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

the no-fertilizer control. Thus, the application of NPK+FYM shows significantly more varia- tion in the response variables. This variation is due to a high performance in livestock produc- tion but low performance in most biodiversity related indicators. Estimation of the number of livestock that can be supporting by the different fertiliser treatments When comparing the stock feeding capacity for each of the treatments, the no-fertilizer control supported the lowest number of cattle and NPK+FYM the highest, with a difference of one animal per hectare of forage between the two treatments (Table 5 in S1 File). However, crude protein content in hay was only adequate for livestock under FYM and FYM+NPK. An impor- tant point is that the calculations focus on requirements during the first and second trimester of the cow’s pregnancy, and supplements may be needed closer to calving given the modest metabolizable energy contents of the hay. While fertilizer addition was needed to meet crude protein needed to feed suckler cows dur- ing the winter, the hay from the unfertilized control plots appeared to provide adequate nutri- tion for mixed ewes. Fertilizer treatment influenced the number of sheep that could be potentially fed per hectare of forage, thus the no-fertilizer control vs FYM has a difference of 17 sheep for the control vs 24 sheep for FYM and 26 sheep for NPK+FYM. Finally, NPK-only 23 sheep. However, in all cases, feeding supplements will still be needed to meet increased nutritional demands at key physiological stages (e.g., pregnancy). Discussion

In this study, we compared the performance and described the trade-offs among livestock pro- duction, plant communities and insect communities, under four different fertilizer treatments in an upland grass forage system managed under agri-environmental schemes. We observed strong trade-offs in all the fertilizer treatments, as all treatments performed well for some indi- cators and less well for others. Overall, FYM was the fertilizer treatment closest to the optimum point between biodiversity and forage production having the lowest trade-offs between indica- tors. Finally, we found that after 27 years of no nutrient addition, forage production is pre- dicted to be sufficient to maintain low sheep densities but not sufficient for upland cattle. In what follows, we first discuss the limitations of the work and then discuss the results with fur- ther literature. Limitations There are three main limitations to our work. First, our results are conservative as in the exper- iment the maximum fertilizer application rate was constrained as a condition of receiving sup- port payments from the English & Welsh agri-environmental scheme. Farms not within the scheme can apply fertilizer rates two to four times higher, particularly in permanent grassland [50,53]. However, our results shows that even at low rates of fertilizer addition there are impor- tant trade-offs between biodiversity and production, the implication being that trade-offs will worsen at even higher levels of input. Second, we did not calculate livestock production using

observations of live animals as our experimental plots are too small to support livestock. Rather we scaled-up the plot-level measurements and linked these to animal production values from the agricultural literature. However, this limitation can also be an advantage as comparisons can readily be made between the different types of livestock used in farming. The third limita- tion of our work is that the proximity between plots could lead to spill over of species among treatments which might not arise with larger distances between plots. In this sense, our results are conservative as spill over among plots will tend to mask effects. 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 12 / 18