This page displays tables of most variables logged by the weather station since set up in October 2007, either on a ‘daily’ or a ‘seasonal’ basis.

All daily values are for the 24-hour period covering midnight-midnight.

Thank you to beteljuice from the Cumulus forum for providing this script: https://cumulus.hosiene.co.uk/viewtopic.php?f=14&t=11397

Notes:

All data is midnight-midnight (not the 0900-0900 ‘meteorological day’ sometimes used). This can ‘hide’ some mild nights (where the temperature drops before midnight the next evening), and also double-count cold nights (where both sides of midnight are cold), although the 0900-0900 system also ‘double counts’ cold minima in winter, as the temperature has often not risen much by 9am.

Sunshine - Sunshine is slightly underestimated due to early morning shading of the sensor. A total of around 13.5 ours represents full sun around the summer solstice. Sunshine is recorded when the strength of radiation is >70% of the ‘theoretical maximum’ (above 60w/m^2). While tweaking settings, values are somewhat overestimated in April 2015, and underestimated later in summer 2015.

Wind direction - Data up to March 2012 was calculated in Weatherlink software, and appears to use a different methodology. It appears the data since then calculated in cumulus software is more representative of the average direction.

Also, due to the N-S Otter Valley, wind direction is somewhat biased towards more northerly and southerly directions.

Humidity - the effect of sensor changes does show. Between August 2009 and February the 22nd 2016, Humidity generally read a bit higher (typically 5-7% too high) when the relative humidity was below about 90%. Additionally, the sensor used before August 2009 topped out around 95% (apart from false spikes to 100%), and this affects humidity maxima readings and averages.

Rainfall - totals from 2012 onwards are corrected with a manual rain gauge. Totals before this are from the weather station’s tipping bucket gauge only.