I would recommend that appropriate regression techniques are used instead of plotting simple linear trends. For example, in figure 2 your dependent variable is a number of observations (counts). These should be predicted using Poisson regression. Also, there seems to be a disproportionally high number of 'zeros'. This would be best modelled using zero-inflated Poisson regression (ZIP models, see this link for example : http://www.ats.ucla.edu/stat/stata/dae/zip.htm).
Using ZIP regression, you would first predict the probability of having 'zero' observations, and then predict counts for non-zero observations. Often, both informations are useful and they are not necessarily related to the same covariates. The GAMLSS of pscl packages in R can help you do this.
You can also choose to receive updates via daily or weekly email digests. If you are following multiple preprints then we will send you no more than one email per day or week based on your preferences.
Note: You are now also subscribed to the subject areas of this preprint and will receive updates in the daily or weekly email digests if turned on. You can add specific subject areas through your profile settings.
Usage since published - updated daily