2. What is INTERPOLATION ?
The prediction of values in the spaces between data points
Deterministic Statistical
“exact” interpolators
Good for data where points are constant
Ex; Inverse Distance Weighting (IDW)
inexact, but with quantification of error
Good for noisy data
Ex; Kriging
4. IDW
INTERPOLATION
The best results from IDW are obtained when sampling is sufficiently
Weight of each sample point is an inverse proportion to the distance
Involves the estimation of variables at non-sampled locations
Estimate the values at unknown points using the distance and values
to nearby known points
A larger number of sample points imply in a smoother surface
WHEN TO USE ?
Temperature of specific country
Productivity of a country
Elevation
population density
5. KRIGING
INTERPOLATION
Delivers a measure of confidence of how likely that prediction will be
true
The estimations are weighted averaged input point values
The weight factors in Kriging are determined by using a user-specified
semi-variogram model
ORDINARY KRIGING
UNIVERSAL KRIGING
+ WHEN TO USE ?
Environmental science (soil type)
Hydrology
Natural resources studies (water, atmosphere, vegetation)
The Key to Kriging is the Semi variogram
Linear | Spherical | Exponential | Gaussian | Circular
6. Surface is constructed according to variance
Also known as Sibson or Area-stealing Interpolation
A geometric estimation technique and a weighted-average
interpolation method
Appropriate where sample data points are distributed with uneven
density
Associated with neighboring Voronoi (Thiessen) polygons
WHEN TO USE ?
When there is a large no of sample points
NATURAL NEIGHBOR
INTERPOLATION
7. SPLINE
INTERPOLATION
A smooth distribution of values
Interpolates a raster surface from points using a two-dimensional
minimum curvature spline technique
Additional spline parameters
Estimates values using a mathematical function
The number of sample values is relatively small
REGULARIZED SPLINE TYPE
TENSION SPLINE TYPE
+ WHEN TO USE ?
Temperature data
1. Weight parameter
2. Number of points parameter
8. TREND
INTERPOLATION
Surface is constructed according to variance
Statistical method
Least-square regression model
One polynomial equation to the entire surface
Minimizes surface variance in relation to the input values
LINEAR TREND
LOGISTICS TREND
+ WHEN TO USE ?
Pollution over an industrial area
wind direction
14. REFERENCE
• VALIDATION OF SPATIAL INTERPOLATION TECHNIQUES IN GIS, V.P.I.S. Wijeratne and L.Manawadu
University of Colombo (UOC)
• Spatial Interpolation of Rainfall Data Using ArcGIS: A Comparative Study, University of South
Florida St. Petersburg
• Spatial Interpolation of Rainfall Data Using ArcGIS: A Comparative Study , Julie Earls & Dr. Barnali
Dixon
• ArcGIS Help : http://desktop.arcgis.com/en/
• Interpolating Surfaces in ArcGIS Spatial Analyst, C. Childs