Vegetation and vertebrate abundance as drivers of bioturbation patterns along a climate gradient

have a comfortable nest site [51] on the other hand. Our data do not allow a test of this hypoth- esis, but further analyses of these allometric relationships are likely to provide a rich source of biological and behavioral information, particularly in studies comparing a large number of animal groups. Our first hypothesis, that bioturbating activity decreases from arid to humid regions [21– 26], was supported by our results for invertebrates, as their hole density decreased from arid to humid climates. Vertebrates, however, created fewer holes in arid than in semi-arid regions. Burrowing vertebrates are, on average, larger than invertebrates [51] such that fewer holes are consistent with a decline in animal density with increasing body size [54]. Accordingly, verte- brates were presumably less frequent in the arid region of our study than in the other climatic zones, such that fewer vertebrate than invertebrate burrows were present over a given area. Similarly, the higher hole density of invertebrates all along the climate gradient can be attrib- uted to the generally higher abundance of invertebrates [55]. However, it is also the case that most invertebrates create their own new burrows while some vertebrates use previously exist- ing burrows as an energy-saving strategy [56, 57]. In particular, larger animals, in our case ver- tebrates, invest more energy in burrowing effort than smaller invertebrates. Previous investigations showed that the energy cost of burrowing is directly proportional to the amount

of soil moved by the bioturbator. Consequently, larger vertebrates, which need to move larger soil amounts to create a burrow of adequate size, will burrow fewer holes [15]. The excavated soil volume did not follow a clear pattern across the climatic gradient and it differed between invertebrates and vertebrates in our study. Similar results were obtained in a recent study measuring the excavated soil volume of bioturbators along the same environmen- tal gradient [36]. The authors found that the excavated soil volume was greater in the semi- arid (0.56 m3 ha-1 yr-1) and Mediterranean (0.93 m3 ha-1 yr-1) than in the arid (0.34 m3 ha-1 yr- 1) and humid (0.09 m3 ha-1 yr-1) climate zones and that the excavation rates were higher for vertebrates (0.01–56 m3 ha-1 yr-1) than for invertebrates (0.01–37 m3 ha-1 yr-1). These findings are in line with several studies showing that, due to their larger body size, vertebrates excavate considerably larger volume of soil (1–5 m3 ha-1 yr-1) than invertebrates (<1 m3 ha-1 yr-1) [55, 58–64] as well as our findings. Those studies together with our own demonstrate the impor- tance of vertebrates as bioturbators along a climate gradient. Our second hypothesis, that bioturbation activity and quantity respond to seasonal changes [23, 29], was supported by the higher hole density during autumn than spring of the southern hemisphere, as observed at both the semi-arid and humid site. In the arid desert, with a consis- tent lack of rainfall events, there was no difference between seasons. This is in agreement with

previous studies and with the observation that in the southern hemisphere the bioturbation season ends in autumn [65]. Moreover, the climate in Central Chile during the study period in 2019 was drier than usual [66], which may have lessened the differences in bioturbation activ- ity and quantity between seasons. While the relationship between seasons and bioturbation patterns is no doubt, our study suggests that, at least in Chile, the impact of bioturbation is largest in semi-arid and humid climate zones after the autumn rainfall. The absence of a clear trend between vegetation cover and either bioturbation activity or quantity along the climate gradient was consistent with previous studies examining the distri- bution of burrow entrances as a function of vegetation [67, 68]. However, we were able to show that the bioturbation patterns of invertebrates and vertebrates differed. The consistently negatively association of invertebrates with vegetation cover supported our hypothesis that some invertebrates are entirely independent of surface resources due to their permanently belowground lifestyle [34]. By contrast, because vertebrates rely on a resource supply from the surface [33], a positive association with vegetation cover occurred only in the middle of the geographic gradient, as in the arid region the vegetation cover is sparse. Vertebrates living in PLOS ONE Vegetation and vertebrate abundance as drivers of bioturbation patterns along a climate gradient PLOS ONE | https://doi.org/10.1371/journal.pone.0264408 March 4, 2022 9 / 14