Trajectory Analysis

Analyse and visualise HYSPLIT trajectory files.

library(ggopenair)
library(ggplot2)
library(sf)
#> Linking to GEOS 3.10.2, GDAL 3.4.1, PROJ 8.2.1; sf_use_s2() is TRUE

Importing Data

Currently, trajectories should be imported using openair as this functionality hasn’t yet been ported to ggopenair.

# import trajectories
traj <- openair::importTraj(site = "london", year = 2010)

# import data for North Kensington
kc1 <- openair::importAURN("kc1", year = 2010)

kc1 <- dplyr::select(kc1, date, nox, no2, o3, pm2.5, pm10)

# now merge with trajectory data by 'date'
traj <- dplyr::left_join(traj, kc1, by = "date")

traj_red <-
  openair::selectByDate(
    traj,
    start = "15/4/2010",
    end = "21/4/2010"
  )

Path Plots

ggopenair exports traj_plot(), which is extremely similar to trajPlot().

traj_plot(traj_red, colour = "pm2.5") +
  scale_opencolours_c() +
  theme_classic()

Binned Plots

traj_level() is also exported, which again behaves almost identically to trajLevel(). All statistic types are supported. Note that using scale transforms on the “fill” aesthetic is almost always recommended.

traj_level(traj_red) +
  scale_opencolours_c(
    trans = "sqrt",
    label = scales::label_percent(scale = 1)
  ) +
  labs(fill = "% Trajectories\n(sqrt scale)\n") +
  theme_classic()

Traj Clustering

traj_cluster() also exists in ggopenair, but is waiting on the next CRAN release of openair to be fully functional.

–> –> –> –> –>