Development of a Model-based Phosphorus Management Tool for Manure Applications in Subsurface Drained Field

作为畜牧业的副产品，粪便是作物养分(如磷、氮和有机物)的宝贵来源。加拿大地下排水农田的粪肥生产和处置的三个主要集群是安大略省西南部、魁北克省东南部和安大略省南部。然而，农业土壤和以无机肥料或粪便形式施加在其上的养分是固定水体的主要非点源污染物。这种水体的富营养化与地下排水对磷(P)和氮(N)的富集性有关。野外研究表明，淋溶到这些水体中的磷大部分是沿地下排水径流淋失的。 通过大量田间试验，研究了在地下排水农田中施有机肥条件下磷的去向和迁移。结合田间试验，基于模型的工具可以帮助理解土壤-植物系统中P和N循环的科学，并促进养分管理，以保护水环境，同时保持作物产量。然而，自20世纪80年代中期以来，大多数模式中的P程序几乎没有更新；因此，需要改进P输送的模拟。具体地说，这些模型中的许多都假设施入的肥料立即与表土混合。之所以进行这种简化，是因为在大多数情况下，这些模型没有界定足够数量的不同有机物质池，而且只模拟了数量非常有限的实地管理做法。大多数情况下，地表径流的计算采用美国土壤保持局(SCS)曲线法，泥沙流失量的估算采用修正的通用土壤流失方程。此外，大多数这类模式既没有地下排水组件，也没有任何确定施入肥料的能力。 本计划的目标是开发一个基于根区水质模型(RZWQM2)的磷管理工具，并用它来模拟田间尺度上磷的去向和转化。这一新的基于模型的工具RZWQM2-P将在加拿大作为有机肥下磷管理的决策支持工具使用。该模型的新特点包括：(1)使用Green-Ampt模型和可变源区来预测径流；(2)使用WEPP(水土流失预测项目)中采用的机械子程序模拟泥沙的剥离、沉积和迁移过程；(3)根据不同的施肥法(地表施肥法或注入法)，将粪便中的磷分配到土壤残留物和土壤腐殖质库中；(4)磷的吸附/解吸将由NICA(非理想竞争吸附)模型计算，该模型使用Fe/Al/Ca含量来预测溶解的P的吸附/解吸；将考虑优先流模拟附着在土壤纳米颗粒上的颗粒磷从土壤表面到排水沟的传输，以及(V)作物磷吸收和胁迫的子程序将从农业技术转移决策支持系统(Decision Support System For Agrotech Transfer)中采用。 新的基于模型的磷管理工具RZWQM2-P将与在加拿大地下排水的农田上测量的磷水平进行测试。已在几个地区测量了地表径流和地下排水中的磷，并可用于校准和验证新的管理工具。我的团队在麦吉尔大学麦克唐纳农场进行了一项正在进行的关于家禽和牛粪下磷迁移的蒸渗仪研究，数据可以用来对新的磷模型进行参数化。这种新的管理模式可以作为有机肥条件下的水质和作物产量管理的决策支持工具。

A by-product of the livestock industry, manure is a valuable source of crop nutrients (e.g., phosphorus, nitrogen, and organic matter). The three major clusters of manure production and disposal on subsurface drained cropland in Canada are south-western Ontario, south-eastern Quebec, and southern Ontario. However, agricultural soils and the nutrients applied thereon, in the form of either inorganic fertilizers or manures, are the major non-point source pollutants of standing water bodies. Eutrophication of such water bodies has been linked to phosphorus (P) and nitrogen (N) enrichment by subsurface drainage. Field studies have shown that the majority of P leached to these water bodies followed subsurface drainage flow. A number of field experiments have been conducted to investigate the fate and transport of P under manure application in subsurface-drained croplands. In combination with field experiments, model-based tools can help to understand the science of P and N cycles in soil-plant systems and facilitate the management of nutrients to protect aquatic environments while maintaining crop yield. However, P routines in most models have seen little updating since the mid 1980s; therefore, the need arises to improve the simulation of P transport. In particular, a number of these models assume applied manure to be immediately mixed with top soil. This simplification was made because in most cases the models had not defined a sufficient number of different organic matter pools and only a very limited number of field management practices were simulated. In most cases, the SCS (Soil Conservation Service, USA) curve method was used to compute surface runoff and the Revised Universal Soil Loss Equation was employed to estimate sediment loss. Furthermore, the majority of such models have neither a subsurface drainage component, nor any ability to speciate applied manure. The objective of the present program is to develop a P management tool based on the Root Zone Water Quality Model (RZWQM2) and use it to simulate the fate and transformations of P at the field scale. This new model-based tool, RZWQM2-P, will be used in Canada as a decision-support tool for P management under manure applications. The new features of this P and nitrogen model will include: (i) the use of the Green-Ampt model and variable source area to predict runoff, (ii) simulation of sediment detachment, deposition, and transport processes using a mechanical subroutine adopted from WEPP (Water Erosion Prediction Project), (iii) partitioning of P in manure into soil residue and soil humus pools depending on the application method (surface applied or injected) ,(iv) adsorption/desorption of P will be computed by a NICA (non-ideal competitive adsorption) model which uses Fe/Al/Ca content to predict dissolved P adsorption/desorption; transport of particulate P attached to soil nanoparticles from the soil surface to tile drains will be simulated considering preferential flow, and (v) a crop P uptake and stress subroutine will be adopted from the DSSAT (Decision Support System for Agrotechnology Transfer) model. The new model-based P management tool, RZWQM2-P, will be tested against P levels measured on subsurface-drained crop field in Canada. P in surface runoff and subsurface drainage has been measured in several areas and can be used to calibrate and validate the new management tool. Data from an on-going lysimeter study on P movement under poultry and cattle manure, conducted by my team at the Macdonald Farm of McGill University, can serve in parameterizing the new P model. This new management model can be used as a decision-support tool for water quality and crop yield management under manure applications.