Adaptation and Validation of the CO2 Balance Method for Ventilation Rate Estimation in Laying Hen Houses

Highlights Adapted CO2 balance improves ventilation rate prediction in poultry housing systems. Respiration quotient set to 0.9 (day) and 0.85 (night) enhances model accuracy. Adjusting the CO2 production rate (+18% day, +8% night) boosts model performance. Model accuracy declines with inlet-to-exhaust CO2 differences below 150 ppm. ABSTRACT. Assessing the environmental impact of poultry farms requires accurately determining ventilation rates (VR) while minimizing resource use. Accurate VR estimates are crucial for precisely evaluating emissions, making VR a vital factor in enhancing the sustainability of poultry production. This study aims to adapt and validate the carbon dioxide (CO 2 ) balance method for VR prediction in laying hen houses. The adaptation was based on the CO 2 balance method from the International Commission of Agricultural and Biosystems Engineering (CIGR), which considers variables such as the CO 2 production rate of hens (PRCO 2 ), animal activity (AA), and the respiratory quotient of laying hens (RQ). Data collected from an experiment involving laying hens in conventional cages (CC) under laboratory conditions were used to adapt the VR prediction model. The adapted model was validated under both laboratory and commercial conditions in three housing systems: conventional cages (CC), enriched cages (EC), and cage-free systems (CF). At the laboratory level, validation was performed using data from two studies conducted in the same controlled laboratory conditions but with different housing systems (the first in CF and the second in the three systems simultaneously). At the commercial level, validation was performed using data collected from 30 commercial farms in Quebec, Canada. The resulting adaptations included setting AA to 1, RQ to 0.9 during the day and 0.85 at night, and PRCO 2 to 18% during the day and 8% at night. The adjusted model demonstrated an R 2 of 0.63 for VR prediction when the difference between exhaled and inlet CO 2 was less than 150 ppm. At the experimental level, all evaluated housing systems showed an R 2 greater than 0.63 and an average RMSE of 0.35 m 3 h -1 hen -1 . At the commercial level, housing systems exhibited an average predicted R 2 of 0.71 and an RMSE of 1.68 m 3 h -1 hen -1 . The adapted CO 2 balance method presented good predictive values across laboratory and commercial experiments. Keywords: Airflow, Animal activity, Egg production, Gas production, Poultry building, Respiratory quotient.