Coupled hydrosystem models (CHMs) can provide useful insight to the hydrological system when used appropriately. However, every model has inherent assumptions and simplifications leading to uncertainty in model predictions. Typically numeric CHMs are used in a deterministic way without rigorous quantification of model output uncertainty. Uncertainty analysis forces the modeller to scrutinize the processes involved when arriving at a modelled result.
To prioritize CHM uncertainty I must first identify the sources of model uncertainty than quantify, normalize, and prioritize (rank) these uncertainty sources. Numeric CHMs have three main inherent uncertainty sources as described by (Gupta et al. 2005), which are:
- i. Data
- ii. Parameter
- iii. Structural or Conceptual
Developing the methodology to create the CHMs and quantify uncertainty is currently a work in progress. With quantifiable measures of model uncertainty, I will prioritize the sources of uncertainty in CHMs and investigate methods to minimize the greatest uncertainty sources. This will ensure that CHMs will be used appropriately and not used in a way that exceed their predictive capacity.
