How to make a horizontal line chart with multiple years in-CodePudding

I want to make a horizontal line chart with date on the x axis and y axis is treatment with there being multiple lines for each year of data for average Julian date of first flower for each treatment. I don't know how to make this plot to include multiple years.

I want something like this (see image below) but there would be 2 lines for each treatment type to represent each year.

Code I used for the above plot:

ggplot(flwr_avg_duration_19)  
        geom_segment(aes(x = avg_first_flwr, xend = avg_last_flwr, y = treatment_key, yend = treatment_key, color= as.factor(treatment_key)), size = 2)  
        labs(title = "KBS 2019 Average Flower Duration", x = "Julian Date", y = "Treatment")  
        scale_y_discrete(labels=c("A0" = "Ambient",
                                  "AI" = "Ambient  \n Reduced\n Herbivory",
                                  "W0"="Warmed",
                                  "WI"="Warmed  \n Reduced\n Herbivory"))  
        theme(legend.position = "none")

Here is a subset of my data:

dput(flwr_avg_duration_umbs[1:8,])
structure(list(site = c("umbs", "umbs", "umbs", "umbs", "umbs", 
"umbs", "umbs", "umbs"), year = c(2016L, 2016L, 2016L, 2016L, 
2017L, 2017L, 2017L, 2017L), state = c("ambient", "ambient", 
"warmed", "warmed", "ambient", "ambient", "warmed", "warmed"), 
    treatment_key = c("A0", "AI", "W0", "WI", "A0", "AI", "W0", 
    "WI"), insecticide = c("no_insects", "insects", "no_insects", 
    "insects", "no_insects", "insects", "no_insects", "insects"
    ), year_factor = c(2L, 2L, 2L, 2L, 3L, 3L, 3L, 3L), avg_first_flwr = c(157.5, 
    162.904761904762, 154.76, 160.090909090909, 160.678571428571, 
    161.269230769231, 159.848484848485, 158.695652173913), avg_last_flwr = c(182.384615384615, 
    186.761904761905, 186.64, 186.727272727273, 187.5, 188.423076923077, 
    186.939393939394, 187.608695652174)), row.names = 29:36, class = "data.frame")

CodePudding user response：

You need position_dodge to show the groups by year. Use the group aesthetic to ensure you are dodging by year. You will need to use geom_linerange instead of geom_segment, but that is the natural choice here anyway.

ggplot(flwr_avg_duration_umbs)  
  geom_linerange(aes(xmin = avg_first_flwr, xmax = avg_last_flwr,
                   y = treatment_key, 
                   color = as.factor(treatment_key),
                   group = year), size = 2, 
                 position = position_dodge(width = 0.5))  
  labs(title = "KBS 2019 Average Flower Duration", 
       x = "Julian Date", y = "Treatment")  
  scale_y_discrete(labels = c("A0" = "Ambient",
                              "AI" = "Ambient  \n Reduced\n Herbivory",
                              "W0" = "Warmed",
                              "WI" = "Warmed  \n Reduced\n Herbivory"))  
  theme_bw()  
  theme(legend.position = "none")