254
WILLIAM B. LARGE
Both Beranger et al. (1999) and WGASF (2000) compare zonal average net shortwave radiation flux from a variety of sources. Two of these
(NCEP and SOC) are shown in Fig. 6 along with the ISCCP-FD product (thin solid). In the tropics, the latter is about 10 − 20W/m 2 greater
than the Langley Research Center (Darnell et al., 1996) satellite product (not shown). There is much better agreement at higher latitudes.
Although the two dataset agree to within their uncertainties, differences
are significant in the tropics, where the two observational datasets, SOC
(Fig. 6) and COADS (not shown) both fall about midway between the
two satellite products.
Figure 7. Comparison of zonally averaged precipitation into the ocean from a variety
of climatologies.
5.3
Precipitation
The precipitation sampling problem is much less severe in satellite
estimates, but the lack of accurate in situ observations makes the development and validation of algorithms difficult. Therefore, satellite
precipitation data sets suitable for GODAE must be regarded as highly
uncertain. Freshwater fluxes change sea surface salinity, but because
these fluxes do not depend on the salinity there is no direct negative
WILLIAM B. LARGE
Both Beranger et al. (1999) and WGASF (2000) compare zonal average net shortwave radiation flux from a variety of sources. Two of these
(NCEP and SOC) are shown in Fig. 6 along with the ISCCP-FD product (thin solid). In the tropics, the latter is about 10 − 20W/m 2 greater
than the Langley Research Center (Darnell et al., 1996) satellite product (not shown). There is much better agreement at higher latitudes.
Although the two dataset agree to within their uncertainties, differences
are significant in the tropics, where the two observational datasets, SOC
(Fig. 6) and COADS (not shown) both fall about midway between the
two satellite products.
Figure 7. Comparison of zonally averaged precipitation into the ocean from a variety
of climatologies.
5.3
Precipitation
The precipitation sampling problem is much less severe in satellite
estimates, but the lack of accurate in situ observations makes the development and validation of algorithms difficult. Therefore, satellite
precipitation data sets suitable for GODAE must be regarded as highly
uncertain. Freshwater fluxes change sea surface salinity, but because
these fluxes do not depend on the salinity there is no direct negative