Chapter 4 Interactive Tools

Here are some tools to help with interactive viewing of data. We use plotly for visualizing data in gpplot, and leaflet and mapview for visualizing spatial data. 

4.1 Plotly - Interactive plotting

Load packages.

library(ggplot2)
library(lubridate)
library(plotly)

We are using the iris dataset that comes with R.

head(iris)

##   Sepal.Length Sepal.Width Petal.Length Petal.Width Species
## 1          5.1         3.5          1.4         0.2  setosa
## 2          4.9         3.0          1.4         0.2  setosa
## 3          4.7         3.2          1.3         0.2  setosa
## 4          4.6         3.1          1.5         0.2  setosa
## 5          5.0         3.6          1.4         0.2  setosa
## 6          5.4         3.9          1.7         0.4  setosa

Make a plot.

iris1 <- ggplot(iris, aes(Sepal.Length, Sepal.Width, color = Species)) + geom_point()

Use plotly to make plot interactive. You can draw a box around an area to zoom in, double click on plot to zoom back out. If you click on virginica in the legend it will remove that from the plot.

ggplotly(iris1)

You might want to change what shows up when you hover over the point. You can do this by adding the “label” parameter to your aes() call.

iris2 <- ggplot(iris, aes(Sepal.Length, Sepal.Width, color = Species, label = Petal.Length)) + geom_point()
ggplotly(iris2)

If you want even more customization of what shows up, you can use the “text” parameter in your aes() call. You can then customize the combination of what shows up: x, y, text.

Note ‘
’ inserts a line break into your popup.

iris3 <- ggplot(data = iris, aes(x = Sepal.Length, y = Sepal.Width,
                    text = paste('Name of species: ', Species,
                                 '<br>Petal length: ', Petal.Length)))+
  geom_point()

# customize whether you want to display x, y, and text, or just text
ggplotly(iris3)

ggplotly(iris3, tooltip = "text") 

ggplotly(iris3, tooltip = c("x", "text"))

If you don’t want the popup at all, you can also get rid of it.

# Use tooltip = NULL to get rid of all tooltip 
ggplotly(iris3, tooltip = NULL)

Resources: * https://plotly.com/r/hover-text-and-formatting/ * https://stackoverflow.com/questions/36325154/how-to-choose-variable-to-display-in-tooltip-when-using-ggplotly

4.2 Leaflet

library(readr)
library(leaflet)
library(viridis)
library(dplyr)

Lets use some station data from water temperature dataset.

stations_URL <- "https://portal.edirepository.org/nis/dataviewer?packageid=edi.591.2&entityid=a059f5eea4f8500fe1a43566451ec593"

stations <- readr::read_csv(stations_URL) %>%
  mutate(County = factor(County)) %>%
  mutate(Temperature = runif(n = nrow(.), min = 8, max = 30)) # Add a column of made up temperature data
dplyr::glimpse(stations)

## Rows: 138
## Columns: 12
## $ Station          <chr> "ANC", "ANH", "BAC", "BDL", "BDT", "BET", "BIR", "BKS~
## $ StationName      <chr> "ANTIOCH (USBR)", "SAN JOAQUIN RIVER AT ANTIOCH", "BA~
## $ StartDateDataset <date> 1999-02-28, 1994-12-31, 2008-04-17, 2008-02-27, 2005~
## $ EndDateDataset   <date> 2016-06-09, 2017-07-20, 2016-11-30, 2017-12-27, 2008~
## $ Agency           <chr> "USBR", "DWR", "DWR", "DWR", "DWR", "DWR", "DWR", "DW~
## $ Latitude         <dbl> 38.01782, 38.01783, 37.96935, 38.18690, 37.86500, 38.~
## $ Longitude        <dbl> -121.8030, -121.8030, -121.5722, -121.9708, -121.3231~
## $ HydrologicArea   <chr> "Sacramento River", "Sacramento River", "Sacramento R~
## $ Basin            <chr> "Delta", "San Joaquin R", "Delta", "Sacramento R", "S~
## $ County           <fct> Contra Costa, Contra Costa, San Joaquin, Solano, San ~
## $ HabitatType      <chr> "Tidal River Channel (brackish)", "Tidal River Channe~
## $ Temperature      <dbl> 10.271113, 21.343277, 23.601055, 16.809101, 21.365530~

Define your color palette. How do you want to color your stations?

