What information do we need to combine effects?
Once we have identified the effects that contribute to the uncertainty on each term in the measurement function, we can begin to think about how these effects can be combined and propagated to give the uncertainty in the measurand. In order to combine effects we need three key pieces of information:
- The standard uncertainty associated with the effect
- The sensitivity coefficient
- The correlation structure over spatial, temporal and spectral scales for errors from this effect
We’ve already examined the first two of these pieces of information, and have seen that sensitivity coefficients are used to translate the standard uncertainty associated with a given effect into uncertainties associated with the measurand. In the next recipe, we will move on to examine the last of these three pieces of information by looking at correlation and correlation structures.