Tidy Tuesday: Historic UK Meteorological Data

170 years of UK weather station data — tracking how temperatures have risen, rainfall patterns have shifted, and sunshine hours have changed across Britain.

Preface

From TidyTuesday repository.

Monthly historical meteorological measurements from 37 UK Met Office stations, spanning from 1853 to 2024. Data includes temperature, rainfall, sunshine, and frost days. The Met Office, established in 1854, is the UK’s national weather service.

• Which UK regions rank highest for rainfall, sunshine, and temperature?
• Which specific years showed extreme meteorological patterns?
• How have monthly weather patterns evolved year-to-year?

Loading necessary packages

My handy booster pack that allows me to install (if needed) and load my usual and favorite packages, as well as some helpful functions.

# Packages ----------------------------------------------------------------

{
  if (!requireNamespace("pak", quietly = TRUE)) {
    install.packages(
      "pak",
      repos = sprintf(
        "https://r-lib.github.io/p/pak/stable/%s/%s/%s",
        .Platform$pkgType,
        R.Version()$os,
        R.Version()$arch
      )
    )
  }

  install_booster_pack <- function(package, load = TRUE) {
    for (pkg in package) {
      if (!requireNamespace(pkg, quietly = TRUE)) {
        pak::pkg_install(pkg)
      }
      if (load) {
        library(pkg, character.only = TRUE)
      }
    }
  }

  if (file.exists('packages.txt')) {
    packages <- read.table('packages.txt')
    install_booster_pack(package = packages$Package, load = FALSE)
    rm(packages)
  } else {
    booster_pack <- c(
      ### IO ----
      'fs',
      'here',
      'janitor',
      'rio',
      'tidyverse',

      ### EDA ----
      'skimr',

      ### Plot ----
      'ggrepel',
      'ggtext',
      'scales',

      ### Misc ----
      'tidytuesdayR'
    )

    install_booster_pack(package = booster_pack, load = TRUE)
    rm(install_booster_pack, booster_pack)
  }

  # Custom Functions ----

  `%ni%` <- Negate(`%in%`)

  geometric_mean <- function(x) {
    exp(mean(log(x[x > 0]), na.rm = TRUE))
  }

  my_skim <- skim_with(
    numeric = sfl(
      n = length,
      min = ~ min(.x, na.rm = T),
      p25 = ~ stats::quantile(., probs = .25, na.rm = TRUE, names = FALSE),
      med = ~ median(.x, na.rm = T),
      p75 = ~ stats::quantile(., probs = .75, na.rm = TRUE, names = FALSE),
      max = ~ max(.x, na.rm = T),
      mean = ~ mean(.x, na.rm = T),
      geo_mean = ~ geometric_mean(.x),
      sd = ~ stats::sd(., na.rm = TRUE),
      hist = ~ inline_hist(., 5)
    ),
    append = FALSE
  )
}

Load raw data from package

raw <- tidytuesdayR::tt_load('2025-10-21')

weather <- raw$historic_station_met
stations <- raw$station_meta

Exploratory Data Analysis

The my_skim() function is a modified version of the skimr::skim() function that returns the number of missing data points (cells as NA) as well as the inverse (e.g.: number of rows that are not NA), the count, minimum, 25%, median, 75%, max, mean, geometric mean, and standard deviation. It also generates a little ASCII histogram. Neat!

Weather Data

weather %>%
  my_skim(.)

Data summary

Name	Piped data
Number of rows	39148
Number of columns	8
_______________________
Column type frequency:
character	1
numeric	7
________________________
Group variables	None

Variable type: character

skim_variable	n_missing	complete_rate	min	max	empty	n_unique	whitespace
station	0	1	4	15	0	37	0

