A Guide To Interpreting NAPRA Results

Modeling Comments...

Models are not always good at predicting exact values without calibration for the specific conditions to which the model is being applied. NAPRA is being applied in a non-calibrated manner within the workshop and in most applications. Therefore, the actual values for runoff, pesticide concentrations in runoff and nutrient losses could vary significantly from the predicted values.

Computer models such as NAPRA though are generally good at estimating the direction and magnitude of change for model outputs. For example, NAPRA should provide reasonable estimates of changes in nitrogen lost to runoff for a change in timing of fertilizer. It will also provide reasonable comparisons among pesticides.

Sensitive NAPRA Parameters..

Models react differently for each model parameter or input. Some parameters affect some results much more than other parameters. Therefore, it is difficult to make general statements about the sensitivity of NAPRA parameters or inputs.

The following are NAPRA inputs that may greatly affect the pesticide losses:

  • Pesticide type
  • Pesticide application rate
  • Timing of pesticide application
  • Pesticide application method (surface applied vs incorporated)
  • Soil properties (soil properties in NAPRA are obtained from the NASIS soil database for the soil symbol/name selected)
  • Root zone depth (once a pesticide moves below the root zone, it is considered lost to shallow groundwater)

The following are NAPRA inputs that may greatly affect nutrient losses:
  • Fertilizer/livestock manure application rate
  • Timing of application
  • Application method
  • Soil properties
  • Root zone depth