# Let's say we are coloring by County. 
unique(stations$County) # There are 6 counties

## [1] Contra Costa San Joaquin  Solano       Sacramento   Alameda     
## [6] Yolo        
## Levels: Alameda Contra Costa Sacramento San Joaquin Solano Yolo

# We will pick 6 colors from the viridis package. 
pal <- colorFactor(viridis::viridis(6), stations$County)

You could alternatively use colorNumeric, colorBin, or colorQuantile as an argument. See https://rstudio.github.io/leaflet/colors.html

Make your plot.

stations %>% # call data frame.
  leaflet() %>% # call leaflet.
  addTiles() %>% # this adds the map in the background.
  addCircleMarkers(
    color = ~pal(County),
    stroke = FALSE, # alters whether there is a border to circle
    fillOpacity = 0.8,
    lng = ~Longitude, # call your longitude column name
    lat = ~Latitude, # call you latitude column name
  label = ~paste(Station, "Lat:", Latitude, "Long:", Longitude)) %>% # edit what you want to show up in your label
  addLegend(pal = pal,
            values = ~County,
            position = "bottomright")

If you want to change the radius of your points based on a variable, you can also do that.

Define your palette colors again. In this example, we picked individual colors rather than using an existing palette, and assign them directly to each basin.

unique(stations$Basin) # need 7

## [1] "Delta"         "San Joaquin R" "Sacramento R"  "East Bay"     
## [5] "Mokelumne R"   "Stanislaus R"  "Cache Cr"

pal2 <- colorFactor(c("purple3", "blue3", "forestgreen", "orange", "orangered", "hotpink", "grey"), domain = c("Cache Cr", "Delta", "East Bay", "Mokelumne R", "Sacramento R", "San Joaquin R", "Stanislaus R"))

Make map, this time defining the “radius” option.

stations %>%
  leaflet() %>%
  addTiles() %>%
  addCircleMarkers(
    fillColor = ~pal2(Basin),
    radius = ~sqrt(Temperature), # modify the size of circles based on the temperature
    stroke = FALSE, 
    fillOpacity = 0.6,
    lng = ~Longitude,
    lat = ~Latitude,
    label = ~paste(Station, "Lat:", Latitude, "Long:", Longitude, "Temperature:", Temperature))%>%
  addLegend(pal = pal2,
            values = ~Basin,
            position = "topright")

You can also have the labels be more permanent. See labelOptions below (noHide = T). If textOnly = FALSE, there will be a white box around your label. Doesn’t work great for this plot, since there are so many stations.

stations %>%
  leaflet() %>%
  addTiles() %>%
  addCircleMarkers(
    fillColor = ~pal2(Basin),
    radius = ~sqrt(Temperature), # modify the size of circles based on the temperature
    stroke = FALSE, 
    fillOpacity = 0.6,
    lng = ~Longitude,
    lat = ~Latitude,
    label = ~Station,
    labelOptions = labelOptions(noHide = T, direction = "right", textsize = "7px", textOnly = TRUE))%>%
  addLegend(pal = pal2,
            values = ~Basin,
            position = "topright")

There are other types of markers too. This is the default.

stations %>%
  leaflet() %>%
  addTiles() %>%
  addMarkers(
    lng = ~Longitude,
    lat = ~Latitude,
    label = ~Station)

This is a square.

stations %>%
leaflet() %>% 
  addTiles() %>%
  addRectangles(
    lng1=min(stations$Longitude) - 0.1, lat1=min(stations$Latitude) -0.1 ,
    lng2=max(stations$Longitude) +0.1, lat2=max(stations$Latitude) +0.1,
    fillColor = "transparent",
    color = "orangered"
  ) %>%
  addCircleMarkers(
    lng = ~Longitude,
    lat = ~Latitude,
    label = ~Station,
    radius = 2)

4.3 Mapview

This one is a nice, simple way to just take a look at your points.

library(mapview)
library(sf)

Convert your stations to a spatial object.

# name the columns for longitude and latitude
# crs is the WGS 1984 projection, which is usually the projection for lat/lon data
stations_sf <- sf::st_as_sf(stations, coords = c("Longitude", "Latitude"), crs = 4326)

Just one line to see a decent plot! This one shows you all the associated data linked with each point.

mapview(stations_sf)

Color by variable of interest

mapview(stations_sf, zcol = "Basin")

Size your variables

mapview(stations_sf, zcol = "Basin", cex = "Temperature")