Ethiopian Strategy Support Program II (ESSP II)

An Introduction to
Geographical Information Systems

Training Manual
Emily Schmidt, Helina Tilahun,
Mekamu Kedir, and Hailu Shiferaw

Ethiopia Strategy Support Program II (ESSP II)

Ethiopian Development Research Institute (EDRI)

ESSP II / EDRI Training Manual

December 2011

IFPRI-HEADQUARTERS ETHIOPIAN DEVELOPMENT RESEACH INSTITUTE

INTERNATIONAL FOOD POLICY RESEARCH INSTITUTE http://www.edri.org.et/
2033 K Street, NW • Washington, DC 20006-1002 USA Blue Building • Addis Ababa Stadium
Tel: +1-202-862-5600 • Skype: IFPRIhomeoffice P.O. Box 2479 • Addis Ababa, Ethiopia
Fax: +1-202-467-4439 • E-mail: ifpri@cgiar.org Tel: +251 11 5 50 60 66; +251 11 5 52 53 15
Fax: +251 11 5 50 55 88
IFPRI-Addis Ababa Email: exe-director@edri.org.et
http://essp.ifpri.info
IFPRI c/o ILRI
P.O. Box 5689, Addis Ababa, Ethiopia
Tel: +251 11 6 17 25 55 Fax: +251 11 6 46 23 18
E-mail: ifpri-addis@cgiar.org
Contact: Bart Minten, Senior Research Fellow and Program Leader
 Contents

Lab 11: Meteorological data presentation............................................................................ 61

An overview of the interpolation tools .............................................................................. 61
Lab 11 - Exercise Part 1: Using NMA data (Station data) .................................................... 68
Lab 11 - Exercise Part 2: Using rainfall estimation (RFE) satellite data ............................... 83
Lab 12: Agriculture Survey Data Integration ........................................................................ 86
Exercise overview ........................................................................................................... 86
Lab 11: Meteorological data presentation
An overview of the interpolation tools
Visiting every location in a study area to measure the height, magnitude, or concentration of a
phenomenon is usually difficult or expensive. Instead, you can measure the phenomenon at
strategically dispersed sample locations, and predicted values can be assigned to all other locations.
Input points can be either randomly or regularly spaced or based on a sampling scheme.
Surface interpolation functions create a continuous (or prediction) surface from sampled point values.
The continuous surface representation of a raster dataset represents height, concentration, or
magnitude—for example, elevation, pollution, or noise. Surface interpolation functions make
predictions from sample measurements for all locations in a raster dataset whether or not a
measurement has been taken at the location. There is a variety of ways to derive a prediction for each
location; each method is referred to as a model. With each model, there are different assumptions
made of the data, and certain models are more applicable for specific data—for example, one model
may account for local variation better than another. Each model produces predictions using different
calculations.

Understanding interpolation analysis

Interpolation predicts values for cells in a raster from a limited number of sample data points. It can be
used to predict unknown values for any geographic point data, such as elevation, rainfall, chemical
concentrations, noise levels, and so on. The illustrations below show the distribution and values of
sample points and the raster generated from them. The left illustration shows a point dataset of known
rainfall-level values. The illustration on the right shows a raster interpolated from these points.
Unknown values are predicted with a mathematical formula that uses the values of nearby known
points.
Why interpolate to raster? The assumption
that makes interpolation a viable option is
that spatially distributed objects are spatially
correlated; in other words, things that are
close together tend to have similar
characteristics. For instance, if it is raining
on one side of the street, you can predict
with a high level of confidence that it is
raining on the other side of the street. You
would be less certain if it was raining across
town and less confident still about the state
of the weather in the next district.
Using the above analogy, it is easy to see that the values of points close to sampled points are more
likely to be similar than those that are farther apart. This is the basis of interpolation. A typical use for
point interpolation is to create an elevation surface from a set of sample measurements.
In the following graphic, each symbol in the point layer represents a location where the elevation has
been measured. By interpolating, the values between these input points will be predicted.

GIS Manual – December 2011 Lab 11 61

Interpolation Techniques
The IDW (inverse distance weighted) and
Spline interpolation tools are referred to as
deterministic interpolation methods because they
are directly based on the surrounding measured
values or on specified mathematical formulas
that determine the smoothness of the resulting
surface. 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.
How Kriging works: Kriging is an advanced geostatistical procedure that generates an estimated
surface from a scattered set of points with z-values. Unlike other interpolation methods supported by
ArcGIS Spatial Analyst, to use the Kriging tool effectively involves an interactive investigation of the
spatial behavior of the phenomenon represented by the z-values before you select the best estimation
method for generating the output surface.
What is kriging? 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.
The kriging formula: Kriging is similar to IDW in that it weights the surrounding measured values to
derive a prediction for an unmeasured location. The general formula for both interpolators is formed
as a weighted sum of the data:

Where, Z(si) = the measured value at the ith location

λi = an unknown weight for the measured value at the ith location

s0 = the prediction location

N = the number of measured values

In IDW, the weight, λi, depends solely on the distance to the prediction location. However, with the
kriging method, the weights are based not only on the distance between the measured points and the
prediction location but also on the overall spatial arrangement of the measured points. To use the
spatial arrangement in the weights, the spatial autocorrelation must be quantified. Thus, in ordinary
kriging, the weight, λi, depends on a fitted model to the measured points, the distance to the
prediction location, and the spatial relationships among the measured values around the prediction

62 Lab 11 GIS Manual – December 2011

location. The following sections discuss how the general kriging formula is used to create a map of
the prediction surface and a map of the accuracy of the predictions.

