The image below is a screenshot of forecasted sky cover in percentage for various parts of Arizona. Is the sky cover percentage the percentage of each pixel that has cloud, or is it the percentage of a circle with a certain radius where a certain pixel is the center? Not sure how to word it more clearly.enter image description here

  • $\begingroup$ Just to confuse things, I always thought cloud cover meant how many octants (eights) of the sky was covered, and managed to find an official reference: forecast.weather.gov/glossary.php?word=sky%20condition -- of course, "sky condition" and "sky coverage" may be different things. Finally, you should be able to get "pure" (non-image) data for what you're seeking. $\endgroup$ – user967 Mar 16 '19 at 14:59

The "sky cover" represents the percentage of sky obscured by clouds, as would be visible to an observer (or weather instrument) at the Earth's surface.

The most direct reference I could find to this quantity was in this archived presentation from the American Meteorological Society's 2014 Annual Meeting: "The United States Federal Meteorological Handbook (FMH) No. 1 defines sky cover as “the amount of the celestial dome hidden by clouds and/or obscurations”. User @MikeChristianson also identified a direct definition for this exact National Digital Forecast Database (NDFD) product here, confirming this interpretation: "Sky cover is the expected amount of opaque clouds (in percent) covering the sky valid for the indicated hour."

There is also supporting evidence with some additional detail in the NWS Products and Services Reference Guidebook, specifically Section 9, Tables and Abbreviations: The “sky condition” describes the average percentage of the sky that is covered by opaque clouds (not transparent to light) at a given time, followed by a tabular description of the percentile sky cover forecast interpolated to the forecast image in question:

Table describing sky condition and cloud cover forecast parameters

Slight differences in terminology aside, both of these descriptions equivalently refer to the portion of sky obscured by clouds. For typical weather forecasts (as in the example in question), either would be interpreted as describing the sky as it would be visible to an observer at a given point on the Earth's surface - the visible "celestial dome". See also the definition quoted here from the 2nd edition of the American Meteorology Society's glossary, defining cloud cover as "observed from a particular location," in this case the Earth's surface.

Additionally, as noted in the comments, the definitions are vague enough that the observer could be a person or a weather instrument - naturally, the exact details of what would be visible and quantified depend on the observer, but for a simple use of this forecast parameter, the FMH & NDFD definitions support this interpretation of sky cover.

| improve this answer | |
  • $\begingroup$ Thanks, a good clarification as both refer to the same quantity but exact terminology is variable depending on the source or forecast. Added that and a reference to the definition at the top of the wikipedia article - interestingly, the AMS glossary is my starting point for questions like this but the latest edition omits the mention of any viewing location - still clearly referring to the same thing, though. $\endgroup$ – dplmmr Mar 15 '19 at 1:16
  • 1
    $\begingroup$ According to graphical.weather.gov/definitions/defineSky.html and digital.weather.gov/staticpages/definitions.php, "[Sky Cover] Is the expected amount of opaque clouds (in percent) covering the sky valid for the indicated hour." In observation, a ceilometer would generally "look" at a fixed position directly above its location and a trained observer would generally split the visible sky into eight partitions, or oktas. I suspect a forecast would be more like a ceilometer than an observer. $\endgroup$ – Mike Christianson Mar 15 '19 at 6:05
  • 1
    $\begingroup$ To add to @MikeChristianson's comment: en.wikipedia.org/wiki/Okta $\endgroup$ – user967 Mar 16 '19 at 15:03
  • $\begingroup$ Thanks @MikeChristianson and Barry Carter, I spent a while searching specifically for NDFD definitions and was somehow never able to find that. The observer vs. instrument question is valid, but I initially felt that would be getting too far into the details for a question focused on simple forecast interpretation. I'll add in a citation and a mention of that, though. $\endgroup$ – dplmmr Mar 16 '19 at 16:21

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.