ggplot combine line and area in same panel (facet-grid)

Question

I am trying to create a facet grid. It will have three rows and four columns. I want two rows to be line graphs, and one to be a stacked geom_area. Except when I plot this, It always turns up as four rows- two lines, two areas. I would like the two areas on top of each other. How can I fix this?

library(ggplot2)
rawdata_Fig1 <- read.table(text="
Date    Number  mMP Day Rep Parameter   FeType  Value
13/03/2016  1   0   2   1   pH  NA  3.11
13/03/2016  2   0   2   2   pH  NA  3.13
13/03/2016  3   0   2   3   pH  NA  3.43
13/03/2016  4   0   2   4   pH  NA  3.49
21/04/2016  0   10  41  1   pH  NA  2.87
21/04/2016  0   10  41  2   pH  NA  2.87
21/04/2016  0   10  41  3   pH  NA  2.87
21/04/2016  0   10  41  4   pH  NA  2.85
13/03/2016  1   0   2   1   Fe2 Fe2 0.965538462
13/03/2016  57  1   2   1   Fe2 Fe2 0.217846154
13/03/2016  113 5   2   1   Fe2 Fe2 0.300153846
13/03/2016  169 10  2   1   Fe2 Fe2 0.226307692
21/04/2016  0   0   41  1   Fe2 Fe2 9.277692308
21/04/2016  0   1   41  1   Fe2 Fe2 13.92
21/04/2016  0   5   41  1   Fe2 Fe2 4.181538462
21/04/2016  0   10  41  1   Fe2 Fe2 1.066153846
13/03/2016  1   0   2   1   'PO4 uM'    NA  9.39
13/03/2016  2   0   2   2   'PO4 uM'    NA  0.39
13/03/2016  3   0   2   3   'PO4 uM'    NA  -0.06
13/03/2016  4   0   2   4   'PO4 uM'    NA  -0.06
13/03/2016  57  1   2   1   'PO4 uM'    NA  0.09
13/03/2016  58  1   2   2   'PO4 uM'    NA  -0.06
13/03/2016  59  1   2   3   'PO4 uM'    NA  0.54
13/03/2016  60  1   2   4   'PO4 uM'    NA  0.69
13/03/2016  113 5   2   1   'PO4 uM'    NA  113.19
13/03/2016  114 5   2   2   'PO4 uM'    NA  120.54
13/03/2016  115 5   2   3   'PO4 uM'    NA  144.24
13/03/2016  116 5   2   4   'PO4 uM'    NA  101.34
13/03/2016  169 10  2   1   'PO4 uM'    NA  235.17
13/03/2016  170 10  2   2   'PO4 uM'    NA  263.52
13/03/2016  171 10  2   3   'PO4 uM'    NA  252.72
13/03/2016  172 10  2   4   'PO4 uM'    NA  261.72
21/04/2016  0   10  41  1   'PO4 uM'    NA  17.04
21/04/2016  0   10  41  2   'PO4 uM'    NA  26.79
21/04/2016  0   10  41  3   'PO4 uM'    NA  29.49
21/04/2016  0   10  41  4   'PO4 uM'    NA  23.64
13/03/2016  1   0   2   1   TotFe   TotFe   1.316027906
13/03/2016  57  1   2   1   TotFe   TotFe   0.233872429
13/03/2016  113 5   2   1   TotFe   TotFe   0.273511824
13/03/2016  169 10  2   1   TotFe   TotFe   0.158557579
21/04/2016  0   0   41  1   TotFe   TotFe   12.51812087
21/04/2016  0   1   41  1   TotFe   TotFe   18.02799674
21/04/2016  0   5   41  1   TotFe   TotFe   6.167889825
21/04/2016  0   10  41  1   TotFe   TotFe   1.625215185
",header=TRUE)

rawdata_Fig1_Fe <- read.table(text="
Date    Number  mMP Day Rep Parameter Value
13/03/2016  1   0   2   1   Fe2     0.965538462
13/03/2016  57  1   2   1   Fe2     0.217846154
13/03/2016  113 5   2   1   Fe2     0.300153846
13/03/2016  169 10  2   1   Fe2     0.226307692
21/04/2016  0   0   41  1   Fe2     9.277692308
21/04/2016  0   1   41  1   Fe2     13.92
21/04/2016  0   5   41  1   Fe2     4.181538462
21/04/2016  0   10  41  1   Fe2     1.066153846
13/03/2016  1   0   2   1   TotFe       1.316027906
13/03/2016  57  1   2   1   TotFe       0.233872429
13/03/2016  113 5   2   1   TotFe       0.273511824
13/03/2016  169 10  2   1   TotFe       0.158557579
21/04/2016  0   0   41  1   TotFe       12.51812087
21/04/2016  0   1   41  1   TotFe       18.02799674
21/04/2016  0   5   41  1   TotFe       6.167889825
21/04/2016  0   10  41  1   TotFe       1.625215185
",header=TRUE)

Fig1 <-ggplot(rawdata_Fig1,aes(x=Day,y=Value, group=Rep))
Fig1Pan=Fig1 + theme_bw()

Fig1panel=Fig1Pan + facet_grid(Parameter ~ mMP, scales="free") +
  geom_line(data=subset(rawdata_Fig1, Parameter == "pH")) +
  geom_line(data=subset(rawdata_Fig1, Parameter == "PO4 uM")) +
  geom_area(data=rawdata_Fig1_Fe, aes(y=pH, x=Day, fill=Parameter, stat='identity'))

Fig1panel

you have 8 columns and 9 column labels, is there an extra one or is one two words? — rawr, May 20 '16 at 03:08
so you want to combine the two area plots into one plot, basically stack them? not simply reorder the rows so that the two area rows are on top of each other? — rawr, May 20 '16 at 03:19
Correct, stack the area plots. I can do this on its own, when not part of the facet_grid. But not when combining facet and stacking. I've tried position="stack", but maybe not in the right spot? I'm pretty new at ggplot. — Valerie S, May 20 '16 at 03:21

score 3 · Accepted Answer · answered May 20 '16 at 04:10

You can combine the two facet rows and create new fill and group variables to match the facet layers

rawdata_Fig1$Parameter[rawdata_Fig1$Parameter %in% 'Fe2'] <- 'TotFe'
# rawdata_Fig1 <- droplevels(rawdata_Fig1)
rawdata_Fig1_Fe$Parameter2 <- rawdata_Fig1_Fe$Parameter
rawdata_Fig1_Fe$Parameter[rawdata_Fig1_Fe$Parameter %in% 'Fe2'] <- 'TotFe'

Fig1 <-ggplot(rawdata_Fig1,aes(x=Day,y=Value, group=Rep))
Fig1Pan=Fig1 + theme_bw()

Fig1Pan + facet_grid(Parameter ~ mMP, scales="free") +
  geom_line(data=subset(rawdata_Fig1, Parameter == "pH")) +
  geom_line(data=subset(rawdata_Fig1, Parameter == "'PO4 uM'")) +

  ## I only changed these lines
  geom_area(data=rawdata_Fig1_Fe, aes(y=Value, x=Day, group = Parameter2, fill = Parameter2)) +
  labs(fill = 'Parameter')

Thank you!! Took some figuring out on my end (understanding what you did, the code is fine), but it works!! — Valerie S, May 20 '16 at 04:29

ggplot combine line and area in same panel (facet-grid)

1 Answers1