No abstract

Subterranean animals act as ecosystem engineers, for example, through soil perturbation and herbivory, shaping their environments worldwide. As the occurrence of animals is often linked to above-ground features such as plant species composition or landscape textures, satellite-based remote sensing approaches can be used to predict the distribution of subterranean species. Here, we combine insitu collected vegetation composition data with remotely sensed data to improve the prediction of a subterranean species across a large spatial scale. We compared three machine learning-based modeling strategies, including field and satellitebased remote sensing data to different extents, in order to predict the distribution of the subterranean giant root-rat GRR, Tachyoryctes macrocephalus, an endangered rodent species endemic to the Bale Mountains in southeast Ethiopia. We included no, some and extensive fieldwork data in the modeling to test how these data improved prediction quality. We found prediction quality to be particularly dependent on the spatial coverage of the training data. Species distributions were best predicted by using texture metrics and eyeball-selected data points of landscape marks created by the GRR. Vegetation composition as a predictor showed the lowest contribution to model performance and lacked spatial accuracy. Our results suggest that the time-consuming collection of vegetation data in the field is not necessarily required for the prediction of subterranean species that leave traceable above-ground landscape marks like the GRR. Instead, remotely sensed and spatially eyeball-selected presence data of subterranean species could profoundly enhance predictions. The usage of remote sensing-derived texture metrics has great potential for improving the distribution modeling of subterranean species, especially in arid ecosystems.

The endangered giant root-rat (Tachyoryctes macrocephalus, also known as giant mole rat) is a fossorial rodent endemic to the afro-alpine grasslands of the Bale Mountains in Ethiopia. The species is an important ecosystem engineer with the majority of the global population found within 1000 km 2 . Here, we present the first complete mitochondrial genome of the giant root-rat and the genus Tachyoryctes, recovered using shotgun sequencing and iterative mapping. A phylogenetic analysis including 15 other representatives of the family Spalacidae placed Tachyoryctes as sister genus to Rhizomys with high support. This position is in accordance with a recent study revealing the topology of the Spalacidae family. The full mitochondrial genome of the giant root-rat presents an important resource for further population genetic studies.

Human activities, directly and indirectly, impact ecological engineering activities of subterranean rodents. As engineering activities of burrowing rodents are affected by, and reciprocally affect vegetation cover via feeding, burrowing and mound building, human influence such as settlements and livestock grazing, could have cascading effects on biodiversity and ecosystem processes such as bioturbation. However, there is limited understanding of the relationship between human activities and burrowing rodents. The aim of this study was therefore to understand how human activities influence the ecological engineering activity of the giant root-rat (Tachyoryctes macrocephalus), a subterranean rodent species endemic to the Afroalpine ecosystem of the Bale Mountains of Ethiopia. We collected data on human impact, burrowing activity and vegetation during February and March of 2021. Using path analysis, we tested (1) direct effects of human settlement on the patterns of livestock grazing intensity, (2) direct and indirect impacts of humans and livestock grazing intensity on the root-rat burrow density and (3) whether human settlement and livestock grazing influence the effects of giant root-rat burrow density on vegetation and vice versa. We found lower levels of livestock grazing intensity further from human settlement than in its proximity. We also found a significantly increased giant root-rat burrow density with increasing livestock grazing intensity. Seasonal settlement and livestock grazing intensity had an indirect negative and positive effect on giant root-rat burrow density, respectively, both via vegetation cover. Analysing the reciprocal effects of giant rootrat on vegetation, we found a significantly decreased vegetation cover with increasing density of giant root-rat burrows, and indirectly with increasing livestock grazing intensity via giant root-rat burrow density. Our results demonstrate that giant root-rats play a synanthropic engineering role that affects vegetation structure and ecosystem processes.

<p>Ecosystem engineers continuously shape and re-shape the spatial and temporal structure of the environment. Burrowing animals are an important group of ecosystem engineers, because of their ability to rework sediments and soils with consequences for e.g. soil formation and vegetation patterns. Simultaneous, burrowing animals depend on climate, local soil characteristics and vegetation. The endemic Giant Molerat (GMR) is a burrowing animal and important ecosystem engineer in the Bale Mountains. As part of the Bale Mountain Exile Hypothesis Project, the aim of this study is to investigate (1) the interlinkages between GMR, climate and vegetation patterns as well as (2) to upscale the influence of GMR on the vegetation pattern across the plateau with Sentinel satellite data. Field data comprise 47 paired plots of 5m x 5m with and without GMR activity. Additionally, 1.500 independent GMR burrow openings have been mapped. For investigating interlinkages, all parameters are first pre-analysed for correlations and their dependencies (1). In the following these results, the remote sensing data and the individual variables are implemented into the prediction model. To increase the accuracy, an error correction of the model is pursued. For this, the area is calculated into likelihoods of areas influenced by GMR, based on the vegetation survey pairs serving as training areas for the correction. The corrected results are used as final input model in a machine learning-based classification approach using Random Forest with forward-feature selection and leave-feature-out option (2). In the following the results of this ongoing upscaling approach used for the Sanetti Plateau, Ethiopia is presented.</p>