TSREADME.TXT 6 March 2002 Surface Temperatures from the SHEBA Atmospheric Surface Flux Group (ASFG) sites. Compiled by Ed Andreas, USA CRREL, eandreas@crrel.usace.army.mil in collaboration with the ASFG, Chris Fairall, NOAA ETL, Chris.Fairall@noaa.gov, Peter Guest, NPS, pguest@nps.navy.mil, Ola Persson, CIRES/CU, Ola.Persson@noaa.gov. The SHEBA ASFG maintained five sites in and around the SHEBA camp for the duration of the experiment. The sites included in this surface temperature data set are The main ASFG Tower site, in file ASFG_Ts.txt The PAM site at Atlanta, in file Atla_Ts.txt The PAM site at Baltimore, in file Balt_Ts.txt The PAM site at Florida, in file Flor_Ts.txt Our fifth site was another PAM station that moved from Cleveland to Seattle to Maui, so there is not a long, uninterrupted record, as there is from these other four sites. We thus do not include it here. Our ASFG web site, http://www.weather.nps.navy.mil/~psguest/sheba/, gives an overview of our SHEBA research and contains many pictures of our instruments. Our NCAR collaborators designed and maintained the PAM (portable automated mesonet) stations. Their web site, http://www.atd.ucar.edu/rtf/projects/sheba/, describes these instruments, the features of the various sites, and the upgrades throughout the experiment to mitigate problems with dome icing. Finally, the main SHEBA web site, http://sheba.apl.washington.edu/, contains a broader look at the experiment. Each of the four files in this set is a tab-delimited text file that gives the Julian day, the latitude, the longitude, and the surface temperature. Julian Day--Calculated from 1 January 1997, where 1 January is Julian day 1. That is, 12:01 a.m. on 1 January 1997, would be denoted Julian day 1.00069. Latitude--In degrees and fractional degrees Longitude--In degrees and fractional degrees. The negative sign indicates the Western Hemisphere. Ts--Surface temperature in degrees Celsius. Almost all of the surface temperatures are based on radiation measurements made with down-looking and up-looking Eppley PIR pyrgeometers mounted a few feet above the surface. The radiometers at the ASFG tower site had efficient ventilation from the beginning of the experiment and thus suffered least from icing on the domes. The radiometers at the PAM sites had ineffective ventilation early in the experiment and thus provided spotty records of radiation values and the surface temperature computed from these. Finally, by March or April, depending on the site, we had installed effective heating and ventilation on all the PAM stations, and the radiation data became more reliable. Again, the UCAR web site describes these improvements. All sites measured the incoming (QLd) and outgoing (QLu) longwave radiation. From these, we calculated the surface temperature (Ts) from the relation QLu = eps*sigma*(Ts^4) + (1 - eps)*QLd , where eps is the surface emissivity, sigma (= 5.67051E-8 W/(m^2 K^4) is the Stefan-Boltzmann constant, and here Ts is the absolute temperature. Our calculations are based on an emissivity, eps, of 0.99 year-round. In the vicinity of our main tower, the ASFG also maintained several other instruments that provided estimates of the surface temperature. These were 1) a Barnes precision infrared thermometer; 2) the dew-point temperature read by a General Eastern cooled-mirror, dew-point hygrometer mounted 8-10 centimeters above the surface; 3) the near-surface air temperature read by a platinum resistance thermometer in this General Eastern sensor. 4) Our profile tower measured the potential temperature at five levels, nominally, between 2 meters and 18 meters. In neutral stratification, this temperature profile was vertical, so the average of these five levels is a good estimate of the surface temperature. From these four auxiliary measurements, we developed an alternative time series of surface temperature. When the radiation surface temperature was not available, we substituted the best available estimate from this alternative series. In particular, from 24 May though 28 June 1998 (Julian days 509-544, inclusive), we identified a loose connection in one of the Eppleys near our tower that caused problems with the longwave radiation. We used the alternative series as our estimate of surface temperature for this whole period. All sites summarized here had ice and snow as the field of view for the radiometers. That is, none of the tabulated surface temperatures are for open water. In the summer, though, some of the down-looking radiometers may have had melt ponds in their field of view, as well as bare ice. The ASFG site, in particular, had a melt pond in view of the down-looking radiometer for part of the summer. Consequently, although some of the reported surface temperatures are above 0 deg C, it would be reasonable to take 0 deg C as the upper limit of reasonable surface temperatures.