Fertilization using manure minimizes the trade-offs between biodiversity and forage production in agri-environment scheme grasslands

Insect community Insect surveys were performed at the same time as the flower abundance survey. Leaf miners were collected from the same transects used for the flower abundance. Leaf miners were reared individually until an adult leaf miner, or a parasitoid emerged. Pollinators were collected using a hand net for 8 min in each plot between 09:00 and 17:00 hrs and sampling was repeated three weeks later for a total of three rounds (for a total of c.576 min). The level of sampling effort was deemed adequate since the species counts recorded 74% of the pollinator species, 97% of the herbivores and 85% of the parasitoid species estimated to occur in the Colt Park experiment (see 33). Here, we used information for the 24 plots or 33% of plots where no seed were added (see S1 File section 1). The dataset includes information records for 62 species of pollinators (n = 352) including 4 species of bumblebees (n = 15), 25 species of herbivores (n = 1565) and 22 species of parasitoids (n = 515). Using this information we calculated the Shannon-Weiner diversity index for each of the three insect groups. Finally, we calculated per- cent parasitism of the herbivores. Forage Mean forage productivity was measured from hay cut at peak biomass over a four year period (2011–2014). The samples were collected from the center of the plot (1 m from the plot edge), dried in an oven at 60˚C for 48 hours and weighed. For the nutrient content of forage, we collected samples in 2017. We used three transects of 5m separated by 50 cm in each plot, and a sample of c. 20 g of plant material per meter, was

clipped to ground level along each transect. The samples were combined to provide a total of 300g of hay per plot. Each sample was then oven dried for 48 hrs. at 65˚C and analyzed for ash, crude protein, neutral cellullase, gammanase digestibility and ruminant metabolizable energy by Sciantec Analytical services (http://www.sciantec.uk.com/services.php?service=forage). Statistical methods Comparing the performance of four different fertilizer regimes with respect to indica- tors relating to plant community, insect community, forage production and forage nutri- ent content. To compare the performance of each fertilizer treatments for each indicator and group of indicators (group average), we compared normalized mean values of the indicators from each group of indicators (plants, insects and forage). To this end, we first we standardized all indicators by subtracting the mean of each variable and dividing it by its standard deviation, this providing values in the same scale for all indicators while retaining the distributional fea- tures of each indicator measured. Finally, we estimated the performance of each fertilizer across all indicators by estimating the weighted mean of the 12 indicators (i.e. a group mean). We calculated a weighted mean to avoid any bias due to unequal number of indicators repre- senting forage and biodiversity. The mean values for each indicator were weighted considering that all indicators related to forage should contribute to 50% of the performance of each fertil-

izer while indicators related to biodiversity should contribute to the other 50% (with indicators relating to plant communities contributing 25% and indicators of insect communities contrib- uting the other 25%). Finally, to compare performance for each group of indicators individu- ally, we also calculated the arithmetic mean for each group of indicators (forage, plant community and insect community). The effect of fertilization regimes on the trade-offs between the plant community, insect community and livestock forage. Because of the relatively low number of replicates (n = 6 per treatment) we used randomization tests for differences among treatments for each of the 12 indicators. We also used the same approach to test for differences among the groups PLOS ONE Manure minimizes trade-offs between biodiversity and forage production in agri-environment scheme PLOS ONE | https://doi.org/10.1371/journal.pone.0290843 October 4, 2023 5 / 18