Abstract:
Maize and sugarcane production has been threatened by declining soil quality due to long-term unsustainable management practices, which have increased the reliance on inorganic fertilization. This study reports on long-term yields and soil organic matter (SOM) trends and further evaluates nitrogen (N) leaching losses from maize and sugarcane as affected by inorganic fertilization and residue management practices. The study aims to investigate the effects of long-term management practices on maize and sugarcane monocropping systems in South Africa, through the application of long-term monitoring data and mechanistic modelling. Data from the University of Pretoria’s Hillcrest Campus Experimental Farm long-term maize trial and SASRI long-term sugarcane trial, in Mt Edgecombe were used. The APSIM model was calibrated and validated using long-term yield and SOM data, and the model was further used to estimate N leaching and evaluate management scenarios that can be used for more sustainable maize and sugarcane production. Although the model could be well calibrated for simulating maize growth for 2016/2017 season, long-term yields were not always accurately estimated. The results indicated a declining trend in maize yields and SOM over the years, with greater decline in the control treatment. APSIM estimated higher drainage in the maize control but higher N leaching in the fertilized NPK treatment. A manure application scenario proved to be more sustainable for long-term maize production, although it requires a good inorganic N fertilizer management programme to minimize N leaching losses. In sugarcane, observed and simulated results indicated that fertilizer application increased yields, and mulching was the best residue management practice for reducing SOM decline. The combination of fertilization and mulching produced higher long-term sugarcane yields and retained SOM content better, but it also led to the highest NO3-N leaching. Modelling reduced fertilizer application did not result in a significant reduction in yield, indicating that mineralized N from SOM can be able to satisfy a proportion of crop N demand, so fertilizer application recommendations should also account for mineralized N to minimize N losses. This can be the best way of improving N management in sugarcane cropping systems, thus reducing inputs, increasing profits and minimizing losses that can lead to environmental pollution. Modelling has the potential of helping us understand the complex long-term C and N dynamics in cropping systems and identification of ways to improve management practices.