Creating a prediction surface map with kriging

To make a prediction with the kriging interpolation method, two tasks are necessary:

Uncover the dependency rules

Make the predictions.

To realize these two tasks, kriging goes through a two-step process:

1. It creates the variograms and covariance functions to estimate the statistical dependence
(called spatial autocorrelation) values that depend on the model of autocorrelation (fitting a
model).
2. It predicts the unknown values (making a prediction).

It is because of these two distinct tasks that it has been said that kriging uses the data twice: the first
time to estimate the spatial autocorrelation of the data and the second to make the predictions.

Variography
Fitting a model, or spatial modeling, is also known as structural analysis, or 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:
2
Semivariogram(distanceh) = 0.5 * average{(valuei – valuej} ]

The formula involves calculating the difference squared between the values of the paired locations.
The image below shows the pairing of one point (the red point) with all other measured locations. This
process continues for each measured point.

GIS Manual – December 2011 Lab 11 63

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 (see diagram beside).
Spatial autocorrelation quantifies a basic principle of geography: things that are closer are more alike
than things farther apart. Thus, pairs of locations that are closer (far left on the x-axis of the
semivariogram cloud) should have more similar values (low on the y-axis of the semivariogram cloud).
As pairs of locations become farther apart (moving to the right on the x-axis of the semivariogram
cloud), they should become more dissimilar and have a higher squared difference (moving up on the
y-axis of the semivariogram cloud).

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 main application of
kriging is the prediction of attribute values at unsampled locations. The empirical semivariogram
provides information on the spatial autocorrelation of datasets. However, it does not provide
information for all possible directions and distances. For this reason, and to ensure that kriging
predictions have positive kriging variances, it is necessary to fit a model—that is, a continuous
function or curve—to the empirical semivariogram. Abstractly, this is similar to regression analysis, in
which a continuous line or curve is fitted to the data points.
To fit a model to the empirical semivariogram, select a function that serves as your model—for
example, a spherical type that rises and levels off for larger distances beyond a certain range (see the
spherical model example below). There are deviations of the points on the empirical semivariogram
from the model; some points are above the model curve, and some points are below. However, if you
add the distance each point is above the line and add the distance each point is below the line, the
two values should be similar. There are many semivariogram models from which to choose.

Semivariogram models
ArcGIS Spatial Analyst provides the following functions from which to choose for modeling the
empirical semivariogram:

 Circular; Spherical; Exponential; Gaussian ; Linear

The selected model influences the prediction of the unknown values, particularly when the shape of
the curve near the origin differs significantly. The steeper the curve near the origin, the more influence
the closest neighbors will have on the prediction. As a result, the output surface will be less smooth.
Each model is designed to fit different types of phenomena more accurately.
The diagrams below show two common models and identify how the functions differ:

64 Lab 11 GIS Manual – December 2011

A spherical model example
This model shows a progressive decrease of
spatial autocorrelation (equivalently, an increase
of semivariance) until some distance, beyond
which autocorrelation is zero. The spherical model
is one of the most commonly used models.

An exponential model example

This model is applied when spatial autocorrelation
decreases exponentially with increasing distance.
Here, the autocorrelation disappears completely
only at an infinite distance. The exponential model
is also a commonly used model. The choice of
which model to use is based on the spatial
autocorrelation of the data and on prior knowledge
of the phenomenon.

Understanding a semivariogram—Range, sill, and nugget

As previously discussed, the semivariogram depicts the spatial autocorrelation of the measured
sample points. Because of a basic principle of geography (things that are closer are more alike),
measured points that are close will generally have a smaller difference squared than those farther
apart. Once each pair of locations is plotted after being binned, a model is fit through them. Range,
sill, and nugget are commonly used to describe these models.

Range and sill

When you look at the model of a
semivariogram, you will notice that at a
certain distance the model levels out.
The distance where the model first
flattens is known as the range. Sample
locations separated by distances closer
than the range are spatially
autocorrelated, whereas locations
farther apart than the range are not.
The value at which the semivariogram
model attains the range (the value on
the y-axis) is called the sill. A partial sill
is the sill minus the nugget. The nugget
is described in the following section.

