Björn Kaiser

If the observation points are not attached to slices (freexyz) you cannot see them in a Slice View.

In following please find a tiny Python snippet addressing [b]doc.budgetFlowCreate()[/b]:
[color=blue]
def postTimeStep(doc):
    abs_sim_time =  doc.getAbsoluteSimulationTime()
    print "Time-step: " + str(abs_sim_time)
    budgetlist = doc.budgetFlowCreate()

    fnBUDGET_out.write(str(abs_sim_time)+ "\t")
    for index in Dirichlet:
        fnBUDGET_out.write(str(doc.budgetQueryFlowAtNode(budgetlist, index))+ "\t")
    fnBUDGET_out.write(str("\n"))
    doc.budgetClose(budgetlist)
[/color]

Please note that [b]doc.budgetFlowCreate()[/b] builds up the matrix for for the budget calculation on each node even if you are not interested on every node. Accordingly, this function could be computationally too demanding if you are interested in a few nodes only. For a small number of nodes I suggest to use doc.budgetCompute().

The budget created by [b]doc.budgetFlowCreate()[/b] runs in legacy mode where Storage/Capture is not discriminated from the Imbalance. In future a new API function will be provided which distinguishes between Storage/Capture and Imbalance.

If I understood you correctly, initially you have a 2D horizontal flow model. 2D horizontal flow models do not have nodal elevations. Instead, the aquifer top elevation as well as the aquifer base elevation are essentially elemental properties within an unconfined architecture. You may see that in the Data Panel.

If you are interested in assigning elevations for a 3D model I suggest to enter the third dimension (Layer configurator) and assign nodal elevations afterwards.

This is a common question in mass transport problem in FEFLOW. Unfortunately, there is no a single approach to answer, instead the conceptualization of the mass-transport boundaries and mass-transport formulation (convective/divergence) in FEFLOW should be addressed at the project scale.
Below I described a non-common boundary configuration, which could help you. If the task is just to apply a “recharge concentration”, this means a combined specification of water inflow and mass load (dispersive and advective) at the same location. The boundary conditions must ensure that the total of dispersive and advective flux equals the prescribed mass load at all times (i.e. your conceptual amount). The best combination is a Mass-Transfer BC (i.e. Cauchy-type and equal to the desired concentration) and a Fluid-Flux BC (Neumann-type) with the default mass transport formulation. The transfer-rate for the Cauchy-BC should be equal to the incoming fluid flux. The fact you are using the convective form of transport allows you a free outflow from the model at outflowing borders. This boundary configuration is valid for weak/strong lateral hydraulic gradients.
Since you mentioned fertilization problems, I would like to mention we have recently extended the FEFLOW capabilities for reactive transport with our FEFLOW plug-in piChem, which is able to couple FEFLOW models (2D/3D) with the reactive engine PhreeqC.

[b]i) If I draw the initial 2D slice with slope and develop 3D layer configuration in transverse(horizontal) direction I cannot model multilayer well. I can fix my gravity in negative y direction. But, if I select the top slice and copy the selection to all layers, it selects nodes in transverse horizontal direction instead of nodes in vertical direction.[/b]
Yes, you are right, if you copy a selection you copy the selection to other slices. If the gravity vector is orientated in negative y-direction a copied nodal selection is subsequently orientated in transversal (horizontal direction). Technically, a Multilayer Well is composed of a 1D Discrete Feature (DF) plus a WellBC at the base of the model. In principle, instead using the “classical” MLW Boundary Condition (BC) you could also assign a DF and a MLW BC manually.

[b]ii) I can have my 2D slice as rectangle and develop 3D layer such that gravity is in negative z direction. After that I can import the elevation for all slice to create slope. But it is very tedious process.[/b]
If the data are externally prepared you do not have a tedious process. You may quickly assign all elevations for different slices at once. Please have a look on our Demonstration Exercise Chapter II.3.4: http://www.mikepoweredbydhi.com/-/media/A1279D59B7E04E1888ADDC9C8A7AC65A.ashx 

If you define a Domain Of Interest (DOI) and a Masking Domain (MD) and if you additionally consider only parts of the DOI boundary that belongs also to the MD, then you take the interface between DOI and MD for the Subdomain Boundary Period Budget Panel into account. You need to activate this option within the Subdomain Boundary Rage Budget Panel before switching to the Subdomain Boundary Period Budget Panel.

If you solve for groundwater flow only you could assign a very small hydraulic conductivity for these layers/elements to account for the impermeable nature of the bedrock. Alternatively, you could deactivate these layers/elements.

What material properties did you change? The heat transport equation in FEFLOW involves advective and thermal diffusive terms. If you decrease the hydraulic conductivity to a very small value you minimize advective forces, but account for thermal diffusion. In contrast, if you decrease the thermal conductivity to a very small value you minimize thermal diffusion, but account for advection.

Please find the notes on fixed bugs here: [url=http://feflow.info/html/bugls_62.pdf] http://feflow.info/html/bugls_62.pdf[/url]

You could change the Runge-Kutta controls in the Problem Settings – Field-Line Computations. Try to adjust the [b]Maximal Runge-Kutta steps per element[/b] and the [b]Maximal element switches[/b] according to your particular flow pattern.