ensure that spatial and temporal integrity are main- 

 tained. If feasible, the associated salinity data are 

 categorized into one of the five new seasonal salinity 

 zones: 



Salinity zone I: - 0.5 ppt. 

 Salinity zone II: 0.5 - 5 ppt. 

 Salinity zone III: 5-15 ppt. 

 Salinity zone IV: 15 - 25 ppt. 

 Salinity zone V: > 25 ppt. 



If both surface and bottom salinities are recorded in 

 the FIM data, the following guidelines for salinity 

 data selection are used to define salinity associations: 



(1) Use bottom salinity if a trawl is the sampling 

 method and water depths exceed 3 m; otherwise, use 

 depth-averaged salinity. 



(2) Use depth-averaged salinity for gill net data if 

 fished at depths not exceeding 50% of the nets height; 

 otherwise, use bottom salinity. For most other passive 

 gear types (e.g., fyke, hoop, and pound nets), use 

 bottom salinity. 



(3) Use bottom salinity for bag and beach seines. 



(4) If bottom salinity data are not available, use surface 

 salinity. 



Data transformation. Survey catch data for each 

 species are classified by species guild (Table 4, p. 8), 

 and sampling gear susceptibility. When multiple 

 sampling gear types are specified in the survey 

 metadata, discrete data sets are created for each sam- 

 pling strategy. If length-frequency counts are re- 

 corded in the data, this information is used to isolate 

 juvenile catch from adults. These data are then sepa- 

 rated into discrete gear type/life stage data sets (e.g., 

 trawl/juvenile). In the absence of length-at-capture 

 information, gear type is used to help identify which 

 life stages should be compared based on gear suscep- 

 tibility. 



An average catch per unit effort (CPUE) for each 

 estuary/species/month/salinity zone is calculated. 

 The CPUEs are then ordered by percentile to identify 

 natural statistical breaks. These percentile breaks 

 serve to parse the catch data into the five ELMR 

 relative abundance rankings: 



(1) If CPUE 

 present." 



0, then relative abundance = "not 



(2) If 1 < CPUE < 10th percentile, then relative abun- 

 dance = "rare." 



(3) If 10th percentile < CPUE < 50th percentile, then 

 relative abundance = "common." 



(4) If 50th percentile < CPUE < 90th percentile, then 

 relative abundance = "abundant." 



(5) If CPUE > 90th percentile, then relative abundance 

 = "highly abundant." 



ELMR data have been quantitatively updated for 

 Texas, Louisiana, Mississippi, Alabama, Florida in the 

 Gulf of Mexico region; North Carolina, South Caro- 

 lina, Georgia in the Southeast region; and Massachu- 

 setts in the Mid-Atlantic and North Atlantic regions. 

 As examples, the application of this methodology to 

 ELMR data for the States of Texas and Massachusetts 

 are described here. 



Gulf of Mexico: Texas case example. In 1997, the Gulf 

 Wide Information System (GW1S) project was initi- 

 ated by the U.S. Department of the Interior's Minerals 

 Management Service (MMS), in cooperation with 

 NOAA, the Gulf of Mexico states (FL, AL, MS, LA, TX) 

 and others (Christensen and Monaco 1998). The ob- 

 jective of the GWIS project is to develop an authorita- 

 tive data base, as mandated by the Oil Spill Pollution 

 Act of 1990, for oil spill contingency planning in the 

 Gulf of Mexico region (NOAA 1997a). NOAA's role 

 and contribution to GWIS included: 



(1 ) Updating and digitally integrating NOAA's ELMR 

 data into the GWIS data base. 



(2) Updating the data for selected coastal and marine 

 fishes in the Gulf of Mexico. 



To complete NOAA's contribution to the GWIS project 

 for Texas estuarine waters, the ELMR Program ac- 

 quired fishery-independent monitoring (FIM) data 

 sets from the Texas Parks and Wildlife Department 

 (TPWD). Estuarine fishery-independent sampling 

 methods include trawl, bag seine, beach seine, and gill 

 net. These data sets were used to revise and update 

 the existing Texas ELMR data to fit the new spatial 

 framework, according to the general procedure de- 

 scribed above. Specific elements of this procedure 

 included: 



(1) Developing a seasonal, five-salinity-zone spatial 

 framework for Texas estuaries. Figure 19 depicts the 

 salinity zones for Galveston Bay during the low, in- 

 creasing, high, and decreasing salinity seasons. 



(2) Grouping fisheries-independent data according to 

 the revised salinity zones for all Texas estuarine wa- 

 ters. 



48 



