In developing data management support for each estuary program, 

 existing systems are evaluated. The state agency responsible for 

 implementing the estuary program's findings should use an existing 

 data management system if possible. Evaluation criteria include 

 the following: 



• Ease of access and use; 



• Use of relatively standard hardware and software; 



• Capability of storing diverse kinds of information, such as 

 physical, chemical and biological data; land use statistics; 

 and point and nonpoint source records; 



• Data analysis features, including analyzing statistics and 

 generating presentations like tables, graphs, charts, and 

 maps; 



• Flexibility to adapt to changing needs; and 



• Cost of usage and maintenance. 



After priority data sets have been collected and entered into the 

 data management system, they can be accessed as needed. 

 Before data sets are analyzed, however, they must be screened 

 (Figure 3.1). 



The screening procedure is designed to review the quality of the 

 data and to identify unusual values or missing information. For 

 example, nonexistent observations from important time periods or 

 sampling stations should be identified. In addition, unusually high 

 or low data values can be isolated for closer inspection. 



Screening can be conducted by combining computerized checks 

 with experts' technical reviews. The results of the screening proce- 

 dure are included in each data set. They help determine which data 

 sets are appropriate for the various evaluations conducted during 

 data analysis. Screening also helps identify insufficient or missing 

 information that may need to be addressed later in the charac- 

 terization process. 



Estuary segmentation — partitioning an estuary into a series of 

 spatial units or segments — is a useful analytical tool. It permits 

 consolidating an extensive amount of environmental information 

 into representative data elements when certain conditions, such as 

 water temperature and salinity, are relatively homogeneous within 

 a segment. During data analysis, historical information collected for 

 each segment is combined to represent the average set of condi- 

 tions encountered in the segment. In this manner, a data base 

 consisting of hundreds of stations can be reduced to a description 

 of conditions based on a relatively smaller number of segments 

 (Figure 3.3). Besides facilitating data integration, segmentation 

 also allows researchers to examine data based on station locations 

 of uncertain origin. This is particularly useful because the lack of 

 information on exact station locations is a limitation frequently 

 encountered with historical data sets. 



Screening Priority 

 Data Sets 



Estuary Segmentation 



29 



