Making Boots on the Ground More Effective: The Potential of Unmanned Aerial Vehicles in Agricultural Development
It’s no secret field time is expensive.
Development projects that aim to improve agricultural production often have 30,000 or more farmers. It’s no surprise that when you truly target the poor, it’s often hard to reach them. You might have to arrive in an all-terrain vehicle or walk over streams. It might take an hour or more if the farmer’s field is inaccessible by motor vehicle. Maybe you have to take a donkey, as the terrain is too steep to walk easily. These are common scenarios when truly targeting the most marginalized.
A single field agent can have 50 or more farmers in such hard-to-reach places. This agent trains others on various new agriculture management practices, manages demonstration plots, delivers improved varieties of plants and conducts regular monitoring among other activities. It makes sense that, to save time and money, field agents often meet with farmers in one location and conduct project activities as a group. This can also improve the adoption rate of new practices, but that is difficult to monitor without walking each famer’s field, no matter how remote or hard to reach.
Sometimes even that isn’t enough. Take Bossou Antoinette’s cashew farm that she sharecrops in Benin. Monitoring projects usually means asking farmers if they have tried the new management practices. But this practice only measures the farmer’s perception — not the reality in the farmers’ fields or the challenges faced there. Maybe in a corner of Bossou Antoinette’s farm there are invasive weeds that keep coming back, and she gave up on that corner because nothing seems to work permanently, and the work is difficult. Perhaps she may not mention that to a project monitor. When asked, she could simply answer, “Weeding is one of my biggest problems. Cutting is a lot of work.”
That’s what she told me about the farm she sharecrops. We count this as a success because she tried weeding, so that box is checked. This doesn’t mean that such monitoring is bad, it just means it is subject to such human error, and we can’t always go to every corner of every farm to confirm these reports.
Technology can help. Many studies show that using images or pictures provides a more accurate measure of field conditions than even highly trained agriculture practitioners on the ground. Until recently, such high resolution imagery came from satellites and was out of the reach of many development programs because of cost and cloud cover. Now, however, the low cost of unmanned aerial vehicles (UAVs), which can fly below clouds, means that development programs can increasingly access high-resolution imagery.
So far UAVs have been mainly used in emergencies. But last week CRS, in collaboration with NetHope, flew a UAV over cashew farms in central Benin. The images told us immediately that there was a need to thin out trees in some places. We could see where there is space available to plant more trees and how many could be planted, where there had been burning, and identify areas for follow-up due to invasive weeds or other problems. This is just from a first look at the image. Further analysis might tell us even more. With this information, the field agent can use his or her limited and expensive time to pinpoint areas that require an in-person visit.
This imagery can even tell us where about a corner of Bassou Antoinette’s farm that has a weeding problem, one that we did not see even though we walked the boundary of her plot.