This is a draft schedule. Presentation dates, times and locations may be subject to change.

620
Estimating Gas Volume from Headspace Pressure in a Batch Culture System

Tuesday, July 11, 2017
Exhibit Hall (Baltimore Convention Center)
A. Romero-Pérez, Lethbridge Research and Development Centre, Agriculture and Agri-Food Canada, Lethbridge, AB, Canada
Karen A. Beauchemin, Lethbridge Research and Development Centre, Agriculture and Agri-Food Canada, Lethbridge, AB, Canada
Estimation of headspace gas volume (GV) from headspace pressure (GP) in in vitro batch culture experiments conducted at our lab is currently done using equations obtained overseas (Mauricio et al., 1999). Thus, the objective of the present work was to generate an equation to estimate GV from GP based on data obtained under our own experimental conditions. Two independent batch culture runs were conducted on different days. Twelve different feed ingredients including forages, grains and by-products were utilized. Headspace GP and GV were measured after 3, 6, 9, 12, 18, 24, 48, 60 and 72 h using a pressure transducer and graduated plastic syringes (20 or 50 mL) connected to a three-way stopcock respectively. A total of 1811 pairs of headspace pressure-volumes were obtained. One half of the dataset was used to generate linear, quadratic and cubic equations with intercepts set to zero or not (six equations in total). Second half was used to evaluate estimated GV from obtained equations. The best equation (equation 1) was then compared against equations reported by Mauricio et al. (1999; Equation 2) and Lopez et al. (2007; equation 3). Boyle’s law adapted to our lab conditions was also evaluated (Equation 4; 63 mL headspace volume; atmospheric pressure 13.15 PSI). Equations were evaluated using the r2 between the observed and predicted values, root mean square prediction error (RMSPE), concordance correlation coefficient (CCC) and error due to the disturbance or random variation (ED). A quadratic equation was the most precise and accurate among the 6 equations obtained (GV = -0.2232 + 4.8021GP + 0.0422GP2; r2 = 0.995; RMSPE = 0.64 mL; ED = 99.9%; CCC = 0.99). When compared with equations 2, 3 and 4, equation 1 ranked first followed by equation 4 (GV = 4.79GP; r2 = 0.994; RMSPE = 1.10 mL; ED = 39.3%; CCC = 0.99), equation 3 (GV = 5.385GP; r2 = 0.994; RMSPE = 1.85 mL; ED = 13.8%; CCC = 0.98) and equation 2 (GV = 0.18 + 3.697GP + 0.0824GP2; r2 = 0.994; RMSPE = 3.33 mL; ED = 4.23%; CCC = 0.93). In conclusion, the quadratic equation obtained in the present study estimates GV precisely and accurately, and can be used in further experiments conducted under similar conditions.