Increasing productivity without compromising environmental sustainability is, perhaps, one of the greatest puzzles of agricultural development. Yet recent research reports provide evidence of stagnating crop yields and regionally variable crop yield patterns globally. On a regional level, households within the same agro-ecological area have different productivity levels which often vary even between fields managed by the same household.
Yield differences have both biophysical and socio-economic causes: soil types, rainfall patterns, farming techniques and access to labour, fertilisers and advisory services all influence farm productivity. This means that in some cases smallholder farmers are unable to benefit from the potential yield improvements offered by breeding. Thus, better understanding and modelling of the links between the different causes of productivity variations can improve management and increase food security for many households in low-income countries, but to have better models you need better data.
Based on the resources of the Sustainable Agricultural Production and Food Security Theme, a group of researchers from the Swedish University of Agricultural Sciences (SLU), Lund University, Swedish Institute of Agricultural and Environmental Engineering (JTI) and from Makerere University have been testing Unmanned Aerial Systems (UAS) technique for gathering data about farms in the Mbale sub-region in Uganda. The method is combined with interviews and field observations.
UAS is a remote sensing framework tool which consists of an aircraft, ground control stations and data links. The used aircraft (or drone) does not have a human pilot. It is controlled either by the on-board computer or remotely from the ground. UAS, initially used by the military sector in conflict and humanitarian relief zones, have proven to gather data sufficient to meet the needs of agricultural research.
Despite the fact that the team encountered some technical problems, namely difficulties with aircraft landing within the investigated fields, the researchers reported positive results and say this approach is very promising. Initial flights made during the workshop provided data from which they were able to differentiate between plants in a mixed cropping setting and even to identify individual maize plants.
“Applying this methodology requires more data than could be collected during the workshop and the equipment needs some refinements, but the technology needed for this is already available. The flyovers and interviews were positively received in all the villages we visited. We believe that this remote sensing framework is feasible and can soon become operational. The next step is to perform a study over larger areas and over at least two seasons”, Sigrun Dahlin, research group member.