Abstract:
It is extremely challenging to measure first-order rate coefficients for soil nitrate
transformation processes directly, either in the laboratory or in the field. In this study, an
improved inverse method was proposed to optimize the first-order rate coefficients by
considering the intermediate changing processes of the integrated functions. A numerical
experiment was designed to test the accuracy of the method in optimizing the coefficients.
Comparisons between the optimized and theoretical results indicated that all the relative
errors were within 10%. Data collected from a field experiment were used to validate the
optimization procedure and to demonstrate its applications in practice. Using the
established model and the estimated values by the inverse method, the simulated sourcesink
term (SST) distributions of September 2-12, 2007, were in good agreement, with the
root mean squared error (RMSE) between them being as low as 0.00021 mg cm-3 d
-1
.
Based on the established nitrate transformation model, the distributions of soil water
content and nitrate concentration during September 2-12, 2007, were simulated, and
compared well with the measured profiles, with the RMSE of 0.023 cm3
cm-3 and 0.017 mg
cm-3, respectively. The improved inverse method should be useful for optimizing the firstorder
rate coefficients for nitrate transformation, establishing the nitrate transformation
model, and simulating the nitrate transport in the soil-plant system.
Keywords: Numerical simulation, Root-nitrate-uptake, Soil nitrate Kinetics, Soil nitrate
transformation
