Questions about the example codes of ElectrostaticPIC

[HoulinLee]

Dear all,
I have some questions about the example codes of ElectrostaticPIC.

1. In the function interpolate_cic_two_levels at amrex-tutorials-main/ExampleCodes/Particles/ElectrostaticPIC/src/particles/field_gather/FieldGather_K.H, the codes if (lev == 1 && mask(i,j,0,0) == 1) {} else {} indicate that the electric field of coarser grid is used near the level boundary when the current level is level 1. I wonder why we should gather the particle field in this way.
2. In amrex-tutorials-main/ExampleCodes/Particles/ElectrostaticPIC/src/field_solver/FieldSolver.cpp, setLevelBC is used in the function computePhi. I noticed that the multifabs phi and rhs are nodal types, but the comments of setLevelBC in amrex/Src/LinearSolvers/MLMG/AMReX_MLLinOp.H say that setLevelBC is only for cell-centered solves only.
3. In the function sum_fine_to_crse_nodal at amrex-tutorials-main/ExampleCodes/Particles/ElectrostaticPIC/src/field_solver/FieldSolver_K.H, should fine(ii-1,jj+1,0,0) be fine(ii+1,jj-1,0,0)?

Thanks.

[atmyers]

1. This is done to mitigate self-force errors that occur when the macroparticle is near the coarse/fine boundaries - the method that this example code implements is described here.

2. Yes, this function is a no-op in the nodal case. The call doesn't really need to be there, but it doesn't hurt anything to have it.

3. I believe you are correct, thank you for finding and reporting this. Fix typo in Electrostatic PIC example reported by HoulinLee. amrex-tutorials#46

[HoulinLee]

Thanks for your reply!