Context dependent effect of an herbivorous subterranean rodent on vegetation in relation to primary productivity
Description
Subterranean rodents could modify the environment becoming keystone species and/or ecosystem engineers. This role may be context dependent if the changes produced shift throughout the species range and are stronger under certain environmental conditions. Our objective was to analyze if the effect of Ctenomys mendocinus on vegetation cover is context dependent, specifically examining variations between areas with different levels of primary productivity. We compared the effect of C. mendocinus on plant cover among four contrasting environments and found that it depended on primary productivity in a predictable manner. In low productivity environments, the rodent significantly reduces vegetation cover, while it has no discernible effect in highly productive ones. These findings suggest that the effect of C. mendocinus on plant cover may be dependent on primary productivity level and highlights the importance of accounting for the underlying environmental factors that influence the intensity of C. mendocinus-plant interaction.
Files
Steps to reproduce
Study area: This study was developed along a precipitation gradient of the arid central west of Argentina, in four environments, two in the phytogeographic province of Puna and two in the Monte. Study sites were situated in protected areas. In Puna the survey was carried out in Don Carmelo Multiple Use Reserve (-30.94744, -69.08427) and in the highland area of Villavicencio Natural Reserve (-32.485623, -69.114907). In the Monte the sites were in the lowland of Villavicencio Natural Reserve (-32.644943, -68.873889), and in the Man and Biosphere Reserve of Ñacuñán (-34.044444, -67.909753). Field survey: Primary productivity differences amongst sites were assessed with green index comparisons. For each study site, SAVI index (soil adjusted vegetation index) was calculated. For this purpose, Landsat 8 images were used from February to March 2016 with processing level 2. In every environment, 150 random points were selected, separated by at least 300 m. Finally, a 3x3 mean filter was used. C. mendocinus is recognized for its solitary behaviour, established by the capture of only one animal by gallery system, even during the reproductive season. These gallery systems can be identified by distinctive disturbance patches characterized by a dense concentration of burrow entrances and horseshoe-shaped mounds of soil resulting from herbivory and excavation activities. In this study, we analyzed the effect of perturbance on vegetation cover at a patch scale. We categorized patches disturbed by C. mendocinus as areas highly disturbed = “disturbed” (areas with a high density of holes, 1.73/m2 or more), and areas relatively undisturbed = “undisturbed” (patches with a low density of holes [0.18/m2 or less] and mounds [0.13/m2 or less]) for comparative purposes. Then, to compare the effect of C. mendocinus on vegetation cover, 10 patches of individual recent C. mendocinus activity were selected on each study area, separated at least 100 m from each other. We set a 30 m long transect in each disturbed patch and repeat it in an adjacent undisturbed patch. Within each transect, vegetation cover was measured in 15 subsamples of 2x1 m each. Only aboveground vegetation cover was measured, because it is the main component of the diet of C. mendocinus. We also estimated the relative effect of C. mendocinus on vegetation cover at each site, enabling us to compare the magnitude of the effect across the environments. The relative effect was calculated as the ratio between vegetation cover in undisturbed (U) patches over disturbed (D) ones (Relative effect = U/D). A ratio close to one implies that there is practically no effect of C. mendocinus on vegetation, as cover is similar in disturbed and undisturbed patches. In contrast, a ratio larger than one indicates an effect of the rodent on vegetation cover, as a result of lower cover in disturbed patches.