Satellite retrievals offer systematic assessments of global radiation and other surface radiative fluxes with complete Arctic coverage. There are two primary data sets - APP-x (25-km grids, covering 1982-99) and ISCCP-D (280-km grids, covering 1985-93). Both were introduced in Chapter 2. The APP-x product is based on data from AVHRR. Briefly, surface (skin) temperature is calculated with a split-window infrared algorithm. Surface albedo retrieval for clear and cloudy skies employs corrections for anisotropic (direction-dependent) reflectance and atmospheric effects. Cloud detection is done with a variety of spectral and temporal tests optimized for high-latitude conditions. Cloud particle phase (ice versus water) is discriminated using near infrared reflectances and infrared brightness temperatures. Cloud optical depth and cloud particle effective radius retrievals use absorbing and non-absorbing wavelengths. Cloud temperature is calculated from the infrared window brightness temperature. The retrieved cloud and surface parameters are then used as input to a radiative transfer model to calculate radiative fluxes. Details are provided by Key and Intrieri (2000) and Key et al., (2001). The global ISCCP-D data set is based on a number of different satellite platforms. Improvements over the precursor ISCCP-C product result primarily from improvements in cloud detection, particularly through addition of the AVHRR 3.7-^m channel. The radiative model has also been enhanced to include both ice and liquidphase clouds. Rossow et al. (1996) and Rossow and Duenas (2004) provide additional information. Surface radiation fluxes are derived based on modifications to the approach described by Schweiger and Key (1994).
Surface fluxes from APP-x are considered to be better than those derived from ISCCP-D. Some of the problems with ISCCP-D are outlined by Schweiger et al. (1999). However, the orbital characteristics of the AVHRR satellites mean that APP-x product provides instantaneous fluxes only twice daily. By comparison, ISCCP-D products are available three-hourly. Our analysis focuses on monthly means, which for ease of interpretation should include averaging of the diurnal cycle. Consequently, we make primary use of monthly averages from ISCCP-D data, which are based on the period 1985 through 1993. For mean surface albedo (discussed later), resolving
the diurnal cycle is not as important and the APP-x product (with its higher spatial resolution) is used. New products coming on line, such as based on MODIS, promise more accurate fields.
Maps of the mean monthly global radiation flux for March through October based on ISCCP-D are provided in Figure 5.1. The basic patterns compare reasonably well with those from Serreze et al. (1998), based on measurements from land stations, the Russian NP program, United States drifting stations and (over the North Atlantic and coastal Greenland) empirically derived estimates from Russian studies. The Serreze et al. maps were compiled by subtracting from each climatological monthly station mean the monthly value of Gcir at that station (based on results from a radiative transfer model), interpolating the adjusted values to a regular grid, and then adding back to each grid point the corresponding value of Gclr. Differences between the two data sets are expected because of the different time periods of data used, the sparse amount of direct information over the central Arctic and northern North Atlantic, and errors in satellite cloud retrievals.
The ISCCP-D global radiation fields for March, September and October exhibit primarily zonal patterns (i.e., the isolines are roughly parallel to latitude circles). This illustrates the dominant effects of the strong latitudinal decrease in Gclr for these months. April through August are dominated by a strong asymmetric pattern. Greenland shows a pronounced peak over its central portions from April through August. This manifests the high elevation - the atmospheric path length is smaller and the central portions of the ice sheet are often relatively cloud free. Fluxes decrease sharply toward the Atlantic side of the Arctic, largely because of increasing cloud fraction (Figure 2.24) as well as greater cloud optical thickness.
There is also some tendency in summer for mean monthly fluxes to increase with latitude, most apparent in June, the month of maximum radiation. For this month, Gclr exhibits a marked increase of about 20 W m-2, from 65° to 90° N. The high values over the central Arctic Ocean also manifest the effects of the high sea ice surface albedo in promoting multiple scattering between the surface and clouds. For June, ISCCP-D fluxes over the high-elevation Greenland Ice Sheet peak at more than 360 W m-2, and values in excess of 300 W m-2 are found near the Pole. For the Arctic Ocean, Serreze et al. (1998) place the June maximum (about 300 W m-2) as extending from near the Pole into the Beaufort Sea. June fluxes of 200-220 W m-2 characterize the Atlantic side in the Norwegian and Barents seas, where cloud cover is both thick and extensive.
Was this article helpful?