Metadata 1. General Project title: Does mixed farming benefit moths? Exploring how different farming systems shape both local features and the wider landscape. Project description: Lepidoptera are important pollinators and provide an essential food source for higher taxa, yet many species that were once widespread are in decline across Europe and in the UK. This is largely due to practices associated with intensive farming, such as pesticide and fertiliser applications and habitat loss. If we are to reverse biodiversity decline whilst feeding an ever-growing human population, we must find ways of farming that are beneficial for insects. ‘Mixed’ farming, where livestock are integrated into the crop rotation, is thought to provide benefits to biodiversity by reducing synthetic inputs and by increasing habitat and crop diversity. However, at commercial stocking levels, livestock can have detrimental impacts on grassland Lepidoptera. We investigate the different pathways through which mixed farming could benefit moths in comparison to arable farming (where livestock are absent). Between June and August 2022, twenty-six farms in Scotland were surveyed for moths using ultra-violet LED light traps. Each mixed farm was paired with a nearby arable farm and each pair was sampled once on the same night, with traps running from dusk until dawn. Structural equation modelling was used to explore whether mixed farming could benefit moths either directly or indirectly (mediated by changes in the landscape associated with mixed farming). Contact details: rochelle.kennedy89@gmail.com, nick.littlewood@sruc.ac.uk Geographic location: North East Scotland, between latitudes of 56.0 and 57.5 Date of data collection: 06/2022 – 08/2022 2. Data and file overview File description: 'Environmental_data.csv': This dataset contains environmental information associated with the location of each moth trap such as temperature, landscape composition, date etc. 'Moth_species_data.csv': This dataset contains information on the moth species and numbers of individuals found at each location and can be used in conjunction with the environmental dataset by using the 'TRAP' column as a reference column. Date of creation: 10/2022 Author/owner: Rochelle Kennedy (project lead) Formats: Data is given in .csv format as this is considered to be stable and can be opened in various applications that are free to use. 3. Data quality assurance Data cleaning: All data from traps where field type was ‘GRASS_GRASS’ were removed for statistical analysis as there were not enough traps of this type. In addition, all data from trap ‘DRMF’ was removed as this trap was potentially disturbed by farming activities (harvest) on the night of deployment. Quality control: Accurate identification of moths to species level was ensured by checking initial identifications with an expert, then using these as a ‘reference’ for further identifications. Micro moths can be difficult to identify to species level, as such, those that were not able to be confidently identified by the creator of this dataset were given to an expert for identification. The final dataset was checked for duplicates and missing data using R. The final species list was also checked by an expert for inaccurate identification (by checking for species unlikely to be found in Scotland). Thorough training and regular supervision was provided to fieldwork assistants who aided in data collection and moth identification. Software: All landscape data was manipulated in QGIS to obtain buffers of different scales and was then analysed using the landscapemetrics package in R to obtain proportion cover or diversity indexes. R version 4.3.2 was used for all statistical analysis and data manipulation. 4. Ethical notes Sensitive data: location data (latitude and longitude) has been removed to protect the location of the farms that participated in this project. Farms were anonymised by being given codes that would not allow for identification of the farm or farmers involved. Ethical approval: Granted by the University of Stirling. 