Animal Feed Science and Technology 174 (2012) 190–200 Contents lists available at SciVerse ScienceDirect Animal Feed Science and Technology journal homepage: www.elsevier.com/locate/anifeedsci Development of an in vitro method for determination of methane production kinetics using a fully automated in vitro gas system—A modelling approach M. Ramin ∗ , P. Huhtanen Department of Agricultural Research for Northern Sweden, Swedish University of Agricultural Sciences, SE-901 83 Umeå, Sweden a r t i c l e i n f o Article history: Received 10 July 2011 Received in revised form 5 March 2012 Accepted 12 March 2012 Keywords: Methane kinetics Gas production Rumen model Sample size VFA a b s t r a c t The objective of the present study was to estimate methane production using the kinetic parameters from an automated in vitro gas production (GP) system in a mechanistic, dynamic rumen model. Four sample sizes [300 mg (5 g/l), 600 mg (10 g/l), 900 mg (15 g/l) and 1200 mg (20 g/l)] of timothy hay were incubated in 60 ml buffered rumen fluid in an automated in vitro system to determine methane and total gas production. A logarithmic model (R 2 > 0.99) was fitted to methane data to estimate methane concentrations at time intervals of 0.2 h. The first-order gas production rates were not different (P=0.18) with val- ues of 0.072, 0.061, 0.061 and 0.059/h for the sample size of 300, 600, 900 and 1200 mg, respectively. The methane production rates were 0.052, 0.046, 0.046 and 0.045/h, respec- tively. Predicted methane production decreased linearly (PLIN < 0.01) as the sample size increased from 300 (36.9 ml/g dry matter, DM) to 1200 mg (28.2 ml/g DM). After 48 h of incu- bation total volatile fatty acids (VFA) production decreased (PLIN < 0.01) as the sample size increased from 300 to 1200 mg (4.41 mmol/g DM and 3.82 mmol/g DM, respectively). Neu- tral detergent fibre digestibility (aNDFomD), apparent organic matter digestibility (AOMD) and true organic matter digestibility (TOMD) decreased (PLIN = 0.01) as the sample size increased (0.479, 0.433 and 0.681 for 300 mg, and 0.369, 0.379 and 0.614 for 1200 mg of sam- ple size, respectively). Actual methane production (24 and 48 h of incubation) was strongly correlated (R 2 = 0.97) with the methane production predicted from VFA stoichiometry (VFA measured at 24 and 48 h of incubation). It is concluded that in vitro GP measurements can be successfully used to estimate kinetic parameters of methane production and consequently to predict methane production. It seems that sample size did not affect the first-order pro- duction rate of methane and therefore, it is possible to use greater amounts of substrate in the in vitro GP system up to 1000 mg (16.6 g/l). © 2012 Elsevier B.V. All rights reserved. 1. Introduction Methane (CH 4 ) is the second most significant contributor to the ‘greenhouse’ effect by trapping 20 times more heat than carbon dioxide (CO 2 ) (Yan et al., 2010). Ruminants contribute to approximately one quarter of all anthropogenic methane Abbreviations: aNDFom, neutral detergent fibre assayed with a heat stable amylase and expressed exclusive of residual ash; aNDFomD, aNDFom digestibility; aNDSom, neutral detergent soluble excluding ash; AOMD, apparent organic matter digestibility; CP, crude protein; DM, dry matter; GP, gas production; HS, headspace; MRT, mean rumen retention time; NDF, neutral detergent fibre; OM, organic matter; OMF, OM fermented; pdNDFD, potential digestible neutral detergent fibre digestibility; TDOM, truly digested organic matter; TOMD, true organic matter digestibility; VFA, volatile fatty acid. ∗ Corresponding author. Tel.: +46 90 786 87 20; fax: +46 90 786 81 62. E-mail address: mohammad.ramin@slu.se (M. Ramin). 0377-8401/$ – see front matter © 2012 Elsevier B.V. All rights reserved. doi:10.1016/j.anifeedsci.2012.03.008