A comparison of three mathematical models of response to applied nitrogen using lettuce

Abstract

Modern fertilization recommendation must optimize crop yield and quality and minimize chances of negative environmental effects due to over fertilization. Data from fertilizer studies can be fitted to several mathematical models to help determine optimum fertilizer rates, but resulting recommendations can vary depending on the model chosen. In this study, lettuce (Lactuca sativa L.) was used as a case study vegetable crop to compare models for estimating fertilizer N requirements. Field studies were conducted to measure yield response to applied N. The area was located between 25 ,21 E longitude and 51 ,38´ N latitude in the North of ´ Varamin city, (Tehran province, Iran) in the alluvial plain of Varamin. Soils family were fine, mixed, active, thermic, typic haplocambids based on Soil Taxonomic system. Plants were grown in Central Research Station of Varamin and received five rates of N (0.0, 150, 200, 250 and 300 kg ha ) as a urea in split applications. Data 1 for plant fresh mass and N uptake were recorded. The logistic model described the data for cultivar quite well, with correlation coefficients of 0.90 and above. Logistic, linear-plateau and quadratic models were compared for the field data. Coefficients for the linear-plateau model were derived from the logistic model. All three models for lettuce production were compared graphically and analytically. The model coefficients were then used to make improved estimates of fertilizer recommendations for field production of lettuce.