Switch plots to ggplot

[j-harbin]

I will also make sure that the legends are lining up

Aquaculture Science functional group

[j-harbin]

also make a consistent color for funding sources in legend

[remi-daigle]

in case it's helpful, I recreated your base plot in ggplot like this (this is a bit more fine tuning than pictured above):

library(dataSPA)
library(ggplot2)
library(scales)
library(dplyr)


om <- getData(type="om", cookie=cookie, age=100)

data <- om |> 
  filter(functional_group=="Aquaculture Science",
         !(fiscal_year %in% c("2025-2026","2026-2027"))) |> 
  group_by(fiscal_year,funding_source_display) |> 
  reframe(funding_amount=sum(amount))

ggplot(data)+
  geom_col(aes(x=fiscal_year,y=funding_amount,fill=funding_source_display))+
  theme_classic()+
  theme(axis.text.x = element_text(angle = 90,vjust=0.5),
        legend.title=element_blank())+
  scale_y_continuous(labels = label_dollar(),
                     expand = c(0,0.05))+
  labs(y="Amount of O&M Funding",
       x=NULL)

We can also manually set a universal palette like this:

om_summary <- om |> 
  group_by(funding_source_display)|> 
  reframe(funding_amount=sum(amount)) |> 
  arrange(funding_amount)

funding_palette <- colorRampPalette(brewer.pal(4,"Paired"))(nrow(om_summary))
names(funding_palette) <- om_summary$funding_source_display

data <- om |> 
  filter(functional_group=="Aquaculture Science",
         !(fiscal_year %in% c("2025-2026","2026-2027"))) |> 
  group_by(fiscal_year,funding_source_display) |> 
  reframe(funding_amount=sum(amount))

ggplot(data)+
  geom_col(aes(x=fiscal_year,y=funding_amount,fill=funding_source_display))+
  theme_classic()+
  theme(axis.text.x = element_text(angle = 90,vjust=0.5),
        legend.title=element_blank())+
  scale_y_continuous(labels = label_dollar(),
                     expand = c(0,0.05))+
  labs(y="Amount of O&M Funding",
       x=NULL)+
  scale_fill_manual(values=funding_palette)

BUT I'm really not sure what logic to follow in terms of assigning colours. We have 68 different funding sources and there is a 0% chance of being able to discriminate 68 funding sources by colour alone. That being said, maybe we can do something like all "A-base" funding sources are 'greens' and all "B-base" are 'blues'? Or even have a look at the frequently used/co-occuring funding sources and try to maximize their visual differences?

[j-harbin]

Some done in commit b721cfb and commit 629df98.

[j-harbin]

commit c3accdb

[j-harbin]

commit f47f5b0

[j-harbin]

We should fix the x axis to be vertical.