5. Data documentation Missing data: all missing data is defined as a blank cell. Variable description: File: 'Moth_species_data.csv' ‘SPECIES’: species abbreviation. Where moths were identified to species level, abbreviation for genus and species are separated by ‘_’, otherwise an abbreviation is given for genus or family only. ‘TRAP’: a reference that uniquely identifies all moths that were found in a singular moth trap. ‘FARM’: anonymised code indicating which farm that the moths were recorded on. Multiple traps were placed on each farm. ‘OCCUR’: Abbreviation of ‘occurrence’: The number of individuals of a species that occurred per trap. ‘LEVEL’: The level of identification that was achieved: either ‘SPECIES’ (down to species level), ‘SPECIES_2’ (down to one of two very similar species that can only be sperated from each other by dissecting), ‘GENUS’ (to genus level), ‘FAMILY’ (to family level), or ‘UNID’ (unidentified moth). ‘BROAD_CAT’: Broad larval foodplant category. Obtained using information available in Waring and Townsend (2018) and Sterling and Parson (2020). ‘Macro_micro’: Defines whether the species is categorised as a macro moth or micro moth according to identification guides Waring and Townsend (2018) and Sterling and Parson (2020). ‘Analysis_cat’: Guilds that moths were assigned to for statistical analysis. Guilds included are ‘Farmland’ (species associated with farmland, feeding on grasses or arable/pasture weeds in their larval form), ‘Woody’ (species that feed on woody plants in their larval form), ‘Broad diet’ (species that feed on both woody plants and grasses/weeds), ‘specialist’ (species that have only one host plant in their larval form), ‘Other’ (species that did not fit descriptions above, such as aquatic species). A guild was sometimes assigned to individuals that were identified to genus only, if all species of that particular genus fitted the guild description. ‘Full_name’: Full species scientific name as given in Waring and Townsend (2018) or Sterling and Parsons (2020). File: Enviornmental_data.csv ‘TRAP’: same as in Moth_species_data.csv ‘Farm_code’: same as ‘FARM’ in Moth_species_data.csv ‘PAIR’: number assigned to each farm of a same pair. Relates to sampling design with farm of the same ‘pair’ number being sampled on the same night. ‘Julien_date’: Julien date that moth trapping took place. ‘Farm_type’: Either ‘mixed’ or ‘arable’. Mixed farms were defined as integrating livestock (either cattle or cattle and sheep) into the crop rotation whilst arable farms were defined as having no livestock integrated or present on the farm. ‘Field_type’: Describes whether the moth trap was placed in between two crop fields (‘CROP_CROP’), in between a crop and pasture field (‘GRASS_CROP’) or in between two pasture fields (‘GRASS_GRASS’). ‘Temperature’: Temperature measured at sunset using a vehicle thermometer. ‘BF’: Wind speed as measured using the Beaufort scale. ‘cloud_cover’: Estimate of percentage cover of cloud at sunset. ‘TREE_HEIGHT’: Estimated maximum height of the hedgerow or treeline directly adjacent to the moth trap. Estimate was always carried out by the same surveyor. ‘Built_1000m’: Proportion of ‘built-up’ areas within a radius of 1000 m from the trap. Calculated in QGIS using OS Open Built Up Areas ‘DIST_WATER’: Distance in meters of moth trap to nearest freshwater body (including ponds, streams and rivers as mapped out in OS mastermaps). ‘Hab_div’: Shannon’s index of habitat diversity within a radius of 250 m, 500 m, or 1000 m, from the moth trap. Obtained using CEH landcover maps with aggregated landcover types: improved grassland, semi-natural (scrub, moorland, unimproved grassland, wetland), woodland (broadleaf and coniferous combined), water (areas of seawater were excluded), buildings, arable (land used for crops). ‘Prop_SN’: Proportion of semi-natural areas within a radius of 250 m, 500 m, and 1000 m from moth traps. Obtained using CEH landcover maps ‘Prop_grass’: Proportion of improved grassland within a radius of 250 m, 500 m, and 1000 m from moth traps. Obtained using CEH landcover maps ‘Prop_arable’: Proportion of arable areas within a radius of 250 m, 500 m, and 1000 m from moth traps. Obtained using CEH landcover maps ‘WED’: Woodland edge density (for all woodland types combined) in meters per hectare within a radius of 250 m, 500 m, and 1000 m from moth traps. ‘Altitude’: Altitude of the farm as measured on Google Earth in meters above sea level, by taking a centre point for each farm. ‘Feature_type’: Classification of the type of woody feature that the moth trap was placed next to. Features was classed as either ‘Hedgerow’, ‘Treeline’ or ‘Woodland’ (woodland strip). ‘Crop_div_250m’: Shannon’s diversity index for crops within a