Eco-phenotypic feedback loops differ in multi-stressor environments

Density dynamics and cell size and shape dynamics of two freshwater ciliate species (Colpidium striatum and Paramecium aurelia) in a full factorial design of 4 different salinity (0, 0.75, 1.5 and 3 NaCl g/L) and 4 different temperature (20, 22, 24 and 26°C) conditions. 

 "ID" : is the unique ID of each experimental bottles. This way we can link each video to each bottle over time. 
 "Species": Col = colpidium striatum, Pau = paramecium aurelia, PC = competition treatment 
 "Culture" :  states the same either mono(culture - single species) or mixed culture (competition)
  "Salinity" : different salt levels
"Temperature" : different temperature levels
 "Replicate"  : replicates
 "Date" : is the julian date of the year which I like to use to plot the time series, it takes into account if we 'skipped' a day, or whether it is measured in the morning or afternoon
  "Time"  : categorical time points, does not take into account skipping days, looks at total time points
 "True.date" : the actual date of the measurement
 "predicted_species" : using the species identification algorithm which species it is, this is only implemented on the competition treatment
"density" : the average density of a sample
 "mean_area" : bio area , here average bio-area across individuals of that sample
"mean_ar" : cell shape, ar = aspect ratio of minor to major cell size axis
 "sd_mean_area" : the standard deviation of mean_area
 "sd_mean_ar"  : the standard deviation of mean aspect ratio
  "n." : the number of individuals that have been used to estimate the trait values from
"act.time" : helps to make a plot from 0 to end point in time (basically julian.date - max(julian.date))

density: #individuals / mL
mean_area: µm²
mean_ar : unitless