databaseExplanation.txt

# Machine learning reveals key drivers of at-vessel mortality in demersal sharks

**Authors omitted during review period**

The present database was used to fit the boosted regression trees models presented in this scientific article. 


### This database contains information on:


#### (1) The relative identification of each individual studied, including:

Database code   Full name 								Units

scientificName	Scientific name of the study species 					Scyliorhinus canicula, Galeus melastomus

spCode		Code given to name each species  					Scyliorhinus canicula == Scanicula, Galeus melastomus == Gmelastomus

organismID	identification number given to each specimen 				ranging from 1 to 3079 

towN		identification number given to each tow analysed			ranging from 1 to 66		

vessel		identification number given to each trawler collaborating in the study	ranging from 1 to 8

date		date when the tow occurred						ranging from 02-12-2020 to 15-06-2022

Vessel name and fishing location was omitted as observation campaigns were conducted on board commercial trawlers and such information is confidential.

 


#### (2) Survival stage of the specimen at the time when sharks were released back to sea:

Database code   Full name				Units

mortality	Mortality stage of the specimen		0 == alive,  1 == dead

 



#### (3) Biological, environmental and fishing operation predictors considered into the modelling approach.

Database code   Full name						Units

TL		Body size						centimeters

MAT		Maturity						0 == immature , 1 == mature
	
SEX		Sex							0 == male, 1 == female

DEPTH		Tow depth						meters
	
DUR		Effective towing duration				hours
		
SPEED		Towing speed						knots

TOWMASS		Total catch biomass in the tow cod-end			kilograms
	
DECKTIME	Time exposed on deck					minutes

CLOUD		Cloud coverage						%

SEASTATE	Sea state						Douglas scale (0 to 9)

WIND		Wind force						Beaufort scale (0 to 12)

ATEMP		Atmospheric temperature					ºC

DTEMP		Change from atmospheric to sea bottom temperature	ºC