ASABE Technical Library - Abstract

Click on “Download PDF” for the PDF version or on the title for the HTML version.
If you are not an ASABE member or if your employer has not arranged for access to the full-text, view access and subscription options

Comparing a Simple Carbon Nitrogen Model with Complex RZWQM Model

Published by the American Society of Agricultural and Biological Engineers, St. Joseph, Michigan www.asabe.org

Citation:  2019 ASABE Annual International Meeting  1901394.(doi:10.13031/aim.201901394)
Authors:   Anupam Bhar, Ratnesh Kumar, Robert W Malone
Keywords:   Model Comparison, RZWQM, Soil Nitrogen Model.

Abstract. Accurate modeling of soil Carbon and Nitrogen system is crucial for the accuracy of the overall agriculture models. This has led to C and N dynamics, as captured in advanced agricultural modeling formalisms such as RZWQM, to be quite complex, involving numerous parameters. Calibrating a large number of parameters requires more data to avoid under fitting, and is also time-consuming. The execution time of the complex model is also higher, and is of concern when performing an optimal decision-making such as of irrigation and fertilizer scheduling. A study of tradeoff among modeling complexities, and the corresponding impact on modeling accuracy, is desirable. In this regard, our paper examines a simpler soil C and N dynamics proposed by Porporato et al. and compares it with a relatively advanced model, namely, RZWQM. The C-N module in RZWQM has three microbial, five organic matter, nitrate, ammonia, and also inorganic C and N pools, and comprises of around 25 parameters. In contrast, the simplified Porporato model has one microbial, two organic matter, nitrate and ammonia pools, and it involves a total of 12 parameters. We subject the two models to same applications of fertilizer and water, and compare the two models on the basis of their daily nitrate and ammonia concentrations. The RZWQM model was calibrated with data from experiments conducted at USDA fields in Greeley, Colorado, 2010. The Porporato model was calibrated to match RZWQM predicted Nitrate and Ammonia. The variability that gets introduced to crop dynamics was discounted, by zeroing out the crop growth. The comparison (between simple Porporato vs RZWQM) results showed coefficient of determination (R2) match of 0.99 and 0.62 (1 for perfect match) for Nitrate and Ammonia respectively, while 35-fold saving in simulation time. This is encouraging as having an accurate and fast executing model enables real-time decision-making as well as fast prototyping for offline decision-designs.

(Download PDF)    (Export to EndNotes)