GIS Manual – December 2011 Lab 11 65

Nugget
Theoretically, at zero separation distance (for example, lag = 0), the semivariogram value is 0.
However, at an infinitely small separation distance, the semivariogram often exhibits a nugget effect,
which is a value greater than 0. If the semivariogram model intercepts the y-axis at 2, then the nugget
is 2.
The nugget effect can be attributed to measurement errors or spatial sources of variation at distances
smaller than the sampling interval (or both). Measurement error occurs because of the error inherent
in measuring devices. Natural phenomena can vary spatially over a range of scales. Variation at
microscales smaller than the sampling distances will appear as part of the nugget effect. Before
collecting data, it is important to gain an understanding of the scales of spatial variation in which you
are interested.

Making a prediction
After you have uncovered the dependence or autocorrelation in your data (see Variography section
above) and have finished with the first use of the data—using the spatial information in the data to
compute distances and model the spatial autocorrelation—you can make a prediction using the fitted
model. Thereafter, the empirical semivariogram is set aside.
You can now use the data to make predictions. Like IDW interpolation, kriging forms weights from
surrounding measured values to predict unmeasured locations. As with IDW interpolation, the
measured values closest to the unmeasured locations have the most influence. However, the kriging
weights for the surrounding measured points are more sophisticated than those of IDW. IDW uses a
simple algorithm based on distance, but kriging weights come from a semivariogram that was
developed by looking at the spatial nature of the data. To create a continuous surface of the
phenomenon, predictions are made for each location, or cell centers, in the study area based on the
semivariogram and the spatial arrangement of measured values that are nearby.

Kriging methods
There are two kriging methods: ordinary and universal. Ordinary kriging is the most general and
widely used of the kriging methods and is the default. It assumes the constant mean is unknown. This
is a reasonable assumption unless there is a scientific reason to reject it.
Universal kriging assumes that there is an overriding trend in the data—for example, a prevailing
wind—and it can be modeled by a deterministic function, a polynomial. This polynomial is subtracted
from the original measured points, and the autocorrelation is modeled from the random errors. Once
the model is fit to the random errors and before making a prediction, the polynomial is added back to
the predictions to give meaningful results. Universal kriging should only be used when you know there
is a trend in your data and you can give a scientific justification to describe it.

66 Lab 11 GIS Manual – December 2011

Sample before and after interpolation
Before interpolation

After interpolation

GIS Manual – December 2011 Lab 11 67

Lab 11 - Exercise Part 1: Using NMA data (Station data)
This exercise is aimed at using ArcGIS-Kriging interpolation technique to estimate and map the
spatiotemporal variation of rainfall for Ethiopia. Rainfall data, for the periods between 1950 and 2000,
are given in excel for 183 observed values across the country.
1. Open Excel, and go to your Lab 11. Open ―RF_July.xls‖. You can see that there is a latitude
(―X‖) and longitude (―Y‖) field, along with 10-58 years observed/measured annual rainfall values
for 183 locations/‖stations‖.
2. In order to bring an Excel worksheet into ArcGIS, you must save it as a .dbf file, or a .csv file
(make sure to choose ms-dos format). This is sometimes problematic because it occasionally
truncates values during the saving process.
3. Resave your file as ―Rf_July.csv‖ into your working directory.
4. Close all worksheets in Excel. You need to have the worksheet that you intend to add to ArcGIS
closed in excel in order to add correctly.
5. Open a new session in ArcGIS and add the file ―Regions‖ from your Lab 11 and the database file:
―Rf_July.csv‖ that you saved in your working directory.
6. Right click on the ―Rf_July.csv‖ file and scroll to ―Display XY
data‖. Click on ―Display XY data‖ and a window should pop
up like the window to the right.
7. Make sure that your ―X Field‖ displays longitude for the
coordinates, and your ―Y Field‖ displays latitude for your
coordinates. Press OK.

8. You may get a warning message stating that your ―Table Does Not Have Object-ID Field‖. This is
a unique identifier that ArcGIS builds into all of its shapefiles. Press OK and ArcGIS will create
this field for you.
9. Now, can you see your ―Rf_July‖ points? Where are they? Right click on the ―Rf_July.csv
Events‖ layer, and from the menu choose ―Zoom to layer‖. The ―Rf_July‖ layer should now be
visible. If the ―Regions‖ layer is not visible, please check the projection of the weather station
points (in this case follow the STEPS 10 to 18, otherwise go to STEP 19.

68 Lab 11 GIS Manual – December 2011

10. Right click on the ―Regions‖ layer and choose ―Zoom to Layer‖. What happens to the ―Rf_July
Events‖ layer? Magic, it disappeared…or is it a projection problem?

