Topic 15 Flashcards
(21 cards)
toblers law
basis for interpolation
close things are more related than other things
kriging
geospatial method
stochastic technique
surface models
spatial interpolation and related models
predictive models used to estimate continuous field phenomena
spatial interpolation
operations that use known locations to precit the values of all other areas
used for creating continuous models or datasets
similar to nearest neighbour techniquie
search strategies
directionality of point in search window can have huge impact on the estimated value
simple
quadrant
octant
briefly..what is spatial autocorrelation
all things related are near each other
all forms of interpolation can be categorized by two fundamental classifications
deterministic vs stochastic (geostatistical) models
global vs local
deterministic models
assume that measurements at sampled points are absolute and essentially error-free
true surface is constant and predictable (no variability)
stochastic (geostatistical) models
make predictions from the statistical properties of the sample data, which means they can incorporate error and variability into the prediction process
true surface modelled as a trend
prediction is probabilistic
example of deterministic model
mapping gravity anomalies
example of stochastic model
soil moisture maps
global interpolators
derive single mathematical function that is applied across the entire prediction surface
all sample data are used to build the prediction function
works best with smooth surfaces
trend surface
local interpolators
use neighbourhoods of points for prediction
convolution filters and moving window analysis
IDW
more susceptible to outliers
what interpolation workflow is more commonly used in GIS workflows
local interpolation techniques
IDW
straightforward methods of interpolation
unknwon points are interpolated from nearby points within a fixed or variable distance (known as a neighbourhood)
how is IDW calculated
the unknown point is calcuated to be the weighted mean of neighbours where the weight is assigned by the inverse of the distance
ordinary kriging
most popular form of interpolation from the kirging family
similar to IDW as it estimates the attribute at an unsampled location as a weighted average of the attributes at nearby locations
how is ordinary kriging caluclated
kriging uses a more sophisticated weight, which includes distance and the spatial structure and arrangement of the sample data
examines how different the attributes are based on distance apart then it estimates the points
two steps of ordinary kriging
step 1) construct a model for the semivariogram
step 2) predict attribute values at unsampled locations
semivariogram
model for spatial dependence, or how proximity influences the similarity of nearby points
how does kriging use the semivariogram
uses to estimate the weights, whcih is an expression of spatial dependence (not just simple distance)
