Uncertainty analysis of solute transport modeling at regional scale using soil survey data

Deterministic simulation models are being developed and used for understanding and quantifying major transport processes involved in the leaching of solutes from soils under various climate conditions, soil properties, and agricultural practices. Such models provide useful tools for water management and chemical application studies of non-point source pollution at regional scales. In contrast, soil information at regional scales is uncertain, in the sense that it can have many possible values.; When uncertain input data is used in simulation models, the model output data can also be expected to be uncertain. If a simulation model is used to support land and water management decisions, the uncertainty of model predictions can be of particular importance in defining the level of confidence of using the model output to influence the decision making process.; In this study a systematic approach to assessing the impact of soil data uncertainty on leaching model predictions was developed. The approach based on Monte Carlo simulations was used to estimate the probability distribution of leaching model outcomes resulting from the uncertainty of soil survey information. Map unit variability was characterized by the data uncertainty of individual soil components. The proposed method was found easy to conceptualize and to implement.; Additionally, two simulation approaches to water flow and solute movement using deterministic simulation models that differ in algorithm complexity, one being a mechanistic-type and the other a functional-type, were compared using the uncertainty analysis approach developed here. The results of the comparison suggested that the simpler functional modeling approach of water and solute movement has promise as an alternative approach with comparable precision to more complex mechanistic models when used for regional scale studies.; Current applications were reviewed and recommendations made for the appropriate use of and improvements to soil survey data for modeling pesticide leaching at regional scales.