11. Go to the main tool bar, and select the ―Zoom to full Extent‖ button
12. You should now see the entire ―Regions‖ layer incomplete overlap each other. If you use the

regular zoom tool , and zoom repeatedly into these layers, you will realize that it is in fact the
―Rf_July.csv Events‖ layer. As it is in a different projection, it is unable to locate and resize itself
correctly in relation to the ―Regions‖ Layer.
13. If we take the assumption that this information was collected by GPS, then reverting to the default
coordinate system used by GPS will correct this issue.
14. So let‘s try our hypothesis! First, Right click on the ―Rf_July.csv Events‖ you select Remove.
15. Now, right click on the ―Rf_July.csv‖ layer. Left click on the ―Display XY Data‖.
16. As you can see, the coordinate system is unknown. Click on the ―Edit‖ button. In the next
window, click the ―Select‖ button and choose the following path:
Use either of the following projection according to your boundary projection
Geographic Coordinate Systems Projected Coordinate Systems
World UTM
WGS1984 WGS 1984
WGS 1984 UTM Zone 37N.prj

17. Click Add.

18. Your ―Display XY Data‖ window should now look like the graphic
to the right. Click OK

19. Your Rainfall measurement points (weather stations points) should be geographically contiguous
with your ―Regions‖ layer.
20. Your ―Rf_July.csv Events‖ is currently only a cosmetic layer. We know this because it has the
word ―Events‖ following the name. It is not yet a shapefile.
21. To create a permanent shapefile from this cosmetic layer, right click on the ―Rf_July.csv Events‖,
scroll down to ―Data‖ and select the ―Export Data‖ option.
22. Leave all the initial options as default, but make sure to save the final file to your working
directory, calling the file ―Rf_July_ETh.shp‖
23. A pop-up window will ask you if you would like to ―Add to map‖. Select OK and the new shapefile
should automatically add to the data frame.

GIS Manual – December 2011 Lab 11 69

24. Look at these data, how the available rainfall observation/measurement points are representative
and spatially distributed across the country and to be used to estimates other unknown locations.
25. Create a map with the rainfall measurement point‘s layer over the province boundary.

Creating a surface using kriging interpolation

1. Start ArcMap by double-clicking the ArcMap icon on your computer desktop. Alternately, click the

―Start‖ menu, point to ―All Programs‖, point to ―ArcGIS‖ and select ―ArcMap‖)

2. You may receive the following welcome screen, if so, select ―a new empty map,‖ and press OK. If
you do not receive this screen, ArcMap has selected a blank map by default.

3. You are now looking at the basic ArcMap screen with its various menus and tools. To begin with,
we will add rainfall data. From the ―File‖ menu, select ―Add Data‖.
From you working directory add ―Regions‖ and ―Rf_July_ETh‖ files.

4. From the standard menu click ArcToolbox window and double click the Spatial Analyst Tools
from the ArcTool Box Window and you will find a number of functionality tools and when you
double click the Interpolation tool you will find IDW, Kriging, Natural Neighbor, Spline,…Trend

70 Lab 11 GIS Manual – December 2011

5. Double click Kriging. Kriging window wizard
opens up.
6. Click the Input point features dropdown
arrow and click the rainfall dataset
Rf_July_ETh point data to use.
7. Click the Z value field dropdown arrow and
click the field you want to use in this case
use ―ANN‖. The field that holds the rainfall
magnitude value for each point.
8. In Output surface raster specify a name for
the outputs (annual_rn) to write in your
working directory, or leave the default to
create a temporary dataset.
9. Click the kriging method you want to use
(left click on the Ordinary radio button;
please refer the Kriging methods section
above).
10. Click the Semivariogram model dropdown arrow and click the model you want to use. (Read
what type of semivariogram models included in ArcGIS 10 above; but for this case use spherical-
default)
11. Change the default output cell size to 0.00833 degree which is equivalent to 1km (Since your
dataset is in lat/long the unit is if it is in degree).
12. Click the Search radius type dropdown arrow and click Variable.
Note: Uses a variable search radius in order to find a specified number of input sample points for
the interpolation.
Variable
Number of points—an integer value specifying the number of nearest input sample
points to be used to perform interpolation. The default is 12 points.
Maximum distance—Specifies the distance, in map units, by which to limit the search
for the nearest input sample points. The default value is the length of the extent's
diagonal.

Fixed
Uses a specified fixed distance within which all input
points will be used for the interpolation. Distance—
Specifies the distance as a radius within which input
sample points will be used to perform the interpolation.

The value of the radius is expressed in map units. The default

radius is five times the cell size of the output raster.
Minimum number of points—An integer defining the minimum
number of points to be used for interpolation. The default value
is 0.

GIS Manual – December 2011 Lab 11 71

13. Optionally, change the default number of points.
14. Optionally, specify a maximum distance.
15. Optionally, output variance of prediction raster.
The optional output variance of prediction raster contains the kriging variance at each output
raster cell. Assuming the kriging errors are normally distributed, there is a 95.5 percent
probability that the actual z-value at the cell is the predicted raster value, plus or minus two times
the square root of the value in the prediction raster.

