The primary data forms part of an ongoing monitoring programme by the Shetland Oil Terminal Environmental Advisory Group (SOTEAG). SOTEAG has undertaken biennial benthic surveys in Sullom Voe and the surrounding areas since 1974 to determine the status of infaunal communities and seafloor sediments. For more information on SOTEAG and the types of monitoring they practice please see: https://soteag.org.uk/. A full description of sampling methods can be found here: https://soteag.org.uk/environmental-monitoring/monitoring-reports/ These published data are a subset of the original SOTEAG data. They were used in a 32-year analysis of benthic infaunal invertebrates to investigate changes in species and trait composition, as well as detect long-term temperature-related shifts in benthic species and communities. The manuscript is entitled “Multidecadal changes in coastal benthic species composition and ecosystem functioning occur independently of temperature-driven community shifts” by Armitage et al., 2024. It should be noted that many of these data are a subset of the original and may have been transformed or standardised for statistical analyses. Below is a description of each .csv file, and whether the data are primary or secondary data, and how they are used in the analyses for the published paper. For details on how the data were reduced, please see the publication. In summary, a total of 15 stations were selected with a biennial year range of 1986 – 2018. The depth range covered 4 – 58m, where stations were classified into shallow, intermediate, and deep depth groups. Env.csv Primary data: The environmental data that were collected during sampling by SOTEAG. The data are in long format with each column denoting: Station, Year, Water depth (m), Gravel %, Sand %, Mud %, Grain size, Water temperature (ᵒC), and water salinity. Indices.csv Secondary data: These data are the various species and functional indices created from an analysis using species and trait data. The data are in long format and have row names consisting of the station and year, as well as column headings species richness (sp.rich), total species abundance (abundance) species diversity: Shannon (shannon), species diversity: Simpson (simpson), species evenness (evenness), functional richness (FRic), functional evenneness (FEve), functional diversity, also knowns as dispersion (FDis), functional redundancy (Red) species vulnerability (Vul), sampled station (Station), year of sampling (Year), classified depth group (Depth_group), and the decade (Decade). Species.csv Primary data: This is the SOTEAG species abundance/community list that has been reduced (rare species were removed, a subset of stations were selected, and data were averaged and standardised across replicates). Species_classification_all.csv Secondary data: This is extra supporting information for the species list highlighting the taxonomic ranking (coloumns) and species (rows). STI_and_EcoFun_per_sp.csv Secondary data: This is data produced from a combination of species abundance and traits. The first two columns are row ID and species names. Columns 3-12 describe five ecosystem functions with the calculations/methodology provided in the publication. These data show the mean and scaled (0-1) scores for each function. The column headings equate to bioturbation (Bio), sediment stability (Sed), nutrient recycling (NutR), low trophic position (lowTP), high trophic position (highTP). The following columns X0 – X100 represent the quantiles of temperature affinities, with 50 being the 50% mark. The last column is the number of records (nrec) related to the thermal affinity search. STImacroSoteag.csv Secondary data: Species thermal affinities (methods described in publication) with the first two columns as row ID and species names. The following columns X0 – X100 represent the quantiles of temperature affinities, with 50 being the 50% mark. The last column is the number of records (nrec) related to the thermal affinity search. Traits.csv Primary data: Trait data where each species is listed as a row and traits as a coloumn heading. For trait column names please see publication. Note that the ranges of scores given to a species affinity to a trait is 0-3, which may be split depening on the affinity of the species (i.e., fuzzy coding). Traits_EcoF.csv Secondary data: This is data produced from a combination of species abundance and traits. The first two columns are row ID and species names. Columns represent the trait used for the calculation of the Ecosystem Function (please see equations 2 – 6 in the publication). The column headings equate to bioturbation (Bio), sediment stability (Sed), nutrient recycling (NutR), low trophic position (lowTP), high trophic position (highTP), while the trait abbreviations can be found in Table 1 of the publication. The scores follow the fuzzy coding approach highlighted above (Traits.csv). TrendsVsSTIHats.csv Secondary data: Results from species thermal index analysis and used to plot figure 4 of the main publication. Data is in long format. tsSSTdata.csv Primary data: Sea surface temperatures used in the species distribution model to fit species thermal indices. See publication for details (section 2.3). Vulnerability.csv Secondary data: A species index that was a result from the main species abundance/community data. Index values are given for each species (rows) by station/year (columns). Community means of the index can be found in “indices.csv”.

2024