Climate change alters slug abundance but not herbivory in a temperate grassland

The effect was particularly evident in winter 2013 when slugs were more abundant in +heat plots. If similar effects occurred in winter 2012 (unfortunately, we did not trap slugs in 2012), this may have resulted in the only significant slug herbivory being observed occurring in the +heat treatment in the spring of 2013. Heat effects observed in slug abundance were also observed in forage quality metrics. Greater slug abundances in non-heated plots at the end of the summer coincides with the largest difference in C:N between heated and non-heated plots. During the late summer and fall, heated plots primarily consisted of C4 grasses, contrasting with C3 dominant grasses in ambient plots, explaining the large difference in C:N of the bulk forage for those plots. In addition, research has shown that invertebrate herbivores prefer C3 grasses to C4 grasses [49, 50]. It is possible that the greater abundance of slugs in non-heated Fig 5. Ordination of plant species cover in the slug exclosures (triangles) and control rings (circles) from each climate treatment plot. The plant species noted at each end of the axes indicate the positive and negative drivers of the ordinations and had an R-value � 0.65. https://doi.org/10.1371/journal.pone.0283128.g005 Table 4. ANOVA table for C:N and %N of bulk forage samples from the 2013 harvests. Degrees of freedom are shown as numerator, denominator. Bolding indicates significance. C:N %N Effect DF F P DF F P Heat 1,44 13.88 0.0006 1,12 16.30 0.0016 Precip 1,44 0.30 0.5837 1,12 0.00 0.9506

Heat × Precip 1,44 7.16 0.0104 1,12 7.55 0.0177 Time 2,44 36.20 <0.0001 2,32 39.33 <0.0001 Time × Heat 2,44 9.92 0.0003 2,32 11.42 0.0002 Time × Precip 2,44 0.47 0.6277 2,32 0.28 0.7636 Time × Heat × Precip 2,44 0.15 0.8584 2,32 0.09 0.9153 https://doi.org/10.1371/journal.pone.0283128.t004 PLOS ONE Slug herbivory & climate change PLOS ONE | https://doi.org/10.1371/journal.pone.0283128 March 14, 2023 12 / 17