Abstract
The relative accuracy of cloud cover assessment from grid overlays on whole-sky images is discussed. Research shows that grid projection, density and cloud retrieval method are the major variables. Equal area grids using an okta or cloud presence/absence scale are recommended. The latter method is significantly faster, allowing a denser grid to be used, which reveals considerably more information on the spatial structure of the cloud field The International Daylight Measurement Programme grid is evaluated and an alternative 72-sector grid is proposed on the basis of a binary yes/no decision for the presence or absence of cloud at the centre of the grid sectors.
Get full access to this article
View all access options for this article.
