README — Turtle feeding experiments
About the data set
Dataset title: Herbivore or Hunter? Ontogenetic shifts in the feeding impact of the invasive turtle Trachemys scripta
Dataset creators and affiliations: Nathan McCard1, 2, 3, Neil E. Coughlan1, 4, J. Robert Britton2, Jaimie T.A. Dick1, 5, and Monica McCard 1, 3, 5
1 Queen's University Marine Laboratory, 12-13 The Strand, Portaferry, BT22 1PF, Co. Down, Northern Ireland, UK. 
2 Department of Life and Environmental Sciences, Faculty of Science and Technology, Bournemouth University, BH12 5BB, UK. 
3 School of Geography and Environmental Sciences, Liverpool John Moore University, James Parsons Building, Byrom Street, Liverpool, L3 3AF, UK
4 School of Biological, Earth and Environmental Sciences, University College Cork, Distillery Fields, North Mall, Cork, Ireland. 
5 Institute for Global Food Security, School of Biological Sciences, Queen’s University Belfast, Medical Biology Centre, Belfast BT9 7BL, Northern Ireland, UK.
Rights holder(s): Data was produced by Nathan McCard and Monica McCard
Publication year: 
Description
This dataset contains experimental feeding observations for small (juvenile) and large (adult) turtles offered different quantities of bivalve or aquatic plant prey. It documents the quantity of prey provided at the start of each trial, the quantity eaten and the quantity remaining, allowing feeding responses to be compared between turtle size classes, prey types and initial prey densities.
Prey recorded are Dreissena polymorpha (zebra mussel), Corbicula fluminea (Asian clam), Elodea canadensis and Potamogeton crispus. Bivalve quantities are recorded as counts of individuals; aquatic plant quantities are recorded by mass in grams, with additional strand counts for E. canadensis. The spreadsheet records a temperature of 22 (°C), and feeding durations of 60 minutes. Plant trials also include measurements 14 days later.
Related publication: NA
Contact details
Principal investigator email: N.G.McCard@ljmu.ac.uk 
Additional contact: M.M.McCard@ljmu.ac.uk 
Terms of use
Copyright: Open Source
Deed - Attribution 4.0 International - Creative Commons
Please cite: McCard, Nathan, Coughlan, Neil, Britton, Robert, Dick, Jamie and McCard, Monica (2026) Herbivore or Hunter? Ontogenetic shifts in the feeding impact of the invasive turtle Trachemys scripta. [Data Collection]
https://doi.org/10.24377/LJMU.d.00000294
Project and funding information
Project title: Herbivore or Hunter? Ontogenetic shifts in the feeding impact of the invasive turtle Trachemys scripta
Project start and end dates: July – September 2019.
Funding organisation and grant number: Not externally funded
Contents
turtleFR.xlsx — Microsoft Excel workbook containing six worksheets, separated by prey species and turtle size class:
* E. canadensis & P. crispus — 64 data rows, juvenile turtles fed E. canadensis and P. crispus.
* E. canadensis — 32 data rows, adult turtles fed E. canadensis.
* P. crispus — 32 data rows, adult turtles fed P. crispus.
* D. polymorpha — 80 data rows, juvenile turtles fed D. polymorpha.
* D. polymorpha — 80 data rows, adult turtles fed D. polymorpha.
* P. crispus — 33 populated data rows, juvenile turtles fed P. crispus.
Workbook format: .xlsx, readable in Microsoft Excel. 
Methods
Feeding experiments were conducted within a round open-top tank (Diameter: 151cm Χ Height: 45cm), using 90 L of de-chlorinated tap water for small turtles and 150 L for the larger turtles. The water was maintained at 22.0 (± 1.0 °C). Predators were allowed to acclimate to experimental arenas for 30 minutes, while prey species were separately acclimated to the temperature regime for 60 minutes prior to experimentation. Each prey species was separately supplied at four or five different densities in a randomised pattern temporally to eliminate time as a confounding factor (bivalves: 2, 4, 8, 16, 20 individuals; macrophytes: 5, 10, 15 and 20 g). Treatment n = 8 per prey species per density. Functional response experiments were initiated through the addition of the allotted prey treatment to experimental tanks containing an individual turtle. Turtles were allowed to feed for a one-hour period before being removed from the experimental arena for enumeration of prey consumed. All 16 turtles were used, eight small and eight large, while no turtle was used more than once at each prey density and prey species to avoid pseudoreplication. Due to the limited numbers of turtles available, individuals were reused systematically in experiments following a designated recovery time (? 48 hours) under standard diet and housing conditions (e.g. Dickey et al., 2021; McCard et al., 2021). Controls consisted of one replicate of each prey type across all densities in the absence of predators.
All statistical analyses were performed in R (R Core Development Team, 2018). To determine the type of FR, logistic regression was applied to analyse the proportion of prey consumed across varying prey densities, with 'prey density' treated as a continuous predictor (Pritchard et al., 2017). A negative linear (first order) coefficient signified a Type II FR, whereas a positive first-order coefficient followed by a negative quadratic (second order) term indicated a Type III FR (Juliano, 2001). To illustrate trends in prey consumption, a locally weighted scatterplot smoothing (LOESS) curve was fitted to the FR data for each of the four prey species using a smoothing parameter of 0.6 (Pritchard et al., 2017). Since prey were not replaced during the trials, Rogers' random predator equation was used to model the functional responses (Rogers, 1972). The Functional Response Ratio (FRR; a/h) was calculated for each prey type using the estimated attack rate (a) and handling time (h), integrating both parameters to provide a composite measure of predatory impact (Cuthbert et al., 2019; South et al., 2019; McCard et al., 2021).
Spreadsheet column dictionary
Shared fields: 'turtle #' = specific turtle identifier; 'prey' = prey taxon; 'temp' = recorded experimental temperature (unit to confirm); 'density' or 'density (g)' = amount initially supplied; 'eaten' or 'eaten (g)' = consumed quantity; 'alive' or 'alive (g)' = quantity remaining; 'time' or 'time (mins)' = feeding-trial duration in minutes.
Bivalve worksheets (x2): 'prey size (mm)' = bivalve size category in millimetres; 'density', 'eaten ' and 'alive ' = counts of individual bivalves. Scientific names are abbreviated as 'D. polymorpha' in some records and written in full in others.
E. canadensis worksheets: 'density (g)', 'eaten (g)', 'alive (g)' = plant mass in grams; 'strands before' and 'strands after' = number of plant strands before and after the trial; 'strands 14 days later' = follow-up strand count; 'weight 14 days later' = follow-up plant weight (presumed grams; confirm).
P. crispus worksheets: 'density (g)', 'eaten (g)', 'alive (g)' = plant mass in grams; 'weight 14 days later' or 'weight 14 days later (g)' = follow-up plant weight. Column naming varies slightly between juvenile and adult worksheets.
RESEARCH DATA README  |  Turtle feeding experiments