16. Click the Environments tab located at the bottom of the kriging window (as shown above figure).
The Environments helps to set up Cell Size, Current Workspace, Output Coordinate System,
Extent, Scratch Workspace, and more.

17. Click the General Setting tab and click the drop down box under Extent and you will find different
extent setting options as
Default-No extent set
Union of inputs- The maximum area of all input(s)
Intersection of inputs-The minimum area common to all input(s)
As specified below (you need to specify the Left, right, bottom and top value of X and Y)
Same as display- the extent of the current ArcMap display wil be used
Same as layer <layer>-The extent will be based on the extent of the specified layer

72 Lab 11 GIS Manual – December 2011

 Output Extent environment will only process features or rasters that fall within the extent
specified in this setting and any feature or raster that passes through the specified extent will be
processed and written to output. Note that the selected extent is used only to select features, not
clip them.

18. For this particular exercise choose the Same as Regions and click Ok the Environments Setting
window

19. Click OK on Kriging window…wait till the process finishes.

Following the above steps, you are expected to create rainfall grids for the remaining months
(January-December), and also for temperature datasets.
Once you have 10-58 rainfall raster grids follow the following step to clip out the grid by the
country boundary and create a lay out for presentation.

Creating layout for presentation

20. Double click the Spatial Analyst Tools from the ArcTool Box Window and you will see a number
of tools
Double click on Extraction tool box;
Double click Extract by Mask tool box: This tool allows you to extract a portion of a
raster dataset based on a template extent.

GIS Manual – December 2011 Lab 11 73

21. On the clip window under the Input
Raster click the dropdown arrow and
select the annual_rn grid as in put to be
clipped.
22. In the Input raster or feature mask data
option click the dropdown arrow and
select the Regions boundary file as a clip
feature.
23. In the Output raster write the output
name in your working directory
(rf_clipped)

Below illustrates the clipped rainfall grid.

Repeat the above step to clip the rest of rainfall grid files.

Prepare your map for printing

1. Now you have clipped rainfall grid and regional level administrative boundary of Ethiopia. The
next step is to prepare this map for printing.
2. First, make sure that your ―Region‖ shapefile and ―RF_clipped‖ grids are all turned on by checking
the checkmark box next to the layer name in the table of contents.
3. Reorder the layers in table of content as Region and RF_clipped. If the Region boundary filled
with color it covers the underneath RF_clipped, therefore make sure that the Region boundary
should be hallow.

74 Lab 11 GIS Manual – December 2011

4. Double click on the Regions symbol and Symbol selector window will display form the options,
select Hollow.
5. The map window that you have been working in up to now is called the ―Data View‖.
We will now move to a different map window called the ―Layout View‖ to prepare the map for
printing. The ―Data View‖ is where you perform the majority of your GIS editing work, while the
―Layout View‖ is where you arrange your map(s) along with preferred graphics for printing or
export.
6. Select the layout view

7. We want to select a paper setup that best fits the

geographical nature of Ethiopia in order to use
the most space on our page for our map. Click
on the File tab in the upper left of your screen and
scroll to ―Page and Print Setup‖. Make sure you
select Landscape for the Orientation. Also click
the checkmark to ―Use Printer Paper Settings‖.
This will give you a dotted outline of how far you
are able to expand your map. (See right for
diagram)
8. Now, resize the window around your map of
Ethiopia, so that it fits within the printable area.

9. Double-click on the layer name ―Rf_clipped‖; this will take you directly to the ―Layer Properties‖ for
the layer. Select the tab ―Symbology‖.

10. In the ―Show‖ box (see graphic), click on the option ―Classified‖, and next click on the ―Classify‖
tab. This will take you to the ―Classification‖ drop down menu (see graphic below).

GIS Manual – December 2011 Lab 11 75

11. Click on the arrow to the right of ―Classes‖ and choose 12. Click on the arrow to the right of
―Method” and choose ―Manual‖.
12. Choose your ―Break
Values‖ set be 174
intervals started from 500
and type the maximum
value 2,233 mm of annual
rainfall as there is no value
more than this. (See
below)
Note: You can use your
own break value and
number of classifications
13. If you click color ramp you
will see different color
ramp options and choose
your color of interest that
best explains the data. For
this exercise choose
spectrum-full bright color
ramp. To select the color
ramp by name instead of
graphic view right click the color ramp and uncheck the graphic view. From the drop down option
find the spectrum-full bright.
Note: You can change the color property by right on the color ramp and point to properties. Edit
color ramp window lets you manipulate the color property and you can also save the style and
reuse it.
14. Click Ok on Classification and the layer properties.

15. Now, if you want to present more than one map per page you
can insert new data frame using ―Insert‖ menu and resize it
to fit page setup and map orientation.
16. The classification scheme you set for the rainfall data can be
saved as a layer file that can be used for similar use.
When you save a layer to disk, you save everything about
the layer, such as the symbolization and labeling. When you
import/add a layer file to another map, it will draw exactly as
it was saved.
One of the main features of a layer is that it can exist outside
your map as a file on disk. This makes it easy for others to
access the layers you've built.
Right click the rain98_clip layer from the table of content and
click Save As Layer File… option and type appropriate name
and save the layer in your working directory.

