Proportional dietary DNA as a quantitative metric to monitor the diet composition of piscivorous terns

This archive of data and code accompanies the article: Clucas, G.V., A.N. Stillman, E.C. Craig. 2026. Proportional dietary DNA as a quantitative metric to monitor the diet composition of piscivorous terns. Ornithology. Data 2017 to 2019 COTE visual diet data.csvVisual observations of successful nestling feeding attempts at the tern colony described in the manuscript. Key columns for this analysis include the following: Watch ID. Unique ID of observation bout. Time. Feeding event start time.Nest ID. Unique ID for each nest.Prey category. Prey species group assigned in the field. Prey size. Size of prey item relative to the bill of the bird. A food item the same size as the bill has a value of 1. Estimated prey weight. The estimated weight of the prey item in grams. COTE_Fecal_RRA_grouped.csvRelative read abundances for each fecal sample, with RRA combined into separate prey "groups" for analysis. Rows show individual samples and columns show sample-level metrics like RRA and metadata. Key columns for analysis include:index. Unique ID for every sample. [fish species group]. Relative read abundance of each fish species group detected in the sample. "Other" includes all species not included in the other 4 groups. Year. Year in which the sample was collected. COTE_Fecal_RRA_ungrouped.csvRelative read abundances for each fecal sample, separated by prey species. Rows show individual samples and columns show sample-level metrics like RRA and metadata. Key columns for analysis include:index. Unique ID for every sample. [fish species name]. Relative read abundance of each fish species detected in the sample. Year. Year in which the sample was collected. Code 1_Data_prep.R: Starting with raw data, clean and organize the data and calculate diet metrics. Export Output.csv as a starting point for downstream analysis. 2_Stats_compare.R: Conduct statistical comparisons of different diet metrics. 3_Simulations_FOO.R: Use subsampling to simulate the effect of fecal sample size on MAE, RMSE, and interannual correlation using the FOO diet metric.3_Simulations_RRA.R: Use subsampling to simulate the effect of fecal sample size on MAE, RMSE, and interannual correlation using the RRA diet metric.4_FOO_RRA_all_species_Supplement.R: Create a supplemental table with information on FOO and RRA metrics for all species separately. Assess correlation between metrics. Figure1_Barplot.R: R code for Figure 1. Figure3_Lineplot.R: R code for Figure 3. Figure4_SimulationPlot_RRA.R: R code for Figure 4.FigureS3_SimulationPlot_FOO.R: R code for supplemental figure S3.