Open Access

Table 5

Factors considered during the boruta regression tree analysis of larval crashes and mortality. Significant predictors as identified by the boruta algorithm are listed in order of importance from top to bottom. Terms retained as significant in the general linear model. *** indicates statistical significance at p<0.001 and ** at p<0.01 in the associated logit linear model. Full results are provided in Table S2.

Possible predictors of hatchery crashes Significant predictors of:


Broodstock related factors Larvae related factors Hatchery crashes Mortality in surviving batches
Hatchery Frequency larval cleaning (days) Temperature larval tank Broodstock density***
Year of production Temperature larval tank (°C) Broodstock origin Temperature conditioning [endpoint]***
Broodstock origin Salinity (ppt) larvae Broodstock density Hatchery
Broodstock release number Larval density at stocking (ind/ml) Week number larval release Frequency broodstock cleaning
Quarantine applied Larval feed delivery frequency Broodstock release number Salinity (ppt) larvae***
Frequency broodstock cleaning Proportion larval feed diatom Duration of conditioning Conditioning salinity (ppt)
Broodstock density at larval release Larval setup Larval feed delivery frequency Broodstock origin
In/Out of season Grading undertaken Proportion broodstock feed diatom Algal paste used in broodstock feeding**
Duration of conditioning (days) Week number larval release Algal paste used in broodstock feeding Week number larval release
Conditioning Photoperiod (min light per day) Number larvae in release (x1000) Temperature conditioning [endpoint] Frequency larval cleaning
Temperature conditioning (°C) [endpoint] In/Out of season Temperature larval tank (°C)
Conditioning Salinity (ppt) Duration collection to larval release (days)
Use air pumps broodstock Quarantine applied
Algal paste used in broodstock feeding
Feed delivery frequency broodstock
Proportion broodstock feed diatom
Duration collection to larval release (days)

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.