Nonlinear Dynamics and Optimal Fertilizer Control of a Soil Nitrogen–Crop Biomass System: Persistence Threshold, Equilibrium Stability and Finite-Horizon Pontryagin Analysis
Fertilizer nitrogen decisions involve a trade-off between crop response, nitrogen retained in the soil, nitrogen lost from the root zone and the cost of application. This paper analyses a reduced mechanistic model of this trade-off: a two-state nonlinear ordinary differential equation system for plant-available soil nitrogen and crop biomass, with saturating (Michaelis–Menten type) uptake and density-limited biomass growth, augmented by a cumulative nitrogen-loss accounting state and a bounded fertilizer input. We prove positive invariance, global existence and uniform boundedness of the active states, and derive a sharp threshold for the existence of a positive crop equilibrium under constant supply. The positive equilibrium is unique when it exists and is locally asymptotically stable; the boundary (crop-free) equilibrium loses stability exactly at the threshold. Because the loss state integrates a non-negative flux, it has no stationary limit and is analysed separately. A finite-horizon optimal control problem trades terminal biomass and residual soil nitrogen against cumulative loss and linear-plus-quadratic fertilizer cost. We establish existence of an optimal control, derive the Hamiltonian, adjoint system and projection condition, and interpret the adjoint as the marginal value of soil nitrogen. In a normalized numerical study, an optimized twenty-segment schedule reaches 99% of the terminal biomass of the high-input policy with 23% less fertilizer and 20% less cumulative loss, and the optimized objective is stable under control-mesh refinement and multistart optimization. One-at-a-time sensitivity, cost-weight and loss-penalty experiments show that the timing of the late-season application ramp, rather than its magnitude, responds most strongly to relative costs. Parameters are illustrative rather than field-calibrated, so the results are mathematical findings and a calibration-ready framework, not agronomic recommendations.
Authors
- Napoleon Ochonogor
- OKONYE Angela Okwuolise
Institutions
- Dennis Osadebay University (NG)
Publication Details
- Journal
- IPS Journal of Physical Sciences
- Published
- 2026-11-21
- DOI
- https://doi.org/10.54117/ijps.v3i2.49
- Primary Topic
- Agricultural Science and Fertilization
- Type
- article
- Field-Weighted Citation Impact
- 0.00