Variable type: numeric

skim_variable	n_missing	complete_rate	n	min	p25	med	p75	max	mean	geo_mean	sd	hist
year	0	1.00	39148	1853.0	1951.0	1978.0	2000.0	2024.0	1971.04	1970.68	37.48	▁▂▃▇▇
month	0	1.00	39148	1.0	4.0	7.0	10.0	12.0	6.50	5.29	3.45	▇▅▅▅▇
tmax	928	0.98	39148	-0.9	8.6	12.4	16.9	28.3	12.80	11.70	5.05	▁▇▇▅▁
tmin	902	0.98	39148	-8.6	2.7	5.6	9.5	17.0	5.99	4.93	4.07	▁▃▇▆▂
af	2327	0.94	39148	0.0	0.0	1.0	5.0	31.0	3.46	4.82	5.16	▇▂▁▁▁
rain	873	0.98	39148	0.0	39.3	62.9	95.5	568.8	73.33	58.39	48.54	▇▂▁▁▁
sun	9168	0.77	39148	2.8	64.7	111.9	163.6	350.3	118.39	99.69	63.24	▇▇▆▂▁

stations

# A tibble: 37 × 5
   station      station_name        opened    lng   lat
   <chr>        <chr>                <dbl>  <dbl> <dbl>
 1 aberporth    Aberporth             1941 -4.57   52.1
 2 armagh       Armagh                1853 -6.65   54.4
 3 ballypatrick Ballypatrick Forest   1961 -6.15   55.2
 4 bradford     Bradford              1908 -1.77   53.8
 5 braemar      Braemar No 2          1959 -3.40   57.0
 6 camborne     Camborne              1978 -5.33   50.2
 7 cambridge    Cambridge Niab        1959  0.102  52.2
 8 cardiff      Cardiff Bute Park     1977 -3.19   51.5
 9 chivenor     Chivenor              1951 -4.15   51.1
10 cwmystwyth   Cwmystwyth            1959 -3.80   52.4
# ℹ 27 more rows

weather %>%
  group_by(station) %>%
  summarize(
    min_year = min(year),
    max_year = max(year),
    n_records = n(),
    .groups = "drop"
  ) %>%
  arrange(min_year)

# A tibble: 37 × 4
   station     min_year max_year n_records
   <chr>          <dbl>    <dbl>     <int>
 1 armagh          1853     2024      2064
 2 oxford          1853     2024      2064
 3 southampton     1855     2000      1743
 4 stornoway       1873     2024      1818
 5 durham          1880     2024      1740
 6 sheffield       1883     2024      1704
 7 bradford        1908     2024      1404
 8 eskdalemuir     1911     2024      1368
 9 lowestoft       1914     2010      1150
10 wickairport     1914     2024      1332
# ℹ 27 more rows

Temperature Trends Over 170 Years

Annual Mean Temperature

annual_temp <- weather %>%
  filter(!is.na(tmax), !is.na(tmin)) %>%
  mutate(tmean = (tmax + tmin) / 2) %>%
  group_by(year) %>%
  summarize(
    mean_temp = mean(tmean, na.rm = TRUE),
    n_stations = n_distinct(station),
    .groups = "drop"
  ) %>%
  filter(n_stations >= 3)

# Baseline period: 1961-1990 (standard climatological reference)
baseline <- annual_temp %>%
  filter(year >= 1961, year <= 1990) %>%
  summarize(baseline_temp = mean(mean_temp)) %>%
  pull(baseline_temp)

annual_temp <- annual_temp %>%
  mutate(anomaly = mean_temp - baseline)

Rainfall Trends

annual_rain <- weather %>%
  filter(!is.na(rain)) %>%
  group_by(year) %>%
  summarize(
    mean_monthly_rain = mean(rain, na.rm = TRUE),
    n_stations = n_distinct(station),
    .groups = "drop"
  ) %>%
  filter(n_stations >= 3)

Station Comparison

Which stations are wettest, warmest, sunniest?

station_summary <- weather %>%
  filter(year >= 1990) %>%
  left_join(stations, by = "station") %>%
  group_by(station_name) %>%
  summarize(
    mean_tmax = mean(tmax, na.rm = TRUE),
    mean_rain = mean(rain, na.rm = TRUE),
    mean_sun = mean(sun, na.rm = TRUE),
    lat = first(lat),
    .groups = "drop"
  ) %>%
  arrange(desc(mean_tmax))

