[ai-geostats] Probability part 2
To add to my previous question, I have a continuous dataset of spectral readings from local crops. Under the idea of spatial continuity I would assume that the degree of relatedness between observations will provide the structure of a functional relationship between observations. For instance, if fertilizer is added to an observation point and it produces a response, what would the probability be for a similar response at an adjacent point, given that the two points are related. Here I assume that the response is a function of the spatial random variable and is basically a scalar.
I feel, whether or not I am right remains to be seen, that this would be similar to posing the question: at point A. there is an observed ore concentration level, what is the probably of a similar level at an unknown point close to point A. I would assume that closer points have a higher probability of being similar, whereas points further apart do not. My thought is that the variogram describes this probability; however, I am unsure how to make the connection or more specifically if it is valid to make the connection between the two. In at least one conference proceeding I found a presentation that referred to probability kriging. Could that be the answer to my question?
Thanks for the responses.
OSU Environmental Sciences
The semi-variogram function (if it reaches a sil at
some point) is directly analogous to the covariance
between samples given the distance (and possibly
direction) between them. If you take:
total sill (final height of semi-variogram)
semi-variogram value at a given distance
you get the covariance between the values at two
locations separated by that distance. If you divided
through by the total sill you (theoretically) get the
correlation between them.
So it isn't really a probability function but a
You can derive probabilities for unsampled locations
by theory or simulation through the kriging process.