Spatial variation in smallholder farms in the Sahelian region makes it difficult to estimate yields using models. One of the reasons for the spatial variation is the Messor Galla ant. This harvester ant species clears an area around their mound of vegetation. Seeds form the main part of their diet. As a part of a larger experiment on 50 farmer fields, a study was conducted to quantify the amount of arable land lost as a cause of Messor galla ants and its relation to soil and crop type. Using these result and other data collected during the experimental season an attempt was made at quantifying parameters that may correct the Beer-Lambert’s light interception formula so that more accurate yields can be estimated for variable smallholder farms. High resolution mosaics based on UAV imagery were used to identify potential Messor galla colonies. An in situ inspection of these potential sites followed afterwards so that the influence of crop- and soil type on the distribution of Messor galla colonies could be tested. The results show that Messor galla ants prefer to make their colonies in deep, well drained, sandy to sandy-loamy soils. Although cotton fields had the highest ant colony density, the farmers use crop rotation so therefore it is believed crop type does not play a role since a nest persists for several years. The arable land lost did not surpass 0,60% of the total field areas in >90% of the cases. Therefore it is suggested that the costs related to chemical control of Messor galla would outweigh the costs of damage they cause. The attempt to calculate the correction parameters was partially successful. Due to noise on the collected data and the minor impact Messor galla ants have, it was not possible to find any significant relations between parameters. Nonetheless the line of thoughts used to analyse the data can be used in future research.
The intership was part of the STARS project. The imagery that was used is also available, please contact Tom Schut.