Time to re-think agricultural phosphorus modelling for the 2020s

<div> <p>Phosphorus (P) is critical to our food production systems with many crop systems dependent on continual inputs to meet yield demands. However, a consequence of the widespread application of P to agricultural soils in the past 60 years has led to concerns about the long-term sustainability of P fertiliser supply and to P being transferred from soil systems to watercourses, causing diffuse pollution. This highlights the multi-scaled and interdisciplinary nature of the past, present and future of P management.  </p> </div><div> <p>The aim of this research is to define a starting framework to consider the best ways to develop a model that addresses the contemporary understanding of P processes, integrating the needs of the crop, with biogeochemical and hydrological modelling considerations, going beyond P transfer to the role of P in both food and water challenges.  </p> </div><div> <p>So, this review explores some of the current P models and the future opportunities for expanding their representation of P processes in agricultural systems. This goes beyond nesting existing models and reshapes approaches to posing research and modelling questions to achieve P models that cross disciplinary boundaries and have meaning and usability in practice. As part of this contribution, we welcome modellers and P scientists to come forward and help drive this complex issue of P in agriculture. </p> </div>

Dynamics of nitrogen-fixing cyanobacteria with heterocysts: a stoichiometric model

A simulation model for nitrogen-fixing cyanobacteria was formulated to predict population and nutrient dynamics in water quality studies. The model tracks population biomasses of nitrogen and phosphorus, which potentially limit population growth. Lack of intracellular nitrogen cues the differentiation of specialised heterocysts for nitrogen fixation. Ecoevolutionary analysis presented here predicts that natural selection optimises heterocyst differentiation in relation to external supplies of nitrogen and phosphorus. Modelling the production of N-rich toxins (e.g. anatoxins, saxitoxins) suggests that both total biomass and the biomass N:P ratio can predict concentrations of toxins. The results suggest hypotheses that major taxa of nitrogen-fixing, nuisance cyanobacteria are differentially adapted to varying nitrogen and phosphorus supplies, and that biomass stoichiometry is related to toxins production in this major group of harmful algae. This approach can be extended into models of community and ecosystem dynamics to explore implications of nitrogen fixation for cyanobacterial biomass and toxins production.

Primary production and phosphorus modelling in LakeVegoritis, Greece

Primary production and phosphorus are two of the most important determinants of the water quality of lakes. Phytoplankton primary production and phosphorus cycling were modelled within a one-dimensional lake water quality model. The model was calibrated and applied to Lake Vegoritis in Greece for two different years (1981 and 1993) using daily meteorological variables and inflow rates as input data. Monthly profiles of temperature, chlorophyll-a, and oxygen concentration for these two years were used to calibrate the model. Simulation results indicate that the thermal regime of the lake strongly affects phosphorus profiles and that phytoplankton concentrations throughout the year are tightly correlated with soluble reactive phosphorus concentrations. The significant decrease in the depth and the volume of the lake from 1981 to 1993 resulted in important changes in phytoplankton and phosphorus concentrations. A sensitivity analysis was conducted to estimate the errors resulting from the uncertainty in the biochemical variables of the model and the limited data on phosphorus and phytoplankton.

Developing an independent, generic, phosphorus modelling component for use with grid-oriented, physically based distributed catchment models

Grid-oriented, physically based catchment models calculate fields of various hydrological variables relevant to phosphorus detachment and transport. These include (i) for surface transport: overland flow depth and flow in the coordinate directions, sediment load, and sediment concentration and (ii) for subsurface transport: soil moisture and hydraulic head at various depths in the soil. These variables can be considered as decoupled from any chemical phosphorus model since phosphorus concentrations, either as dissolved or particulate, do not influence the model calculations of the hydrological fields. Thus the phosphorus concentration calculations can be carried out independently from and after the hydrological calculations. This makes it possible to produce a separate phosphorus modelling component which takes as input the hydrological fields produced by the catchment model and which calculates, at each simulation time step, the phosphorus concentrations in the flows. This paper summarises the equations and structure of such a Grid Oriented Phosphorus Component (GOPC) developed by the authors for simulating phosphorus concentrations and loads using the outputs of a fully distributed physical based hydrological model. The GOPC performance is illustrated by an example of a simplified hypothetical catchment subjected to some ideal conditions.