Selection of foraging microsites by granivorous rodents in the Monte desert: a field experiment

LOPEZ DE CASENAVE, J.; MILESI, F.A.; CUETO, V.R.; MARONE, L.

Mendoza

Congreso; 10th International Mammalogical Congress; 2009

Most desert granivorous rodents forage preferentially under shrub and trees. Though usually associated with perceived risk of predation, this preference is particularly likely if seed abundance peaks in those microsites, as in the open algarrobo woodlands of the central Monte desert, Argentina. To evaluate if rodents use of space is associated with the main structural features of the vegetation, we  recorded, once per season, the nocturnal removal of a single Setaria italica seed from each of 300 stations arranged every 5 m in three 10×10 grids. Nocturnal consumers were confirmed as small mammals in two smaller, similar field experiments, by identifying their footprints over sieved fine soil around seed stations. We characterised every microsite (1m-diameter circle around each station) by measuring percent cover of each plant species with a point-interception technique, and then used PCA to summarise the correlation patterns of the multivariate matrix. We also measured distance to the nearest-neighbour tree. We assessed microsite selection graphically and statistically. As expected, seed removal was higher in autumn-winter than in spring-summer. In only one of the three grids there was some evidence of positive spatial autocorrelation of neighbouring stations. Although woody cover and seed abundance were strongly associated with the main PCA axes, there was no avoided microsite type, and the structural characteristics of used sites did not differ from availability. The use of particular microsites among seasons did not differ from random. In conclusion, in a field experiment in which seed addition was almost nil, rodents did not show a particular use of space at the microsite scale. In contrast, they did show a tendency at a bigger spatial scale, foraging farther from trees than expected by chance. We contrast these results with those found simultaneously for granivorous birds at the same place.