User Graph DIST

[Charlie]

Im trying to understand how to make a distance vs concentration graph for a transport model for a specific shift/time.

How do I make sure that the distance vs conc. graph is for a particular shift in the model (e.g. the final shift)?

I can plot pore volumes (or time) vs concentrations but I'm just not 100% which shift the DIST command relates to.
Some parts from my input below. Thanks in advance
>>

TRANSPORT         
-cells   5      
-lengths    5*2
-shifts   500      
-time_step    20 yr   
#-porosities   5*0.4
-flow_direction forward         
-boundary_conditions         flux constant
-dispersivities      5*1
-correct_disp         true
-diffusion_coefficient         1.0e-9   ### m2/s
#-punch 5                          ### for PORE VOL AND TIME GRAPHS -  Punch a specific cell i.e. last
-punch_cells        1-5          ## FOR distance GRAPHS ???
-punch_frequency    500    ## LINKED TO LINE ABOVE FOR DISTANCE GRAPHS - plots a specific shift. ie. LAST SHIFT (END) ???

USER_GRAPH 1             ##### to link this to a particular cell - SEE the -punch 5 line in TRANSPORT
    -chart_title      "CELL RESULTS - PORE VOLUMES"
    -headings               PORE_VOL EQUI_calcite KIN_dolomite TOT_Cl-
    -axis_titles            "Pore Volumes" "KIN_mineral" "TOT Cl"
    -initial_solutions      false
    -connect_simulations    true
###-plot_concentration_vs time
-start
10 graph_x (step_no + 0.5)/cell_no       ## tranfer cell centered conc to column end
20 graph_y EQUI("calcite"), KIN("dolomite") 
30 GRAPH_sy TOT("Cl")
-end


USER_GRAPH 2    ##### to link this to a particular cell - SEE the -punch 5 line in TRANSPORT

    -chart_title      "FINAL CELL RESULTS - TIME SERIES - BREAKTHROUGH"
    -headings               TIME EQUI_calcite EQUI_dolomite TOT_Cl-
    -axis_titles            "TIME YEARS" "mineral" "TOT Cl"
    -initial_solutions      false
    -connect_simulations    true
  ##  -plot_concentration_vs  x
  -start
10 GRAPH_X TOTAL_TIME/(60*60*24*365)       ## CONVERTS TO YEARS
20 graph_y EQUI("calcite"), KIN("dolomite")
30 GRAPH_sy TOT("Cl")
-end


USER_GRAPH 3       ###### Do I need to link to the PUNCH_CELLS AND PUNCH_FREQUENCY lines in TRANSPORT ??
   -chart_title      "DISTANCE CELLS"
    -headings               TIME EQUI_calcite KIN_dolomite TOT_Cl- 
    -axis_titles            "Distance, in m"  "mineral" "TOT Cl"
    -initial_solutions      false
    -connect_simulations    true
    -plot_concentration_vs  x
  -start
10 GRAPH_X DIST
20 GRAPH_Y EQUI("calcite"), KIN("dolomite")
30 GRAPH_sy TOT("Cl")
  -end
   
END

[Charlie]

I think I have the solution now based on this Q&A;

https://phreeqcusers.org/index.php/topic,187.msg532.html#msg532

[dlparkhurst]

DIST is dependent only on the -lengths definition. DIST for cell 1 is half the length of cell 1, and in general DIST for a cell is the midpoint of the cell.

[Charlie]

Thanks David that makes sense. I am still unsure as to how I can plot a distance vs concentration plot in USER GRAPH for a specific step or shift. Using the link above I modified my input to the below, but it doesn't appear to have made a difference. My user graphs for pore volumes and time vs concentrations work OK, however the distance graph doesn't show the same breakthrough for the conservative tracer (Cl). 

TRANSPORT         
-cells   5      
-lengths    5*2
-shifts   500      
-time_step    20 yr   
#-porosities   5*0.4
-flow_direction forward         
-boundary_conditions         flux constant
-dispersivities      5*1
-correct_disp         true
-diffusion_coefficient         1.0e-9   ### m2/s
#-punch 5    ### for PORE VOL AND TIME GRAPHS - TOGGLE OFF FOR DISTANCE GRAPHS! Punch a specific cell i.e. last
-punch_cells        1-5 ## FOR distance GRAPHS
-punch_frequency    500 ## KEEP LINKED TO LINE ABOVE FOR DISTANCE GRAPHS - plots a specific shift. (last)


USER_GRAPH 3 ######  need to link to the PUNCH_CELLS AND PUNCH_FREQUENCY lines in TRANSPORT
   -chart_title      "DISTANCE CELLS"
    -headings               TIME EQUI_calcite KIN_dolomite TOT_Cl- 
    -axis_titles            "Distance, in m"  "mineral" "TOT Cl"
    -initial_solutions      false
    -connect_simulations    true
    -plot_concentration_vs  x
  -start
10 if (step_no <> 500) then goto 100
20 GRAPH_X DIST
30 GRAPH_Y EQUI("calcite"), KIN("dolomite")
40 GRAPH_sy TOT("Cl")
100 REM end
  -end
   
END

[dlparkhurst]

With 5 cells, SOLUTION 0 will be advected to the end of the column in 5 shifts.

Assuming you have no Cl in your column initiallly, and 1 mmol/kgw in solution 0, if you have a flux boundary at the end of the column, the concentration at cell 5 will be slightly greater than 0.5 at shift 5. It should be 0.5 at the boundary, but the node for the cell is the midpoint of cell 5--a half length before the end of the column. Ultimately, the concentration of Cl in the column would equal the concentration in SOLUTION 0, assuming Cl is conservative.

However, you have chosen a constant boundary condition for the far end of the column. In this case the concentration at the end of the column will be determined by the concentration of SOLUTION 6, or, if undefined, the initial solution for cell 5 (I think). Thus, regardless of the number of shifts, the concentration at the far boundary does not change, and given some dispersion, the concentration at the node of cell 5 will be similar, but greater than the boundary. Ultimately, a steady state evolves that will have concentrations near SOLUTION 0 for Cl near the inflowing boundary with a transition to the far boundary condition in the higher numbered cells. With your dispersivity (and a tiny effect from diffusion), the steady state for a conservative constituent sets up in 20 to 50 shifts.

[Charlie]

Thanks David for clarifying and picking that up! Much appreciated.