Abstract
Gridded temperature datasets covering the UK at 5 km × 5 km resolution have been analysed for temporal and spatial patterns of change and variability. Decadal‐scale variability is visualized by smoothing the data series using a kernel smoother. Trends have been analysed by comparing averages for a baseline period with averages for a recent period, as well as using linear regression. Spatial patterns of change have been mapped by calculating trends for each grid point. Monthly gridded datasets of daily maximum, minimum and mean air temperatures are available from 1910 to date. These show statistically significant increases for all parts of the UK , with the rate of change increasing from the north and west to the south and east. Seasonally, the least significant changes have occurred in the winter, with significant increases for most districts for all of the other seasons. Derived indices based on daily temperature are available from 1961 onwards. There have been significant decreases in heating degree days and increases in growing degree days for all districts, while changes in cooling degree days are less significant as the values are low and inter‐annual variability dominates. There have been significant decreases in frost days, especially in spring and autumn, and this is linked to significant increases in growing season length, especially for Scotland.