Cloud cover has first order impacts on the Arctic surface radiation balance (see Chapter 5). Cloud microphysical and radiative properties are hence a vibrant area of research. While much is being learned from modeling and special observation programs (e.g., SHEBA and the Department of Energy Atmospheric Radiation Monitoring program), a continuing problem is the general lack of accurate data on even cloud amounts over the Arctic. Curry et al. (1996) provide a comprehensive review of the problem.
Over the Arctic, one can observe many of the same "generic" low-level (e.g., stratus, stratocumulus), middle-level (e.g., altocumulus) and high-level (cirrus) cloud types as seen in middle latitudes. At the summer mesopause (roughly 85 km) noctilucent clouds may form. They are composed of minute ice crystals through upwelling of trace amounts of water vapor. Despite the high latitude of the Arctic, convective cloud cover is common over Eurasian and Alaskan land areas during summer. Convective cloud cover is also frequent during winter over the Norwegian Sea - cold outbreaks from the north, when reaching the fairly warm, ice-free waters off the Norwegian coast, result in destabilization of the air.
However, cloud classification can be problematic in the Arctic. Curry et al. (1996) identify four "unusual" boundary layer cloud types over the Arctic Ocean, the first two falling into the general category of low-level, optically thin Arctic stratus: (1) extensive summertime boundary layer clouds with multiple layers; (2) mixed-phase (ice and water) boundary layer clouds during the transitional season; (3) low-level ice crystal clouds and "clear-sky" ice crystal precipitation (the latter often termed "diamond dust") in stable wintertime boundary layers; (4) wintertime ice crystal "plumes" emanating from open leads (Schnell et al., 1989). Measurements in the western Arctic during SHEBA (Intrieri and Shupe, 2004) show that "true" clear-sky diamond dust occurs about 13% of the time during November through mid-May. However, the occurrence of diamond dust is often greatly over-reported (145% of the time) by observers, mainly during polar darkness. Lidar profiles showed that for nearly all these misreports, ice crystals were actually precipitating from clouds.
Overall, cloud cover is least extensive during winter and most extensive during summer. This seasonality is well expressed over the central Arctic Ocean and is primarily driven by the seasonality in low-level stratus. By comparison, the Atlantic sector of the Arctic Ocean is quite cloudy for all months. Nevertheless, even mean monthly cloudiness is rather poorly quantified, perhaps only within 5-10%. As reviewed by Curry et al. (1996), there are several reasons for this unsatisfying state of affairs. While cloud fraction (the part of the celestial dome covered by cloud) is a standard variable of synoptic weather reports, spatial coverage of observations is spotty and the measurement itself is somewhat subjective. The most comprehensive surface-based data set for the central Arctic Ocean is still from the Russian NP program. Also, during the long polar night cloud observations are difficult, especially under moonless skies. An analysis by Hahn et al. (1995) of Arctic Ocean cloud amounts observed in winter under moonlit and moonless conditions suggests an underestimate of total cloudiness by about 5%.
There are a number of satellite-derived gridded data sets. For example, gridded satellite-derived estimates of cloud parameters (cloud amount, cloud top height, optical thickness, etc.) with global coverage are provided through the International Satellite Cloud Climatology Project (ISCCP) (Rossow and Schiffer, 1991). Gridded polar-specific cloud products from 1982 onwards are provided as part of the AVHRR Polar Pathfinder effort (Maslanik et al., 1997). Cloud detection in both data sets is based on a combination of visible and infrared wavelength imagery. For example, the ISCCP-C1 (daily) and C2 (monthly) data sets for the polar regions are based on AVHRR data from channels 1 (0.58-0.68 ¡m, 1 ¡xm = 10-6 m) and channels 4 (10.3-11.3 ¡xm). The Arctic Polar Pathfinder (APP-x) effort includes cloud properties along with surface radiation fluxes and other variables on a daily basis (Key and Intrieri, 2000; Key et al., 2001).
There are fundamental problems with satellite retrievals in high latitudes. In the visible wavelengths, clouds and the snow/ice surfaces have essentially the same
reflectance, making cloud discrimination very difficult. Infrared measurements are limited by the fact that temperature differences between clouds and the surface are usually small. Because of the low-level temperature inversion, cloud tops may be warmer than the surface.
Schweiger and Key (1992) show a wide discrepancy between the annual cycle of total cloud amount in the Arctic reported by Warren et al. (1988) from surface stations and from ISCCP-C2 data, although except for the central Arctic Ocean the basic spatial patterns are in general agreement. Wilson et al. (1993) suggested that some of the discrepancies in winter might be attributable to the frequent occurrence of ice crystals (diamond dust) in the atmosphere, which might be identified as low-level cirrus by the ISCCP algorithms. The ISCCP-C products have been superseded by the ISCCP-D series, which includes improvements to mitigate the problems found over polar surfaces. However, Schweiger et al. (1999) show that the ISCCP-D1 product still has significant problems and overestimates cloudiness in winter and summer compared with the NP observations. Moreover, the annual cycle of cloudiness is reversed from that observed at surface stations.
