December 2011 article (early view): “Importance of Crop Yield in Calibrating Watershed Water Quality Simulation Tools,” by Sujithkumar Surendran Nair, Kevin W. King, Jonathan D. Witter, Brent L. Sohngen, and Norman R. Fausey
Confidence in a simulation tool’s ability to accurately represent and capture the inherent variability of a watershed is dependent upon high quality input data and subsequent calibration. It is essential that these models allow representation of a wide variety of crops and technology combinations for evaluation and provide reasonable estimates of crop yield and environmental impact of crop production.
In general, calibration and validation of SWAT typically has been evaluated by comparing the simulated surface runoff, and nutrient concentration in runoff, against measured values at a watershed outlet. However, the processes affecting the water and nutrient balance in an agricultural watershed are highly influenced by crop production. The goal of this study was to build upon previous research by further investigating the model’s accuracy in simulating crop biomass and yield, surface flow, nutrient loading, and their interaction in a watershed. The authors propose a calibration approach for SWAT by sequentially integrating hydrology, crop, and nutrient components (in the immediate study only nitrogen).
[Please note: I have quoted and paraphrased freely from the article, but the interpretation is my own!]