Pick a period (annual, a season, or a specific month) and a
metric to see that metric for that part of the year across the
whole record. Temperatures are means; rainfall is the total. Winter
groups December with the following January and February. This lets you
answer questions like "is April actually getting drier?" or "how have
summer maximums shifted?"
Linear and quadratic trends are compared by AIC; quadratic is preferred
only if it improves AIC by ≥ 2. The shaded grey band marks the
1960–1990 baseline period.
Incomplete years (fewer than 340 daily obs for the whole year,
~84 for a single season, ~28 for a single month) are shown in striped
grey: their values are drawn so gaps are visible, but they're excluded
from the trend fit, baseline mean, 5-year rolling mean, and high/low
stats. Hover any bar for its observation count.
Ground temperature is Cambridge Botanic Garden, 30 cm and 100 cm
depths (1911–2019), shown as two separate series.
Rainfall indices computed from daily NOAA GHCN-D values.
Years with fewer than 350 daily observations are shown in striped grey
and excluded from the trend fit.
Heat metrics
Cold metrics
Counts (frost days, hot days, etc.) are modelled with Negative Binomial
regression: a log link keeps projections non-negative, and overdispersion
(variance much greater than mean here) is handled honestly.
Years with fewer than 340 daily observations are shown in striped grey
and excluded from the trend fit.
Soil temperature extremes from Cambridge Botanic Garden, 1911–2019, at
two depths. Shallow soil (30 cm) responds quickly to air temperature
and frosts; deep soil (100 cm) is heavily damped and is the deepest
indicator of long-term climate signals. Many metrics are very rare —
e.g. true freezing at 30 cm only happened in extreme winters.
Resolution:Smoothing (daily only):
Show:Lines:
Compare two periods:
Daily mean temperature (°C)
Daily maximum temperature (°C)
Daily minimum temperature (°C)
One line per year (or per 5-year window in pentad mode), coloured from
blue (older) to red (recent). Click any line to pin it. Black line +
grey envelope = 1960–1990 mean ±1σ for each day/week.
Variable:
Granularity:Colour scale:
Mean monthly air temperature (°C)
Each cell is one (period, year) value. Rows are time within the year
(top = January, bottom = December), columns are years left → right.
Monthly averages over ~30 days per cell, smoothing out
week-to-week noise to make long-term shifts more visible.
Weekly gives finer detail but is noisier.
Absolute uses the raw value; the colour bar covers the full
data range. Anomaly subtracts the 1960–1990 mean for the same
period, so warmer-than-baseline cells are red and cooler-than-baseline
cells are blue — climate change shows up visually as a leftward-blue /
rightward-red gradient.
Incomplete years (< 340 daily observations) are excluded.
Timescale:View:
SPEI — the Standardised Precipitation-Evapotranspiration Index —
measures drought by comparing the climatic water balance
(precipitation minus potential evapotranspiration, "PET") against the
historical distribution for that location and time of year. It is
expressed like a z-score: 0 is normal, negative is drier than usual,
positive is wetter. Unlike a precipitation-only drought index, SPEI
folds in the evaporative demand from warming, so a hotter year
registers as drier even with the same rainfall.
PET is computed with the Hargreaves method from daily max/min air
temperature. The index is fitted to a Pearson Type III distribution,
calibrated over the WMO 1961–1990 normal period. SPEI-12 reflects
slow-building hydrological drought (reservoirs, groundwater);
SPEI-3 reflects shorter seasonal dry spells (soil moisture,
crops). Bars are coloured on the standard drought scale: browns for
dry, blue-greens for wet.
Date:
Type any date within the covered range and see the daily readings for
that day, along with how the day compared to its day-of-year average
from the 1960–1990 baseline period. Missing values are marked "n/a".