Figure 2.22 shows the mean annual cycle of cloudiness (percentage cover) over the central Arctic Ocean based on surface observations from the Comprehensive Ocean Atmosphere Data Set (COADS) through 1995. The COADS records for this region are from a combination of observations from the NP program and ship reports. No adjustments for moonlit versus moonless conditions or for ice crystal clouds are incorporated. Results are given for the percentage of the celestial dome covered by clouds of all types (total cloud cover) and by only low cloud cover. During the winter months, there is typically about 60% total cloud cover, and about 45-50% low cloud cover. During summer, corresponding values are about 80% and 70%. Hence the annual cycle is largely driven by low-level (stratus) clouds. Note the abrupt increase in central Arctic Ocean cloud cover between May and June and the abrupt decline between October and November. The results in Figure 2.22 are broadly consistent with those from other studies.
There has been considerable work addressing summer Arctic stratus. Herman and Goody (1976) envision the formation of summer stratus as an airmass modification process, whereby relatively warm and moist air of continental origin in an initially unsaturated airmass moves over the sea ice cover and condenses due to radiative and diffusive cooling to the colder surface and longwave emission to space. Once the cloud becomes optically thick enough so that cloud-top radiative cooling becomes large, turbulent kinetic energy is produced and vertical mixing occurs, producing more condensation due to adiabatic cooling. The subsequent study of Curry et al. (1988) supports this basic view. The springtime increase in cloud cover takes place prior to the onset of widespread surface melting (Barry et al., 1987), implying that evaporation from the pack ice surface is not important. The observed persistence of the stratus is viewed in terms of the sluggishness of effective dissipating processes (e.g., precipitation, radiative heating, convective heating, synoptic activity). Several mechanisms have been proposed to account for the observed multiple layering, which are reviewed by Curry et al. (1996).
As pointed out by Beesley and Moritz (1999), for the central Arctic Ocean, there is a fairly sharp rise in the vapor flux convergence between May and June (Walsh et al., 1994; Serreze et al., 1995a; Cullather et al., 2000). This increase occurs about a month earlier than the sharp spring rise in the amount of Arctic stratus, hence casting Herman and Goody's (1976) interpretation into question. Based on results from a radiative-turbulent column model, Beesley and Moritz (1999) suggest that a more dominant role on the seasonality of low-level stratus is played by the temperature-dependent formation of atmospheric ice. Briefly, at temperatures below freezing, the saturation vapor pressure over ice is lower than over liquid water, such that ice particles grow at the expense of the liquid condensate. The concentration of ice crystals is smaller than that of CCN. Hence a given mass of frozen condensate is distributed among smaller numbers of larger nuclei that grow rapidly to precipitable sizes when the environment is supersaturated with respect to ice, disfavoring the development of stratus. As discussed by Beesley and Moritz (1999), this can be viewed as a "preemptive dissipation" of stratus during the cold season, as the process can prevent the humidity of clear air from reaching saturation with respect to liquid water. This idea is supported by numerous observations of ice crystal precipitation during the winter months. During summer, when temperatures are higher, the ice-crystal scavenging processes are less effective and stratus is more likely to form and persist.
Figure 2.23 provides the corresponding COADS-derivedmean annual cycle in cloud cover for the Atlantic sector of the Arctic Ocean, taken as the region between 60° N to 65° N latitude and 40° W to 40° E longitude. Note, in comparison to the central
Figure 2.23 Mean annual cycle of cloud cover (total cloud and low cloud, in %) for the Atlantic sector of the Arctic Ocean based on COADS data through 1995 (by the authors).
Arctic Ocean, the lack of seasonality in the means of both total and low cloud cover, which hover at about 85% and 75%, respectively. This manifests the persistence for all months in this region of frequent synoptic activity (see Chapter 4) and fairly high atmospheric humidity and temperatures.
The advantage of satellite data, of course, is the regular Arctic-wide coverage. Figure 2.24 provides mean fields of cloud cover (%) for the four mid-season months based on the APP-x data set, which is considered to be improved over ISCCP-D. The fields are averaged for the years 1982 through 1999. To summarize: (1) cloud cover is rather extensive over most of the Arctic during all seasons; (2) lesser amounts are found over central Greenland, as the ice sheet extends above much of the tropospheric water vapor; (3) in accord with the surface-based COADS observations, cloud cover is particularly extensive in the Atlantic sector for all months.
Was this article helpful?