Denitrification is a key but poorly quantified component of the N cycle. Because it is difficult to measure the gaseous (NO_x, N₂O, N₂) and soluble (NO₃) components of denitrification with sufficient intensity, models of varying scope and complexity have been developed and applied to estimate how vegetation cover, land management and environmental factors such as soil type and weather interact to control these variables. In this paper we assess the strengths and limitations of different modeling approaches, highlight major uncertainties, and suggest how different observational methods and process-based understanding can be combined to better quantify N cycling. Representation of how biogeochemical (e.g. org. C., pH) and physical (e.g. soil structure) factors influence denitrification rates and product ratios combined with ensemble approaches may increase accuracy without requiring additional site level model inputs.

反硝化是N循环的关键但量化程度不高的组成部分。因为很难测量气态（NO _x，N ₂ O，N ₂）和可溶态（NO ₃）具有足够强度的反硝化成分，已开发了范围和复杂程度不同的模型，并将其用于估算植被覆盖率，土地管理和环境因素（例如土壤类型和天气）如何相互作用以控制这些变量。在本文中，我们评估了不同建模方法的优势和局限性，强调了主要的不确定性，并提出了如何将不同的观测方法和基于过程的理解相结合以更好地量化N循环的方法。关于生物地球化学（例如有机酸，pH）和物理（例如土壤结构）因素如何影响反硝化率和产物比率与整体方法相结合的表示，可以提高准确性，而无需额外的场地水平模型输入。