Dataset for paper "Machine learning reveals key drivers of at-vessel mortality in demersal sharks"

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