GCOS GOOS WCRP/OOPC IX/3 

page 52 

research  (Diaz  et  al.,  2002).  CLIMAR II  was  organized  partly  in  the  light  of  these 

recommendations, and this report summarizes our progress in fulfilling them so far. 

Proceedings  

Like  the  Boulder  workshop,  CLIMAR II  was  divided  into  three  main  sessions.  In 

Session I, on cross cutting issues, presentations included databases, metadata, quality control 

(QC),  homogeneity,  biases,  statistical  analysis  techniques,  reanalyses,  and  user  products. 

Presentations  in  Session  II  concentrated  on  sea  level  pressure  (SLP),  wind  and  waves;  and 

those in Session III dealt with marine temperatures and sea ice. Estimation of uncertainty was 

a common theme in all the sessions. Many of the presentations in each session were based on 

the International Comprehensive Ocean Atmosphere Data Set (ICOADS) named I COADS at 

the  Boulder  workshop  but  now  re named  to  ease  citation  and  web  paging.  ICOADS  is  an 

upgrade of COADS, created by blending  COADS with the Met Office's  Marine Data Bank 

and millions of newly digitized logbook records, with careful elimination of duplicates (Diaz 

et al., 2002). A final summary session reviewed progress since the 2002 Boulder workshop, 

and discussed future activities. 

Comparison  with  the  Boulder  recommendations  (Diaz  et  al.,  2002)  revealed  good 

progress on :  

1)  Increased coverage of data, especially for data sparse times and places.  

2)  Understanding and reduction of biases, e.g. in in  situ marine air temperature (MAT) 

and in satellite based sea surface temperature (SST) data.  

3)  Specification of uncertainties and their inclusion in analyses. 

4)  Comparison of QC techniques.  

5)  Availability of additional land station SLP data to support marine analysis. 

6)  Development  of  techniques  for  reanalysis  of  atmospheric  circulation  in  the  pre 

radiosonde era. 

There has also been some progress in: 

1)  Approval by the WMO Executive Council of a format for metadata from Ocean Data 

Acquisition Systems (ODAS) including buoys.  

2)  Analysis of diurnal cycles in SST using geostationary satellite data.  

3)  Availability of satellite based temperatures for inland seas and large lakes.  

4)  Research to improve the specification of SST in marginal ice zones. 

5)  Assembly of the first version of a blended sea ice dataset for the Arctic for 1950 98 by 

the JCOMM Expert Team on Sea Ice. 

6)  Improvement  of  cloud clearing  techniques  for  satellite based  SST.  For  example,  the 

SSTs from the Tropical Rainfall Measuring Mission (TRMM) have yielded substantial 

improvements in cloudy and poorly sampled tropical regions. 

7)  Assessment of biases in the Maury SLP data. 

Furthermore, we note the substantial international effort to prepare recommendations 

for enhancements to GCOS (GCOS, 2003).