| Surface Skin
Temperature (SKTEMP) |
 |
||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
|
 |
||||||||||
 |
|||||||||||
|
Scatterplot and time series of surface skin temperature from SHEBA meteorological observations and from corresponding Path-P retrievals. Time series data are plotted as 1-day running means. |
|||||||||||
The Path-P product provides a Surface Skin Temperature. It is retrieved by combination of methods (c.f. Francis, 1994). Here we compare it to 2-m surface temperatures measured at the SHEBA camp. Surface temperature retrievals are done using a mixture of methods depending on whether the retrieval box was classified as clear, partly cloudy or overcast. For clear and partly cloudy retrieval boxes, the surface temperature inversion is done using varying combinations of HIRS channels 8, 18 and 19 using only the clear HIRS pixels. Completely overcast boxes (all HIRS pixels have some cloud) are "cloud cleared" and surface temperature retrieval is performed using the "cloud cleared" radiances. This method is only applied in cases when the effective cloud fraction for the respective box is less than 90%. The method based on MSU-2 correlations described in Francis (1994) is no longer in use. Given this mix of retrieval methods a clear definition of whether the Surface Skin Temperature represents a "Clear Sky" value or an "All Sky" value is difficult. In the Arctic, where surface temperature can vary drastically between clear and cloudy this lack of a definition may pose a problem. The situation is further complicated by the fact that the response in surface temperature due to a change in cloud cover varies depending on other conditions such as wind speed etc. Since it takes longer for the surface to warm after a cloud has moved in, than it takes to cool when a cloud disappears (Walsh and Chapman, 1998) the time when "hole in the clouds" is seen by a satellite remote sensing system will also influence the retrieved temperature. As can be seen from the figures and comparison statistics below, the Path-P SKTEMP variable corresponds closely to the measured 2-m temperature. The coldest temperatures appear to be overestimated slightly.
| Surface Skin
Temperature (SKTEMP) |
|
||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
|
|
||||||||||
|
|||||||||||
|
Scatterplot and time series of surface skin temperature from SHEBA meteorological observations and from corresponding Path-P retrievals. Time series data are plotted as 1-day running means. |
|||||||||||