Excess nutrients in aquatic ecosystems is a major water quality problem globally. Worsening eutrophication issues are notable in cold temperate areas, with pervasive problems in many agriculturally dominated catchments. Predicting nutrient export to rivers and lakes is particularly difficult in cold agricultural environments because of challenges in modelling snow, soil, frozen ground, climate, and anthropogenic controls. Previous research has shown that the use of many popular small basin nutrient models can be problematic in cold regions due to poor representation of cold region hydrology. In this study, the Cold Regions Hydrological Modelling Platform (CRHM), a modular modelling system, which has been widely deployed across Canada and cold regions worldwide, was used to address this problem. CRHM was extended to simulate biogeochemical and transport processes for nitrogen and phosphorus through a complex of new process-based modules that represent physicochemical processes in snow, soil and freshwater. Agricultural practices such as tillage and fertilizer application, which strongly impact the availability and release of soil nutrients, can be explicitly represented in the model. A test case in an agricultural basin draining towards Lake Winnipeg shows that the model can capture the extreme hydrology and nutrient load variability of small agricultural basins at hourly time steps. It was demonstrated that fine temporal resolutions are an essential modelling requisite to capture strong concentration changes in agricultural tributaries in cold agricultural environments. Within these ephemeral and intermittent streams, on average, 30%, 31%, 20%, and 16% of the total annual load of nitrate (NO₃), ammonium (NH₄), soluble reactive phosphorus (SRP), and particulate phosphorous (partP)NO₃, NH₄, SRP and partP occurred during the episodic snowmelt freshet ($\sim$9 days, accounting for 21% of the annual flow), but shows extreme temporal variation. The new nutrient modules are critical tools for predicting nutrient export from small agricultural drainage basins in cold climates via better representation of key hydrological processes, and a temporal resolution more suited to capture dynamics of ephemeral and intermittent streams.

水生生态系统中的营养过剩是全球主要的水质问题。在寒冷的温带地区，富营养化问题日益严重，许多以农业为主的集水区普遍存在问题。由于在模拟雪、土壤、冻土、气候和人为控制方面存在挑战，因此在寒冷的农业环境中预测向河流和湖泊的养分输出尤为困难。先前的研究表明，由于寒冷地区水文的代表性较差，在寒冷地区使用许多流行的小盆地养分模型可能会出现问题。在这项研究中，寒冷地区水文建模平台 (CRHM) 是一种模块化建模系统，该系统已在加拿大和全球寒冷地区广泛部署，用于解决这个问题。CRHM 被扩展到模拟氮和磷的生物地球化学和传输过程，通过一个复杂的基于过程的模块，代表雪、土壤和淡水中的物理化学过程。农业实践，如耕作和施肥，对土壤养分的可用性和释放产生强烈影响，可以在模型中明确表示。一个流向温尼伯湖的农业流域的测试案例表明，该模型可以在每小时的时间步长内捕获小型农业流域的极端水文和养分负荷变化。事实证明，精细的时间分辨率是捕捉寒冷农业环境中农业支流中强烈浓度变化的必要建模条件。在这些短暂和间歇性的流中，平均有 30%，₃ )、铵(NH ₄ )、可溶性活性磷(SRP)和颗粒磷(partP)NO ₃、NH ₄、SRP和partP发生在间歇性融雪期间($～$9 天，占年流量的 21%），但呈现出极端的时间变化。新的养分模块是通过更好地表示关键水文过程和更适合捕捉短暂和间歇性河流动态的时间分辨率来预测寒冷气候下小型农业流域的养分输出的关键工具。