HadCRUT5 is a gridded dataset of global historical surface temperature anomalies relative to a 1961-1990 reference period. Data are available for each month from January 1850 onwards, on a 5 degree grid and as global and regional average time series. The dataset is a collaborative product of the Met Office Hadley Centre and the Climatic Research Unit at the University of East Anglia.
The current version of HadCRUT5 is HadCRUT.5.1.0.0, available from the download page.
— source: HadCRUT5 Index
A detailed description of the datasets can be found in the
Answers to Frequently Asked Questions.
List of the datafiles that are loaded by the Python script:
HadCRUT.5.1.0.0.analysis.summary_series.global.annual.ncHadCRUT.5.1.0.0.analysis.summary_series.northern_hemisphere.annual.ncHadCRUT.5.1.0.0.analysis.summary_series.southern_hemisphere.annual.nc
HadCRUT5 data are downloaded from: https://www.metoffice.gov.uk/hadobs/hadcrut5/data/HadCRUT.5.1.0.0/download.html
The following plots have been generated by the Python scripts hadcrut5_plot.py and hadcrut5_bars.py.
They require the Python libraries: Matplotlib, netCDF4, NumPy, and Requests.
If Python and the required libraries are not installed on your system, you can simply
install uv and run the commands listed below prefixed
with uv run. For example uv run ./hadcrut5_plot.py.
$ ./hadcrut5_plot.py --help
usage: hadcrut5_plot.py [-h] [-a ANNOTATE] [-f OUTFILE] [-g] [-m SMOOTHER]
[-n] [-p PERIOD] [-s] [-r] [-t TIME_SERIES] [-v]
Parse and plot the HadCRUT5 temperature datasets v2026.1 (stable)
Copyright (C) 2020-2026 Davide Madrisan <d.madrisan@proton.me>
License: GNU General Public License v3.0
options:
-h, --help show this help message and exit
-a, --annotate ANNOTATE
add temperature annotations (0: no annotations, 1
(default): bottom only, 2: all ones
-f, --outfile OUTFILE
name of the output PNG file
-g, --global plot the Global Temperatures
-m, --smoother SMOOTHER
make the lines smoother by using N-year means
-n, --northern Northern Hemisphere Temperatures
-p, --period PERIOD show anomalies related to 1961-1990 (default),
1850-1900, or 1880-1920
-s, --southern Southern Hemisphere Temperatures
-r, --trends add linear-fit trend lines for 1970-2010 and
2010-present to the Global Temperatures, showing the
acceleration of global warming
-t, --time-series TIME_SERIES
do plot the "annual" time series (default) or the
"monthly" one
-v, --verbose make the operation more talkative
examples:
hadcrut5_plot.py
hadcrut5_plot.py --global --annotate=2
hadcrut5_plot.py --period "1850-1900"
hadcrut5_plot.py --period "1850-1900" --smoother 5
hadcrut5_plot.py --period "1880-1920" --outfile HadCRUT5-1880-1920.png
hadcrut5_plot.py --period "1880-1920" --time-series monthly --global
hadcrut5_plot.py --period "1880-1920" --global --trends
hadcrut5_plot.py select the period 1961-90 by default but supports (see the command-line switch--period) two other base periods found in the literature: 1850-1900, and 1880-1920.
$ ./hadcrut5_plot.py --annotate=2 --outfile plots/HadCRUT5-1961-1990.png
$ ./hadcrut5_plot.py --annotate=2 --period "1850-1900" --outfile plots/HadCRUT5-1850-1900.png
$ ./hadcrut5_plot.py --annotate=2 --period "1880-1920" --outfile plots/HadCRUT5-1880-1920.png
The command-line option --trends overlays two linear-fit trend lines on the Global Temperatures: one for 1970-2010
and one for 2010-present, each labelled with its slope in °C/decade. Comparing the two makes the acceleration of
global warming immediately visible: the more recent trend line is both steeper and clearly diverges above the older
one.
$ ./hadcrut5_plot.py --annotate=2 --period "1880-1920" --trends --outfile plots/HadCRUT5-global-trends.png
The 2010-present trend line is also extrapolated 10 years past the last available year, with a boxed annotation
and a matching note in the legend showing the forecast temperature for that future year.
This forecast value is not simply the last observed anomaly plus ten years of the slope. It is the regression
line's own value, i.e. slope × forecast_year + intercept, projected forward. Because annual anomalies are noisy,
the fitted trend line for the last available year rarely matches the actual observed anomaly for that same year
exactly, so forecast value != last observed anomaly + slope × decade. Extrapolating from the regression line
rather than from the last single (noisy) data point gives a more stable estimate of where the underlying trend is
heading.
By adding the command-line option --smoother N you can create the same three plots, but using the N-year means data.
For instance --smoother 5 will get you a better idea of the trend lines.
Image generated for the anomalies related to the period 1880-1920.
$ ./hadcrut5_plot.py --period "1880-1920" --smoother 5 --outfile plots/HadCRUT5-1880-1920-smoother.png
The command-line option --time-series monthly selects the monthly HadCRUT5 datasets (by default the dataset providing the annual means is selected).
Image displying the monthly anomalies related to the period 1880-1920, for the global temperatures only.
