Running the Experiments

This page covers what to do once your design is approved and you have your study ID: what to record while the experiment runs, how to handle short (acute) exposures, and how to fill in the data file. For the form itself and the submission process, see the User Guide.

Record every experimental detail

Please keep a full record of how each experiment was run, beyond the values that go in the data file. Experiments in DARE run in many labs, and these details are what let us compare results fairly and explain differences between them. If in doubt, write it down.

Record at least:

What Details to keep
Timing Dates and times of collection, the start of acclimation, the start of the treatments (day 0), every measurement, and the end of the experiment.
Temperature and salinity Every reading you take, the instrument used and when it was last calibrated. Keep logger files as they come off the logger.
Handling How organisms were moved into the treatments (directly or step by step), stocking density, vessel type and volume, and any medium renewal.
Feeding and light What, how much and when organisms were fed; light:dark cycle and light intensity.
Deviations Anything that didn’t go to plan: temperature or salinity excursions, equipment failures, contamination, deaths outside the expected pattern, missed or extra sampling, and any change from the approved design.
Raw files Images, instrument output and lab notebook pages behind each value. Keep them until the DARE data are published; the core team may ask for them.

Where it goes:

  • Notes about a single measurement (a damaged sample, a value re-measured, the reason for an NA) go in the notes column of that row.
  • Everything else goes in the Message to the core team box when you upload your data, or in an email to the core team. Attach logger files or a short summary if they are long.
  • Changes to the design (a different temperature, an extra treatment, a shorter run) also need an updated metadata file: open yours in the builder, make the change and upload it with your data.

Short (acute) exposures

DARE asks for organisms to be exposed continuously for the whole experiment. Some responses, such as metabolic rate, survival of a sudden change, or behaviour, are measured over short exposures of hours to a few days. If your experiment is short:

  • Keep the full design. Run the ambient control, the elevated temperature and the salinity stress level side by side, for the same length of time, exactly as in a longer experiment.
  • Treat every vessel the same way. Move organisms into the control and the treatments at the same time and by the same method, so handling doesn’t differ between treatments.
  • Decide how organisms reach the treatment conditions, and report it. Either transfer them directly into water already at the target temperature and salinity, or change the conditions step by step. For a stepwise change, record the size of each step and how long it took. Use the same approach in every treatment.
  • Define day 0 clearly. Day 0 is the moment organisms are in their treatment conditions: the time of transfer for a direct transfer, or the time the target was reached for a stepwise change. Say which in your notes, and whether your initial measurement was taken before or after that moment (the builder asks for this too).
  • Record exact times. Fill in the time column for every row, and give elapsed_days as a decimal from day 0, for example 0.25 for 6 hours and 0.5 for 12 hours. The experiment duration in the builder is a whole number of days, so round it up.
  • Allow for handling stress. If moving organisms could affect a measurement (respiration and activity especially), let them settle for the same length of time in every treatment before measuring, and note how long.
  • Keep feeding the same across treatments, and note any period without food before a measurement, such as fasting before respiration.
  • Measure more often if responses are fast. The start and end are the minimum. For exposures of a day or two, measurements a few hours apart show how quickly organisms respond.

Contact the core team before starting if you plan an exposure shorter than one day, or a design that differs from the one approved.

Filling in the data file

Once your design is approved, open the data file in any spreadsheet program and fill in these columns. Rows are sorted by time point, so each sampling day is one block. Each block holds only the measurement types scheduled for that day.

Column What to enter
study_id Your study ID from the core team, in every row
date The actual measurement date, as YYYY-MM-DD
time 24-hour HH:MM, or leave empty
elapsed_days Days since T0
value A number with a decimal point, in the units given in the metadata. Write NA if a value is missing.
qc_flag ok, suspect, below_detection, or empty
notes Free text, including the reason for any NA
  • Don’t edit the other ID and target columns, and don’t delete rows. To add an unplanned time point, copy an existing row.
  • Vessels and individuals: vessel_id names each vessel (for example T20_S620_V3). individual_id names what each value was measured on: a single individual (T20_S620_V3_I2) or a pool of individuals (T20_S620_V3_P1). It is empty for whole-vessel and population-level measurements. experimental_rep is 1 unless you repeat the whole experiment.
  • Measurements: the measurement column numbers the values recorded for a vessel on a day (1, 2, 3…), matching the measurements per vessel per day set in the form.
  • Survival: recorded once per vessel, as the proportion alive.
  • Measured conditions: rows named realized_temp_c and realized_conductivity (or realized_salinity) are for the temperature and salinity you measured. Record them at least at the start and end of the experiment, and preferably throughout.
  • Saving: save as CSV UTF-8 (comma delimited). In Excel, choose this under Save As. LibreOffice, Numbers and Google Sheets use UTF-8 by default.
  • Dates: spreadsheets sometimes reformat dates when saving. To prevent this, format the date column as text before typing.

When the file is complete, run it through the checker and upload it: see Checking the data file in the User Guide.