Microbial pretreatment of lignocellulosic feedstocks is an environment friendly alternative to physiochemical pretreatment methods. A better understanding of the interactive fungal mechanisms in biological systems is essential for enhancing performance and facilitating scale-up and commercialization of this pretreatment technique. In this study, mathematical models were developed for describing cellulose and hemicellulose consumption, lignin degradation, cellulase and ligninolytic enzyme production and oxygen uptake associated with the growth of Phanerochaere chrysosporium during a 14-day shallow stationary submerged fungal pretreatment process on cotton stalks. Model parameters were estimated and validated by Statistics Toolbox in MatLab 7.1. Models yielded sufficiently accurate predictions for cellulose and hemicellulose consumption (R-2 = 0.9772 and 0.9837), lignin degradation (R-2 = 0.9879 and 0.8682) and ligninolytic enzyme production (R-2 = 0.8135 and 0.9693) under both 1-day and 3-day oxygen flushing conditions, respectively. The predictabilities for fungal growth (R-2 = 0.6397 and 0.5750) and cellulase production (R-2 = 0.0307 and 0.3046) for 1-day and 3-day oxygen flushing, respectively, and oxygen uptake (R-2 = 0.5435) for 3-day oxygen flushing were limited. (C) 2012 Elsevier Inc. All rights reserved.