Thursday, February 3, 2011

Sunshine and wheat yields

Malcolm has been plugging away at data from the UK Meteorological Office, calculating mean hours of sunshine in August for our 715 parishes. Now, the data comes from earlier last century and not from 1836, when the wheat yields were recorded, but that’s the best we can do.

August sunshine is a critical factor in wheat yields: best is hot of course, allowing the ear to ripen fully before harvest. I’ve done a regression of wheat yields against hours of sunshine in August, as well as the other data we have: elevation, amounts of rainfall in different groupings. There is a map of sunshine to the right and the regression output is below. The map is interesting, because you can see the parishes in bright yellow, indicating the most sunshine. Along the coasts, and down to the tip of Cornwall. And of course this matches up with where people like to go on holiday. The best wheat yields come from Somerset, inland and with plenty of August sunshine.

Here is the regression output. Elevation has a negative sign, meaning less yield with height. Sunshine is expected. Rainfall is negative above a certain amount. More than 1000 mm in a year waterlogs the wheat plant. This is not a bad result at all, and it will improve once I get data for soil types and amounts of available water into the regression.

No comments: