Lecture 1

Unit III: Spatial Interpolation: Concepts,

techniques and uses
What is spatial interpolation?
• Spatial interpolation is a method used to estimate values at
unmeasured locations based on data from known points. It is widely
used in geographic information systems (GIS), meteorology,
environmental science, and various spatial analyses.
• Key Points of Spatial Interpolation:
• Predicts unknown values using existing spatial data.
• Assumes spatial autocorrelation, meaning nearby points are more
similar than distant ones.
• Creates continuous surfaces from discrete data points (e.g.,
temperature maps, elevation models).
Importance
• Spatial interpolation is essential because it helps us estimate and
analyze spatial data in various fields.
1. Completes Missing Data
• Not all locations have direct measurements (e.g., weather stations,
soil samples). Spatial interpolation fills in gaps to create continuous
datasets
2. Supports Decision-Making
• Meteorology: Predicts temperature, rainfall, and pollution levels in
areas without sensors.
• Agriculture: Estimates soil moisture for better crop management.
• Urban Planning: Helps assess population density, land use, and
environmental risks.
3. Enhances Maps & Visualizations
• Interpolation transforms scattered data into useful maps, making
complex information more accessible for analysis and
communication.
4. Enables Predictions
• Many models use interpolation techniques to forecast future trends—
like climate change, land use changes, and economic shifts.
Geographic Information Systems (GIS) rely on interpolation to
create elevation models, terrain surfaces, and resource distribution
maps.
Spatial Data & Geographic Information in Spatial
Interpolation
• Spatial interpolation relies on spatial data—measured points with
geographic coordinates—to estimate values at unmeasured locations.
• Point Data: Example—temperature readings from weather stations.
• Raster Data: Example—satellite images representing elevation.
Techniques
• Spatial interpolation techniques can be divided into two main
categories: deterministic and geostatistical approaches.
• Deterministic Methods use mathematical formulas to estimate values
based on known data points. Examples include:
• Inverse Distance Weighting (IDW): Assigns higher importance to closer points.
• Spline Interpolation: Creates a smooth surface from known values.
• Geostatistical Methods rely on statistical models that consider spatial
correlation. Examples include:
• Kriging: Uses variograms to estimate values with uncertainty analysis.
• To put it simple, deterministic methods do not try to capture the
spatial structure in the data. They only make use of predefined
mathematical equations to predict values at unsampled locations (by
weighing the attribute values of samples with known locations).
• On the contrary, geostatistical approaches intend to fit a spatial
model to the data. This enables the generation of a prediction value
at unsampled locations (like deterministic methods) and provides
users with an estimate of the accuracy of this prediction.
Inverse Distance Weighting (IDW)
• Inverse Distance Weighting (IDW) is a deterministic interpolation
method used to estimate values at unknown locations based on the
weighted average of nearby known points. The key assumption is that
closer points have more influence on the estimated value than distant
points.
• How IDW Works
• Each known point contributes to the unknown value, but closer
points have greater weight.
• The weight is inversely proportional to the distance—meaning the
farther the point, the less influence it has.
• The interpolation follows the formula:
W(d) = 1/d^p, where W(d) is the weight, d is the distance,
and p is the powered parameter that controls the influence of
the nearby points.
• As p increases, the weight attributed to distant observations
decreases. This factor can be considered a smoothing parameter.
• In fact, if p is small, neighboring and distant observations exercise a
strong influence on unsampled locations. As a consequence, the
prediction might be heavily smoothed because it does not solely
consider local processes, which might blur some interesting variations
or make the variable look much more homogeneous than it actually
is.
• On the other hand, if p is set too high, only the very local phenomena
will be accounted for, and the interpolated map might present a
jagged appearance.
Trend Surface Analysis
• Trend surface analysis intends to fit a trend to the data. In other words, the
objective is to find the equation that best matches the attribute values of
the available data. Trend surface analysis often involves multiple linear
regressions or polynomial functions to find a relationship between the
predicted variable and a combination of the spatial coordinates.
• This approach is a global technique because it considers all the data
available to put into place the linear regressions or the polynomial
functions.
• More local approaches could be implemented to improve the
prediction accuracy. Note that the trend surface analysis can suffer
from outlying values.
• In fact, an abnormal observation is likely to severely influence the
proposed model and might lead to a really bad fit to the data.
• To make the trend surface analysis
more reliable, there is a need to
calibrate and then validate the
selected regressions or polynomial
functions.
• Key Points
• It helps identify large-scale spatial patterns
and trends.
• The method uses polynomial equations to
approximate the surface based on
observed data points.
• It's widely used in terrain modeling, soil
distribution studies, and climate analysis.
• Though useful, it may oversimplify complex
spatial variations, requiring careful
interpretation.
• Polynomial functions differ based on their degree (the highest
exponent of the variable).
• Linear Function (First order): Graph: A straight line; Constant rate of
change (slope)
• Quadratic Function (Second order): Graph: A parabola (U-shaped
curve)
• Cubic Function (Third order): Graph: Can have an S-shape, with up to
two turning points
• Quartic Function (Fourth order): Graph: More complex, can have
three turning points; Can exhibit multiple curves and bends.
Here,
• z → The dependent variable (elevation, temperature, pollution, etc.)
• b,c → Independent spatial coordinates
• a → Constant term (baseline value)
• u,v → Coefficients representing linear trends in the b and c directions
Spline
• A spline is a piecewise-defined polynomial function, meaning it's
composed of multiple polynomial segments connected together in a
way that ensures smoothness. Instead of fitting one large polynomial
across all data points (which can lead to erratic behavior), splines use
smaller, localized polynomials to maintain flexibility and accuracy.
• By building piece-wise polynomial functions that connect data points
gradually, spline interpolation prevents abrupt changes in the
interpolated function.
• The data points are divided into
intervals (basically in pairs)
• Each interval is assigned a polynomial
segment (usually cubic)
• These segments are connected at knots
(the given data points)
• Smoothness conditions are applied:
✓The function itself is continuous
✓First and second derivatives are
continuous (for cubic splines)
• This ensures a smooth curve without
sharp breaks.
Kriging
• A second family of interpolation methods consists of geostatistical methods,
such as kriging, which are based on statistical models that include
autocorrelation—that is, the statistical relationships among the measured
points. Because of this, geostatistical techniques not only have the capability
of producing a prediction surface but also provide some measure of the
certainty or accuracy of the predictions.
• Kriging assumes that the distance or direction between sample points reflects
a spatial correlation that can be used to explain variation in the surface.
• The Kriging tool fits a mathematical function to a specified number of points
or all points within a specified radius to determine the output value for each
location.
• Kriging is a multistep process; it includes exploratory statistical analysis of the
data, variogram modeling, creating the surface, and (optionally) exploring a
variance surface.
• Kriging is most appropriate when you know there is a spatially correlated
distance or directional bias in the data. It is often used in soil science and
geology, environmental studies, urban planning, and epidemiology to analyze
spatial trends, such as disease outbreaks or land-use patterns
Creating a prediction surface map with kriging
• Creating the variograms and covariance functions to estimate the
statistical dependence (called spatial autocorrelation) values that
depend on the model of autocorrelation (fitting a model).
• Predicting the unknown values (making a prediction).
• Variography
• In spatial modeling of the structure of the measured points, you begin
with a graph of the empirical semivariogram, computed with the
following equation for all pairs of locations separated by distance h:
Semivariogram(distanceh) = 0.5 * average((valuei –
valuej)2)
• The formula involves calculating the difference squared between the
values of the paired locations.
Often, each pair of locations has a unique distance, and
there are often many pairs of points. To plot all pairs
quickly becomes unmanageable. Instead of plotting
each pair, the pairs are grouped into lag bins. For
example, compute the average semivariance for all
pairs of points that are greater than 40 meters apart
but less than 50 meters.

The empirical semivariogram is a graph of the averaged

semivariogram values on the y-axis and the distance
(or lag) on the x-axis
Fitting a model to the empirical semivariogram
• The next step is to fit a model to the points forming the empirical
semivariogram. Semivariogram modeling is a key step between spatial
description and spatial prediction.
The Kriging tool provides the following functions from which to choose
for modeling the empirical semivariogram:
• Circular
• Spherical
• Exponential
• Gaussian
• Linear
1. Circular Kriging
•Assumes spatial correlation increases rapidly at short distances and then levels off.
•The transition is smooth, making it useful for gradual spatial changes.
•Often applied in environmental studies, where data transitions smoothly.
2. Spherical Kriging
•Similar to circular, but the correlation drops to zero beyond a certain range.
•Provides a clear cutoff point for spatial influence.
•Commonly used in geology and soil science.
3. Exponential Kriging
•Assumes correlation decreases gradually with distance but never truly reaches zero.
•Suitable for long-range dependencies or continuous spatial changes.
•Often used in meteorology and hydrology.
4. Gaussian Kriging
•Assumes correlation decreases very smoothly over distance.
•Best for highly continuous spatial phenomena.
•Used in air pollution modeling and temperature distribution.
5. Linear Kriging
•Assumes a linear decrease in correlation with distance.
•Works well for simple spatial relationships.
•Used in basic interpolation tasks.