Foraging on crops by wild ungulates may create human-wildlife conflicts through reducing crop production. Ungulates interact with and within complex socio-ecological systems, making the reduction of crop damage a challenging task. Aside from ungulate densities, crop damage is influenced by different drivers affecting ungulate foraging behavior: food availability and food quality in the landscape (i.e. the foodscape) as well as fear from hunting and scaring actions (i.e. the landscape of fear) may together affect the degree of damage via both direct and indirect effects. A better understanding of the individual effects of these potential drivers behind crop damage is needed, as is an appreciation of whether the effects are dependent on ungulate density. We investigated this by applying path analysis to test indirect and direct links between ungulate density, foodscape, landscape of fear and, human management goals on crop damage of oats and grass, respectively. Our results suggest that crop type is the major driver behind crop damage, with more damage to oats than to leys, implying that human decisions (i.e changing crop type) influence the level of crop damage. We found that management goals and actions influenced the foodscape and the landscape of fear, by affecting the amount of forage produced in the agricultural landscape and the amount of scaring actions. Additionally, we found that supplementary feeding influenced the local ungulate densities in the area. Our results highlight the importance of including human actions on multiple levels when assessing drivers behind damage by ungulates in managed landscapes. We suggest that more studies using path analysis on multiple scales are needed in order to tackle complex issues such as crop damage and other human-wildlife conflicts.

The ecological data varirables was collected as the following: crop damage: the biomass inside an exclosure and outside of the exclosure was cut using electric scissors, the difference in biomass betweent the exclosure and the plot outside was compared to generate a biomass loss (crop damage) variable. This was processed into an average % biomass loss of the 3 pais of grazed and un-grazed plots on the fields. 
Forage availability in the surrounding was measured along transects surrunding the field as an absent/presence inventory. This was processed into a propotion of forage species per transect (average of 4 transects). Biomass on fields was a measure of the biomass inside exclosures on fields in gram per field. Unuglate density was a measure of the number of pellet groups of all ungulate species measured along transects surrounidng the fields.This was processed into the number of pellet groups of all unuglate species from the entire area sampled. The social data was collected using surveys, giving information about if supplementray feeding, hunting, and scaring had been conducted on the fields. Furthermore, there was also one suervey question about what management goal the farm had, whether it was hunting/game keeping or agricutlure.

Excel is necessary in order to open the data file.