76 Lab 11 GIS Manual – December 2011

How to import symbolization in ArcMap
The symbology of the previous map will be imported into this map. This tool can be useful in
workflows that require much more intricately detailed symbology that would take considerable time to
recreate for each new map.
1. In the ArcMap table of contents, right-click the layer for which you
want to import symbology (saved layer).
2. Click Properties.
3. In the dialog that appears, click the Symbology tab.
4. Click the Import button.
5. If importing from another layer, choose the layer from the
dropdown list.
6. If importing from a layer file, click the Browse button and choose
the file from which you want to import. Click OK.
7. Click OK to close the Import Symbology dialog box.
8. Using the ―Insert‖ menu, add a Title, Scale Bar, and North Arrow
to your map. Using the insert title option, name your map ―Mean
Annual rainfall‖ (or something similar).
There are many Scale Bars, and North Arrows styles to choose
from. Pick your favorite.
9. Return to the menu option ―Insert‖ and click on ―Legend‖.
10. In the Legend Wizard dialog that pops up, accept all the default options by Clicking ―Next‖
repeatedly and finally ―Finish‖.
11. You will notice a small legend appear somewhere in the middle of your layout window. Position it
correctly.
12. If you wish to change the name of your layers in the legend, edit the corresponding text in the
table of contents, and it will also change in the legend. Just click twice on the text in the table of
contents to edit it (as you would to rename a file in Windows Explorer).
13. When you are happy with the appearance of your map, you can export this map as a .jpeg, .pdf,
or .png, by choosing the menu option ―File‖, scroll to ―Export‖, then save with as in working
directory.

GIS Manual – December 2011 Lab 11 77

 Annual rainfall two maps per page

Calculating rainfall coefficient of variability

The coefficient of variation calculated as the standard deviation divided by the mean is often used to
represent how rainfall is variable/ stable across the country.

1. To calculate mean: (ArcTool Box Spatial Analyst tools

Map Algebra double click on Single output Map
Algebra.
2. In the Input raster or feature data (optional) click the drop
down arrow and select all input data for the mean
calculation one by one.

78 Lab 11 GIS Manual – December 2011

3. In the Map Algebra Expression type
MEAN and create an open bracket ―(―and
from the Input raster or feature data
(optional) select one layer at a time and
drag and drop in the Map Algebra
Expression field. Separate each layer by
a comma and finally close the bracket.
4. Specify the output directory and name
output as ―MEAN_Rainfall‖ in the output
raster
5. Click Ok on to calculate the mean

Calculating Standard deviation

1. To calculate standard deviation: (ArcTool Box Spatial Analyst tools Map Algebra double
click on Single output Map Algebra.
2. In the Input raster or feature data (optional) click the drop down arrow and select all input data
for the standard deviation calculation one by one.
3. In the Map Algebra Expression type STD and create an open bracket ―(―and from the Input
raster or feature data (optional) select one layer at a time and drag and drop in the Map
Algebra Expression field. Separate each layer by a comma and finally close the bracket.
4. Specify the output directory and name output as ―STD_Rainfall‖ in the output raster
5. Click Ok on to calculate the STD

Calculating covariance or coefficient of variability

1. To calculate coefficient of variability: (ArcTool Box Spatial Analyst tools Map Algebra
double click on Single output Map Algebra.
2. In the Input raster or feature data (optional) click the drop down arrow and select
―MEAN_Rainfall‖ and ―STD_Rainfall‖ for the CV calculation.
3. Since CV is calculated as STD divided by MEAN; therefore from the Input raster or feature data
(optional) select, drag and drop the STD_rainfall and ―MEAN_Rainfall‖ layers in Map Algebra
Expression and put the division ―/‖ between them.
4. Specify the output directory and name output as ―CV_Rainfall‖ in the output raster
5. Click Ok on to calculate the CV on the raster calculator

GIS Manual – December 2011 Lab 11 79

6. Finally organize your map as shown below and give your answer for
a. Where high rainfall variability is observed?
b. What is the implication of high rainfall variability on agriculture?

