The distribution of three types of nautical craft was considered in the study. 

 These are sport (recreational) craft, fishing/shrimp boats, and ships (including all large 

 vessels not engaged in fishing or sport activities). 



Two levels of association between different animal taxa were considered: (1) 

 Incidental association resulting from sharing physical habitat such as inshore waters or 

 perching sites, and (2) direct associations where interactions of various taxa are expected 

 or where animals are responding to a shared resource such as krill, bait fish, and carrion. 



OBJECTIVE TWO: DATA MANAGEMENT 



Data Manipulation and Analysis 



The identification, development, and testing of software necessary to capture, 

 analyze, and archive field data relevant to the study are important elements of the Pilot 

 Study. The Statistical Package for the Social Sciences (SPSS), an information 

 management system with adequate flexibility for input, modification, analysis, and 

 storage of a large data source, was employed in this project. SPSS offers a powerful 

 combination of data input and data modification subroutines with potential for entry into 

 a spectrum of internal or external statistical programs. 



Data from all surveys were transcribed into a computer compatible format and 

 entered into a data bank in the DEC-20 computer system at Tulane University, New 

 Orleans, Louisiana. Transliteration was performed to standardize units of measure in the 

 metric system and to facilitate data analysis. Latitude and longitude for each 

 observation were stored as decimal equivalents of degrees in place of degrees and 

 minutes as used on charts and maps. Species and other taxonomic groups were identified 

 with a four letter alphanumeric code and a four digit numeric code which facilitated 

 grouping and sorting of observations. The altitude of the aircraft, the group speed, and 

 the water depth at the position were recorded in feet, knots, and fathoms, and 

 subsequently converted to meters, kilometers per hour, and meters, respectively. 



Population Enumerators 



The investigation of marine animal abundance is fraught with numerous problems 

 concerning the quantification of data due to the complexities of the living world. 

 Although estimates of animal abundance are often imprecise, they provide a relative 

 estimate of abundance that can answer a broad range of ecological and management 

 questions to an acceptable level of approximation. 



In the present study a combination of strip and line transect population 

 enumerators was used to analyze data. Because of the diversity of methods for 

 estimating animal abundance using line and strip transect methods and the need to 

 compare the results of varied methods, a computer program capable of calculating the 

 various estimates using the basic data input was needed. Linetran, the computer 

 program chosen for data analysis, was developed by Professor Charles E. Gates at Texas 

 A & M University. As no single estimator is best under all conditions, Linetran uses up to 

 14 different estimators to compute abundance. Linetran is very flexible and will handle 

 right angle (perpendicular) distance and radial (sighting) distance in any combination. 

 This program can provide a variety of parametric or non-parametric estimates depending 



12 



