Results of a simulation are obtained as time series either of variables, which represent individual layers in the soil such as:
• temperature
• content of ice
• content of unfrozen water
• water potential
• vertical and horizontal flows of heat and water
• water uptake by roots
• storage’s of water and heat
• nitrogen and carbon content in different storages in the soil and the flux of matter between these storages
In addition some output variables are represented as a single variable such as:
• snow depth
• water equivalent of snow
• frost depth
• surface runoff
• drainage flow
• deep percolation to ground water
• carbon and nitrogen content in the plant
• carbon assimilation and respiration
• nitrogen uptake
It is a well-known fact that no simulation model yields better results than what can be expected from the quality of input data. Assessment of the uncertainty in the input data is therefore the first step when the model is to be used. Sometimes field measurements are available which enable a quantitative test of the model. The interpretation of discrepancies found between the measurements and the model predictions requires a lot of care and a basic knowledge of the different processes in the system. An improvement of the fit can normally be obtained after adjustments of some soil or plant properties. Nevertheless, it is not necessarily so that all input data including the physical properties of the system are correctly estimated just because a good fit is obtained when testing the model.
Note that we can always simulate a much more complete picture of both the temporal pattern and of the interaction between variables than what can be achieved by intensive field measurements. However, this should not lead us to believe more in the model predictions than in observations of the real system. Instead we have to design our field measurements to achieve an optimum test of the simulated results. We should concentrate on variables which are easy to measure and which have a strong connection to other variables in the soil-plant-atmosphere system. A typical example is soil water tension, which is easy to measure with a conventional tensiometer, but in addition reflects other factors such as soil water flow and water uptake by roots. Unsaturated water flows are very difficult to measure in field soils and in this case we must always rely on model predictions. However, tracers can be used as indicators of the actual water flow paths in the soil.