$ ./hadcrut5_plot.py --global --period "1880-1920" --time-series monthly
$ ./hadcrut5_animate.py --help
usage: hadcrut5_animate.py [-h] [-b BITRATE] [-f OUTFILE] [--fps FPS]
[--hold HOLD] [-g] [-n] [-p PERIOD] [-s] [-v]
Create an animated plot of the HadCRUT5 temperature datasets v2026.1 (stable)
Copyright (C) 2020-2026 Davide Madrisan <d.madrisan@proton.me>
License: GNU General Public License v3.0
options:
-h, --help show this help message and exit
-b, --bitrate BITRATE
bitrate (in kbps) of the encoded MP4 video, ignored
for GIF output (default: 1800)
-f, --outfile OUTFILE
name of the output file (.mp4 or .gif); if not set the
animation is displayed interactively instead
--fps FPS frames per second of the encoded video (default: 25)
--hold HOLD seconds to hold the last frame at the end of the
animation (default: 2.0)
-g, --global plot the Global Temperatures
-n, --northern Northern Hemisphere Temperatures
-p, --period PERIOD show anomalies related to 1961-1990 (default),
1850-1900, or 1880-1920
-s, --southern Southern Hemisphere Temperatures
-v, --verbose make the operation more talkative
examples:
hadcrut5_animate.py
hadcrut5_animate.py --global
hadcrut5_animate.py --period "1880-1920"
hadcrut5_animate.py --period "1880-1920" --outfile HadCRUT5-1880-1920-animation.mp4
hadcrut5_animate.py --period "1880-1920" --outfile HadCRUT5-1880-1920-animation.gif
hadcrut5_animate.py --fps 30 --hold 5 --outfile HadCRUT5-animation.mp4
This script reveals the selected regions year by year instead of drawing the whole time series at once. Without
--outfile the animation is displayed interactively. An .mp4 output requires ffmpeg to be installed; a .gif
output only needs Pillow (already a Matplotlib dependency). GitHub's Markdown renderer strips <video> tags for
repo-hosted files, so the GIF format is what's embedded below.
$ ./hadcrut5_animate.py --period "1880-1920" --outfile plots/HadCRUT5-1880-1920-animation.gif
usage: hadcrut5_bars.py [-h] [-f OUTFILE] [-p PERIOD] [-v]
Parse and plot the HadCRUT5 temperature datasets v2024.1 (stable)
Copyright (C) 2020-2025 Davide Madrisan <d.madrisan@proton.me>
License: GNU General Public License v3.0
options:
-h, --help show this help message and exit
-f OUTFILE, --outfile OUTFILE
name of the output PNG file
-p PERIOD, --period PERIOD
show anomalies related to 1961-1990 (default), 1850-1900, or 1880-1920
-v, --verbose make the operation more talkative
examples:
hadcrut5_bars.py
hadcrut5_bars.py --period "1850-1900"
hadcrut5_bars.py --period "1880-1920"
hadcrut5_bars.py --outfile HadCRUT5-global.png
The image for to the anomalies related to the period 1880-1920 follows.
$ ./hadcrut5_bars.py --period "1880-1920" --outfile plots/HadCRUT5-global-1880-1920.png
usage: hadcrut5_stripe.py [-h] [-f OUTFILE] [-r {global,northern,southern}] [-v] [-l]
Parse and plot a stripe image of the HadCRUT5 temperature datasets v2024.1 (stable)
Copyright (C) 2020-2025 Davide Madrisan <d.madrisan@proton.me>
License: GNU General Public License v3.0
options:
-h, --help show this help message and exit
-f OUTFILE, --outfile OUTFILE
name of the output PNG file
-r {global,northern,southern}, --region {global,northern,southern}
select between Global (default), Northern, or Southern Temperatures
-v, --verbose make the operation more talkative
-l, --no-labels do not disply the header and footer labels
examples:
hadcrut5_stripe.py
hadcrut5_stripe.py --no-labels --region northern
hadcrut5_stripe.py --region global --outfile HadCRUT5-stripe-global.png
Below is a generated striped image for global anomalies.
$ ./hadcrut5_stripe.py --region global
usage: hadcrut5_close.py [-h] [-f OUTFILE] [-p PERIOD] [-r {global,northern,southern}] [-v]
Parse and plot the approach of HadCRUT5 dataset to a threshold temperature v2025.1 (stable)
Copyright (C) 2020-2025 Davide Madrisan <d.madrisan@proton.me>
License: GNU General Public License v3.0
options:
-h, --help show this help message and exit
-f, --outfile OUTFILE
name of the output PNG file
-p, --period PERIOD show anomalies related to 1961-1990 (default), 1850-1900, or 1880-1920
-r, --region {global,northern,southern}
select between Global (default), Northern, or Southern Temperatures
-v, --verbose make the operation more talkative
examples:
hadcrut5_close.py
hadcrut5_close.py --period "1850-1900" --region global
hadcrut5_close.py --period "1880-1920" --outfile "HadCRUT5-1880-1920-threshold.png"
Below is a generated plot image for global anomalies related to the period 1880-1920 with a threshold set to 1.5°C.
$ ./hadcrut5_close.py --period "1880-1920" --region global
The Python code of this project is released under the GPL-3.0 license. The graphics have a CC-BY4.0 license, so can be used for any purpose as long as credit is given to Madrisan Davide and a link is provided to this website.