station_summary

# A tibble: 37 × 5
   station_name               mean_tmax mean_rain mean_sun   lat
   <chr>                          <dbl>     <dbl>    <dbl> <dbl>
 1 Heathrow                        15.8      51.5     132.  51.5
 2 Hurn                            15.2      73.3     140.  50.8
 3 Oxford                          15.1      57.2     135.  51.8
 4 Cardiff Bute Park               15.0     101.      130.  51.5
 5 Southampton Mayflower Park      15.0      64.7     142.  50.9
 6 Yeovilton                       15.0      61.0     127.  51.0
 7 Cambridge Niab                  15.0      47.0     126.  52.2
 8 Ross-on-wye                     14.8      64.2     130.  51.9
 9 Chivenor                        14.8      78.1     NaN   51.1
10 Eastbourne                      14.6      65.8     156.  50.8
# ℹ 27 more rows

Visualizing Britain’s Warming Climate

# Climate stripes inspired palette
ggplot(annual_temp, aes(x = year, y = anomaly)) +
  geom_col(
    aes(fill = anomaly),
    width = 1
  ) +
  geom_hline(yintercept = 0, color = "#333333", linewidth = 0.5) +
  geom_smooth(method = "loess", se = FALSE, color = "#000000", linewidth = 1, span = 0.3) +
  scale_fill_gradient2(
    low = "#2166AC",
    mid = "#F7F7F7",
    high = "#B2182B",
    midpoint = 0,
    name = "Anomaly (C)"
  ) +
  scale_x_continuous(breaks = seq(1860, 2020, 20)) +
  labs(
    title = "Britain Is Getting Warmer",
    subtitle = "Annual temperature anomaly relative to 1961–1990 baseline | UK Met Office stations",
    x = "Year",
    y = "Temperature Anomaly (C)",
    caption = "Source: TidyTuesday 2025-10-21 | UK Met Office via Jack Davison"
  ) +
  theme_minimal(base_size = 13) +
  theme(
    plot.title = element_text(face = "bold", size = 18, color = "#1B1B1B"),
    plot.subtitle = element_text(size = 11, color = "#555555"),
    plot.caption = element_text(size = 9, color = "#888888"),
    legend.position = "bottom",
    panel.grid.minor = element_blank()
  ) +
  guides(fill = guide_colorbar(barwidth = 15, barheight = 0.5))

ggplot(station_summary, aes(x = lat, y = mean_tmax)) +
  geom_point(aes(size = mean_rain), color = "#2166AC", alpha = 0.7) +
  geom_text_repel(aes(label = station_name), size = 3, max.overlaps = 12) +
  geom_smooth(method = "lm", se = TRUE, color = "#B2182B", linewidth = 1, alpha = 0.15) +
  scale_size_continuous(name = "Mean Monthly\nRainfall (mm)", range = c(2, 8)) +
  labs(
    title = "Latitude Drives Temperature — But Rain Has Its Own Rules",
    subtitle = "Mean daily max temperature vs. latitude for UK stations (1990–present)",
    x = "Latitude",
    y = "Mean Daily Max Temperature (C)",
    caption = "Source: TidyTuesday 2025-10-21 | UK Met Office"
  ) +
  theme_minimal(base_size = 13) +
  theme(
    plot.title = element_text(face = "bold", size = 17, color = "#1B1B1B"),
    plot.subtitle = element_text(size = 11, color = "#555555"),
    plot.caption = element_text(size = 9, color = "#888888"),
    legend.position = "right",
    panel.grid.minor = element_blank()
  )

Final thoughts and takeaways

The UK Met Office dataset is one of the longest continuous meteorological records in the world, and the warming signal is unmistakable. Every decade since the 1990s has been warmer than the 1961–1990 baseline, with recent years showing anomalies exceeding +1°C. This is entirely consistent with global climate projections.

The latitude gradient tells the familiar story — southern stations are warmer — but rainfall breaks the pattern. Western and upland stations receive far more precipitation due to prevailing Atlantic weather systems, regardless of latitude. This maritime influence makes UK climate uniquely complex for such a small geographic area.

Note

The 1961–1990 baseline period is the World Meteorological Organization’s standard reference for climate normals, though WMO has since added a 1991–2020 reference period. Using the older baseline makes the recent warming trend more visually dramatic — but it’s the standard, not a rhetorical choice.