Stretched values Classified values

Use of zonal statistics function in ArcGIS

Overview
Zonal Statistics are a way of summarizing the information in a raster map layer using the boundaries
of zones in a second map layer. For example, we might have a raster grid depicting the rainfall and
have a second map layer depicting Woreda/Zone/Kebele boundaries. Zonal Statistics can be used to
calculate the average rainfall in each of the administrative boundaries (woreda, kebele or zone).
These boundaries can be stored in either raster or vector format - Zonal Statistics will work with either
format for the zones.
Following parameters can be calculated (per operation, only one parameter can be calculated, to get
several parameters it is necessary to save obtained table in a separate file):
sum - sum of all values
min - minimum value of all values
max - maximum value of all values
count - number of value (number of pixels inside zone, excluding pixels with NODATA value)
area - area covered by pixels with value different than NODATA
range - values range
std - standard deviation
mean - average value (NODATA pixels, they are not used in calculation)
median - median value (NODATA pixels, they are not used in calculation)
majority
minority
variety

80 Lab 11 GIS Manual – December 2011

Using Zonal Statistics As Table
To use Zonal Statistics, make sure that you have Rf_clipped map layer open in ArcMap and another
map layer with boundaries in it (woreda boundaries). You can calculate Zonal Statistics as follows:

1. In the ArcToolBox, choose Spatial Analyst Tools, then choose Zonal

and then Zonal Statistics as Table. The difference between Zonal
Statistics and Zonal Statistics as Table is that Zonal Statistics produces
a map layer as output, whereas Zonal Statistics as Table produces a
summary table. If you are working out mean, min, max, STD, rainfall for
each woreda boundaries, Zonal Statistics as Table would produce a
table showing mean, min, max, STD for each province boundaries.
Zonal Statistics would produce the same information, but as a map
display. In practice, Zonal Statistics as Table tends to get used a little
more often.

2. Next to 'input raster or feature

zone data', you should enter the
name of the map layer that has the
zones – woreda/kebele/zone
boundaries which is the shapefile.
3. The Zone field is the name of one of
the fields in the table of attributes for
your zone map layer. To summarize
rainfall levels for different
administrative boundaries, enter a
W_NAME here, which is a unique
name stored in it for each woreda.
4. Next to 'input value raster', choose
the information that you wish to
summaries. Enter the name of the
rainfall grid map layer here
(Rf_clipped)
5. Next to 'output table', you can enter in the name of a table where your results will be stored.
6. Check the Ignore NoData in calculations (optional)
7. Next ‗Statistics type (optional)‘ you can choose a specific type of statistics to be performed, left
the default (All) as it is; but if you want only mean, you can choose MEAN.
8. If you choose OK, you should find that ArcGIS will generate a table of results for you.

GIS Manual – December 2011 Lab 11 81

9. From the table of content right click on the table and point to open (See below by regional rainfall
for July)

82 Lab 11 GIS Manual – December 2011

Lab 11 - Exercise Part 2: Using rainfall estimation (RFE)
satellite data
RFE is the rainfall estimate from satellite imageries, which is acquired by NOAA of NASA with 8km
grid size around the world. RFE is a freely available dataset.
You can download the free Rainfall estimation data from the following link:
http://earlywarning.usgs.gov/fews/africa/web/datatheme.php
You will have dekadal datasets (10 days) images from NOAA and you have 3 datasets for one month.
Hence, annually you can have 36 image files. The data format is .BIL (Band Interleaved by Line
Format). This BIL format will be converted to ArcGIS format, and the dekadal images were summed
up to develop monthly rainfall images. Finally Rainfall values can be mapped for the whole of specific
areas; and also can be compared with weather station values. To do these, follow the following steps
in ArcGIS.
1. After you acquired the required images as zipped files, you can open them as follows: ArcTool
box conversion tool to Raster Raster to Other Format (multiple).

2. Unzip the file

3. Select the file of the BIL format to
change to Grid format in ArcGIS (here
the RFE data for 2001 in BIL format,
which has 36 files for one year).
4. Click ―Add‖.

GIS Manual – December 2011 Lab 11 83

5. Specify the output directory
6. Choose the format type to ―Grid‖
7. Ok. Now, you have grid format of
these data.

8. The data file name of ea01011 means east Africa 2001, January and first decadal (January 1-10).
9. ArcToolboxspatial Analyst Tools  Map Algebra Raster calculator

10. Sum up 01011, 01012 & 01013, and give the

name for output file Jan2001.
11. Do the same for other months of the same year
and also other years.
12. Now, you have monthly Rainfall estimation for
different years.
13. You can also sum up all month images to have
annual Rainfall estimation of that year (s) as
shown in the above tool.
14. Do the same for other years to have annual rainfall estimations.
15. If you need long term (>=10 years) mean annual rainfall estimation, we can add all the annual
rainfall estimates and divided by the number of years. For example, if we have datasets from
2001 to 2010 for ten years, we can use raster calculator tool in ArcGIS and calculate mean
rainfall estimation for the last 10 years. Using the formula:
16. To calculate the Standard deviation follow the above steps!!!

84 Lab 11 GIS Manual – December 2011

17. We can also do covariance (coefficient of variance) or rainfall variability across years and space.
To do this, please follow the STEPS you used in the previous exercise.

Covariance (CV) = STD / MEAN.

