall coconut trees, which can obstruct the signal and result in the drone not capturing images. “For this, we shortened the range of the drone’s flight path, or found higher ground to launch the drone from,” says Reynolds.
“After we download the images from the drone, we process them into orthomosaics: stitched-together images that have been digitally corrected for distortion, so that they can be overlaid onto a map. We open these image layers in a free, open-source G.I.S computer program known as Q.G.I.S. In Q.G.I.S, we are then able to digitize key features located on the farms. With this high resolution drone imagery, we are also able to clearly see individual coconut trees, allowing us to conduct a visual count of total tree numbers.”
Skyey uses a G.I.S tool known as a Web Feature Service (WFS), which allows us to grant users access to our geoserver - an open source server made for sharing geospatial data.
With W.F.S, users are able to download individual layers of information, such as the layer containing information about a farm’s coconut trees. They can then make their own changes and updates to the digital map. In this way, they are able to divide the labour and make the process of analyzing the drone imagery faster and more centralized.
To speed up the process of mapping, Skyeye shows farmers images of their farms from the air so the farmers can draw their boundaries. After that, there is still a ground-truthing exercise to go through to estimate the ages of the coconut trees so W.I.B.D.I can forecast the yield and production of virgin coconut oil, one its main exports.
Reynolds says as Samoa and the Pacific continue to realize how drone technology can be used in various industries, especially in agriculture, the region will become better able to reach large markets and keep up with modern advancements.