Agro-ecosystem models, such as the DNDC (DeNitrification and DeComposition) model are useful tools when assessing the sustainability of agricultural management. Accuracy in soil temperature estimations is important as it regulates many important soil biogeochemical processes that lead to greenhouse gas emissions (GHG). The objective of this study was to account for the effects of snow cover in terms of the measured snow depth (mm of water), soil texture and crop management in temperate latitudes in order to improve the surface soil temperature mechanism in DNDC and thereby improve GHG predictions. The estimation of soil temperature driven by the thermal conductivity and heat capacity of the soil was improved by considering the soil texture under frozen and unfrozen conditions along with the effects of crop canopy and snow depth. Calibration of the developed model mechanisms was conducted using data from Alfred, ON under two contrasting soil textures (sandy loam vs. clay). Independent validation assessments were conducted using soil temperatures at different depths for contrasting managements for two field sites located in Canada (Guelph, ON and Glenlea, MB). The validation results indicated high model accuracy (R-2 > 0.90, EF >= 0.90, RMSE < 3.00 degrees C) in capturing the effects of management on soil temperature. These developments in soil heat transfer mechanism improved the performance of the model in estimating N2O emissions during spring thaw and provide a foundation for future studies aimed at improving simulations in DNDC for better representations of other biogeochemical processes. Crown Copyright (C) 2017 Published by Elsevier Ltd on behalf of IAgrE. All rights reserved.

来源平台：BIOSYSTEMS ENGINEERING