Relationship Between Thermal Environment and Welfare Indicators of Laying Hens: A Canonical Correlation Approach

TL;DR

This study explores how thermal environments affect laying hens' welfare using statistical methods to better understand complex interactions.

Contribution

The study introduces canonical correlation analysis as a more effective method for evaluating complex thermal and welfare relationships in laying hens.

Findings

Canonical correlation analysis identified more significant relationships than Pearson correlation.

The highest correlation was between thermoregulatory responses and the thermal environment (rc = 0.7449).

The method captures interactions between climatic variables and hens' responses effectively.

Abstract

Adaptation of laying hens to heat stress involves variables and multiple mechanisms, which are complex and interconnected by linear responses and interactions. Here, we compare the Pearson correlation method and canonical correlation to evaluate relationships between thermoregulatory, behavioral, and productive responses and thermal environment variables in laying hens reared in a semi‐arid region of Brazil. A total of 270 lightweight Hy‐Line White laying hens, 58 weeks old, with body weight of 1.60 ± 0.092 kg and egg production of 77.30% ± 3.62% were used. Simple Pearson correlation analysis showed fewer significant relationships than those identified by canonical correlation analysis. The results showed low to moderate canonical correlations (0.2576 ≤ rc ≤ 0.7449) between sets of indicators. Relationships between the thermal environment, thermoregulatory responses, and productive responses were significant (p ≤ 0.05), with the pair thermoregulatory responses × thermal environment presenting the highest correlation (rc = 0.7449; rc2 = 0.5548). Canonical correlation analysis is recommended to assess the behavior of complex relationships in laying hens. This multivariate approach provides a comprehensive understanding of linear and interactive relationships and captures interactions between climatic variables and thermoregulatory, behavioral, and productive responses.

Figures