radius of 250 m of moth traps. Crop maps obtained from farmers were used to map out crop types and proportion cover in QGIS. ‘Prop_AES_250m’: Proportion of Agri-Environment Scheme sown mixes within 250 m of moth traps. Agri-Environment Schemes are schemes where farmers are financially rewarded for implementing practices that are mean to benefit the environment. ‘Stocking_rate’: Number of cattle per hectare. Calculated as an average across the whole farm (ie: total cattle divided by total area of pasture available across the farm). For all columns relating to moth abundance, species richness and diversity, raw data from ‘Moths_species_data.csv’ was used to populate columns. ‘Abundance’: Total number of moths in a trap ‘Species richness’ or ‘Richness’: Total number of different moth species in a trap. ‘Diversity or Shannon_div’: Shannon’s diversity index. ‘Farmland’: all moths categorised as ‘farmland' moths (in Moths_species_data.csv ‘Analysis_cat ‘ = ‘Farmland’) ‘Woody’: all moths categorised as feeding solely on woody plants in their larval form (in Moths_species_data.csv ‘Analysis_cat ‘ = ‘Woody’) ‘Macro’: Macro moths ‘Micro’: Micro moths Data processing: All data processing and statistical analysis was carried out in R version 4.3.2. 6. Methods for data collection and processing 6.1. Moth trapping Moth traps were placed in agricultural field margins next to hedgerows, tree lines or woodland strips, as moths use these linear features to navigate the agricultural landscape. Traps were always placed in a cereal field (either winter wheat or spring barley) to reduce the effect of crop type on moth catch). Field margins either had crop fields on either side (‘field-type’ = ‘crop-crop), a crop field on one side with pasture on the other (‘grass-crop) or pasture on both sides of the field margin ‘grass-grass’. Each mixed farm was paired with a nearby arable farm with each pair being surveyed on the same night. Light traps were run from dawn until dusk and each location (or ‘TRAP’) was surveyed once between June and August 2022. Abundance, species richness and Shannon’s diversity index was calculated for macro moths, micro moths, farmland macro moths and farmland micro moths using the vegan package in R (Okansen et al, 2022). 6.2. Landscape metrics Buffers of 250 m, 500 m, and 1000 m radius were created around each moth trap location in QGIS. Proportions of each landcover type, Shannon’s diversity index of landcover and woodland edge density was at each different scale was then calculated for each data point using the landscapemetrics package in R (Hesselbarth et al, 2019). 7. Responsibilities: Persons responsible for data management and collection: Project lead & research assistants. Quality control: project lead and supervisory team Storage back-up: University of Stirling DataSTORRE, Onedrive SRUC. 8. References: Hesselbarth MH, Sciaini M, With KA, Wiegand K, Nowosad J (2019). “landscapemetrics: an open-source R tool to calculate landscape metrics.” Ecography, 42, 1648-1657. Oksanen, J., Simpson, G., Blanchet, F., Kindt, R., Legendre, P., Minchin, P., O'Hara, R., Solymos, P., Stevens, M., Szoecs, E., Wagner, H., Barbour, M., Bedward, M., Bolker, B., Borcard D, Carvalho G, Chirico M, De Caceres M, Durand S, Evangelista H, FitzJohn R., Friendly, M., Furneaux, B., Hannigan, G., Hill, M., Lahti, L., McGlinn, D., Ouellette, M., Ribeiro Cunha, E., Smith, T., Stier, A., Ter Braak, C., Weedon, J. (2022). _vegan: Community Ecology Package_. R package version 2.6-4, . OS Open Built Up Areas [GeoPackage geospatial data], Scale 1:25000, Tiles: GB, Updated: 4 November 2022, Ordnance Survey (GB), Using: EDINA Digimap Ordnance Survey Service, , Downloaded: 2023-06-29 14:30:24.608 OS Open Map Local [TIFF geospatial data], Scale 1:10000. Updated: 26 April 2023, Ordnance Survey (GB), Using: EDINA Digimap Ordnance Survey Service, , Downloaded: 2023-08-08 16:05:57.049 Land Cover map of Great Britain (2021) [TIFF geospatial data], Scale 1:250000, Updated: 10 August 2022, CEH, Using: EDINA Environment Digimap Service, , Downloaded: 2024-05-28 16:45:55.204 Sterling, P., and Henwood, B., illustrated by Lewington, R., 2020. Field Guide to the Caterpillars of Great Britain and Ireland. Bloomsbury Wildlife Guides. Waring, P., Townsend, M., illustrated by Lewington, R., 2018. Field Guide to the Moths of Great Britain and Ireland. Bloomsbury Wildlife Guides. Bllomsbury Publishing.