Peter Schätzl Grundwassermodellierer

No, this is currently not possible.

No, you cannot directly add to a model. You can, however, combine two 2D meshes (in any map format) if they fit at the common boundary via the fem import when creating a new model.

You don't need a script for this, simply use a Concentration-type BC along with a minimum mass flow constraint with a value of 0. I think there's also an example model for this in the benchmarks collection available for download (with the demo data package), reproducing the Henry problem. Using the convective form of the transport equation, an outflowing boundary will let the salt mass pass with whatever concentration it has at the boundary. In divergence form, this works as well, but might be less stable as the mass outflow is based on an internal calculation that might not always be precise enough rather than the formulation of the equations as in convective form.

This would be a job for either a Python script or a plug-in for FEFLOW. The FEFLOW support might have a plug-in ready that fulfills your needs (not sure, though).

I fully second Durchlässigkeitsbeiwert's answers. Very likely you need to smoothen your curves. IN addition, you might also pay some attention to mesh quality which may also have a negative impact on convergence.

Indeed with a seepage face you would expect to get less abstraction than with wells. Why this is the opposite in your model can only be found out by analyzing the model.

There's a panel called "Content" that can be used to derive the integrated mass.

There's no straightforward / supported way of loading external packages into the python installation brought by FEFLOW (and used when using the Python window in FEFLOW). If you need other packages, you should prefer the way Carlos is referring to: run your script on a separate Python installation and load FEFLOW (as package ifm) into your script (option two on http://www.feflow.info/html/help74/feflow/13_Programming/Python/python.html).

With some internet research, you might be able to find some ballpark numbers for thermal resistance between air and soil. These values depend on vegetation, wind speed, etc. - but at least you'll get an idea (I remember having found some values about two years ago - but I don't have access to them anymore). Indeed all processes are involved by some degree, but typically air temperature will dominate.

There may be multiple influences on groundwater temperature over the year. With your BCs you simulate the dependency on air temperature, either assuming that air temperature is equal to the temperature of the soil top, or by applying an additional resistance (heat-transfer BC). Depending on local conditions, you may have influences of solar radiation (on blacktop, for example), infiltration of rainwater, evaporation, ... Typically, however, I'd assume that by using a temperature BC you'd rather overestimate than underestimate the influence of air temperature. So hard to tell why in your model you observe such a small impact. I guess you've checked cp and thermal conductivity already. If there's an unsat zone, it could have an influence, but with typical porosity values, this would be small, too (FEFLOW assumes air to be non-conductive for heat).