Effect of fertilizer application on watermelon yield in light sierozem soils: A mathematical modeling approach

Abstract

This study presents mathematical models developed to predict watermelon (Citrullus lanatus) productivity in response to nitrogen (N), phosphorus (P), and potassium (K) fertilizer application on light sierozem soils in the Turkestan Region of Kazakhstan. Field experiments were conducted using a factorial design to observe growth dynamics, biomass accumulation, and yield performance under varying fertilizer combinations. Regression models were used to describe crop response, with high accuracy indicated by coefficients of determination (R²) ranging from 0.86 to 0.98. The highest observed yield - 48.6 t/ha - was recorded with N200P150K60, closely matching the model’s predicted yield of 47.5 t/ha. The results emphasize the significance of including both the main effects and interactions of N, P, and K factors up to the second degree in predictive modeling. These models offer a practical tool for optimizing fertilization strategies, improving yield forecasting, and supporting precision agriculture practices. The study underscores the value of integrating mathematical modeling with agronomic decision-making to enhance productivity and economic sustainability. Future research should focus on extending these models to various soil and climatic conditions to broaden their applicability.