Search Results

You are looking at 1 - 3 of 3 items for

  • Author or Editor: J. London x
  • Refine by Access: All Content x
Clear All Modify Search
Carole J. Hahn, Stephen G. Warren, and Julius London


Visual observations of cloud cover are hindered at night due to inadequate illumination of the clouds. This usually leads to an underestimation of the average cloud cover at night, especially for the amounts of middle and high clouds, in climatologies based on surface observations. The diurnal cycles of cloud amounts, if based on all the surface observations, are therefore in error, but they can be obtained more accurately if the nighttime observations are screened to select those made under sufficient moonlight.

Ten years of nighttime weather observations from the Northern Hemisphere in December were classified according to the illuminance of moonlight or twilight on the cloud tops, and a threshold level of illuminance was determined, above which the clouds are apparently detected adequately. This threshold corresponds to light from a full moon at an elevation angle of 6°, light from a partial moon at higher elevation, or twilight from the sun less than 9° below the horizon. It permits the use of about 38% of the observations made with the sun below the horizon.

The computed diurnal cycles of total cloud cover are altered considerably when this moonlight criterion is imposed. Maximum cloud cover over much of the ocean is new found to be at night or in the morning, whereas computations obtained without benefit of the moonlight criterion, as in our published atlases, showed the time of maximum to be noon or early afternoon in many regions. Cloud cover is greater at night than during the day over the open oceans fair from the continents, particularly in summer. However, near-noon maxima are still evident in the coastal regions, so that the global annual average oceanic cloud cover is still slightly greater during the day than at night by 0.3%. Over land, where daytime maxima are still obtained but with reduced amplitude, average cloud cover is 3.3% greater during the daytime. The diurnal cycles of total cloud cover we obtain are compared with those of ISCCP for a few regions; they are generally in better agreement if the moonlight criterion is imposed on the surface observations.

Using the moonlight criterion, we have analyzed 10 years (1982–91 ) of surface weather observations over land and ocean, worldwide, for total cloud cover and for the frequency of occurrence of clear sky, fog and precipitation. The global average cloud cover (average of day and night) is about 2% higher if the moonlight criterion is imposed than if all observations are used. The difference is greater in winter than in summer, because of the fewer hours of darkness in summer. The amplitude of the annual cycle of total cloud cover over the Arctic Ocean and at the South Pole is diminished by a few percent when the moonlight criterion is imposed.

The average cloud cover for 1982–91 is found to be 55% for Northern Hemisphere land, 53% for Southern Hemisphere land, 66% for Northern Hemisphere ocean, and 70% for Southern Hemisphere ocean, giving a global average of 64%. The global average for daytime is 64.6%; for nighttime 63.3%.

Full access
Stephen G. Warren, Carole J. Hahn, and Julius London


Cloud observations from land stations and from ships in the ocean are used to investigate the frequency of observation and the co-occurrence of different cloud types, and the geographical and seasonal variations of these co-occurrences. Ground-based observations are used because they provide a more definitive identification of clouds by type than do satellite observations. The clouds are grouped into six types (cirrus + cirrostratus + cirrocumulus, altostratus + altocumulus, stratus + stratocumulus, nimbostratus, cumulus, and cumulonimbus). The results are expressed as frequency of occurrence of different cloud types and as contingency probabilities; that is, given that one cloud type is present, the probability that another particular type is also present is computed. Several sources of bias are identified, and their effects on the results are estimated.

It is found that, on the average at all latitudes and in all seasons, clear skies occur more frequently, by a factor of about 4, over land than over the oceans; cumulus occurs twice as frequently over the oceans than over land but cirrus is reported with a somewhat higher frequency over land.

In general, cirrus and altostratus tend to occur together but altostratus and cumulus do not. The probability of co-occurrence of cirrus and cumulonimbus is much higher in the tropics 30°S–30°N than at mid- to subpolar latitudes. When cirrus or altostratus occurs over land, it is much more likely to be alone than when it occurs over ocean. Some of the reasons for these variations are discussed.

Full access
C. A. Barth, R. W. Sanders, G. E. Thomas, G. J. Rottman, D. W. Rusch, R. J. Thomas, G. H. Mount, G. M. Lawrence, J. M. Zawodny, R. A. West, and J. London
Full access