18. Now, you can compare the annual rainfall either for long term or for annual rainfall estimation from
satellite data and weather station data. To do this, you can pick the points of RFE data by taking
sample statistics from which weather stations are found.
19. Locate the mean annual rainfall
from RFE in the input raster, and
also locate Weather station
points as input point feature, and
specify the output table, and click
OK.
20. Link the output table to the feature
points by creating common Id
(RowId for the new table, and
new Id by calculating FID+1 for
the point feature ID).
21. Compare the figures obtained
from the RFE and weather
stations; and reason out if there
are differences on the same
point.

GIS Manual – December 2011 Lab 11 85

Lab 12: Agriculture survey data integration
Exercise overview
You have received area by crop production information from the Central Statistics Agency of Ethiopia
for the years 2007, 2008, 2009 & 2011. The data is in Microsoft Word format, and you need to
integrate information on area covered by cereal into ArcMap to create a thematic map.
In order to do this you first need to extract the data of area coved by cereal and copy the same to
Microsoft Excel. Then after you need to join this table to the attribute of a map table as discussed in
Lab 07.
1. From Lab12 folder open the Word document titled ―ProductionReport2006-07‖. This table
contains data obtained from the 2006/07 Agricultural Sample Survey conducted by CSA. It lists
crop type, number of holders, amount of area coved by corresponding crops, production in
quintals, and yield of zones covered by the 2006/07 Agricultural Sample Survey. For this
exercise your concern is only area coverage of cereals by zone.
2. Open an empty Excel sheet, write three headings ―Zone_Code‖, ―Zone_Name‖, ―Cereal_Area‖
and save it as ―Cereals‖ in Lab12 folder.
3. Starting from Table 4.1 extract and copy zone code and name (written at the top of each table).
Also type the area covered by cereals which is indicated under the column heading ―Area in
hectare‖ in the ―ProductionReport2006-07‖ to the Excel sheet you just created,.
4. Now you have an Excel sheet that lists the total area covered by cereals at zone level.
 REMEMBER: ArcMap doesn‘t allow space in field names, hence notice that the field names in
your Excel are edited accordingly.

5. Open ―Zonal_cereal_2007‖ excel sheet from your Lab12 folder.

6. Now, you need to join this information with the attribute of a map table.
7. To join your Excel table to the shapefile, as you practiced in Lab 07, start by saving the
―Zonal_cereal_2007‖ table into a .csv file format so that it can be imported into ArcGIS (make
sure none of your fields were truncated during the save). Name the .csv file
―Zonal_cereal_2007‖.
8. Open a new session of ArcGIS and add the layer ―Zone_Cereal_Area.shp‖ from your Lab12
folder.
9. Add your database file ―zonal_cereal_2007.csv‖ from Lab12 folder.
10. Right-click on your ―zonal_cereal_2007.csv‖ table to open it. Make sure that all field names
carried through correctly, and that all data appears in working order. If it looks good, close the
attribute table and move on!
11. The next step is to join the ―zonal_cereal_2007.csv‖ table to the ―Zone_Cereal_Area‖ layer so that
you can utilize its spatial properties to visualize the area covered by cereals.

86 Lab 12 GIS Manual – December 2011

12. Right click on the
―Zone_Cereal_Area‖ layer, Scroll to
―Join & Relates‖ > ―Join…‖ Make the
following dropdown selections, and
click OK.

13. When complete, open the attribute table of the ―Zone_Cereal_Area‖ layer to make sure the join
was successful. As you will see, some of the fields will say <null>, this is okay for this specific
join because data were not collected for these specific zones.
14. Before continuing, save your map to your Lab12 folder as ―Lab12_YOURNAME‖.
15. Double click on ―Zone_Cereal_Area‖. The ―Layer Properties‖ Window should now pop-up. If the
tab ―Symbology‖ is not selected, then select that tab.

GIS Manual – December 2011 Lab 12 87

16. In the box labeled ―Show‖, select the option ―Quantities‖, and click on the sub-option ―Graduated
Colors‖.
17. In the drop down menu ―Value:‖ scroll to the the field named ―Cereal_07‖: This stands for total
area covered by cereals in the indicated zones in 2007.
18. Once this field has been selected, choose a color ramp that you like by clicking on the down
arrow next to the color ramp. Now click, Apply, and OK.
19. Look at how the colors are distribued. The classification brackets
chosen by ArcMap are based on their default statistical
classification; ―Natural Breaks‖.

20. Reopen the ―Layer Properties‖ dialog for the ―Zone_Cereal_Area‖ layer. Return to the symbology
tab. Under the Classification menu (top right-hand corner), Click the ―Classify‖ button.
21. In the ―Classificaton Wizard‖ you will see a histogram illustrating the data distrubution along the
number line.
22. In the ―Method‖ drop down list, you will see several classification alernatives to the ―Natural
Breaks‖ system. You also have the opportunity to change the number of classes that you use.
23. Experiment with the different classification schemes, and look at how they alter the classification
breaks (blue lines) on the histogram data. By clicking OK on both wizards, you will see the effect
of your class scheme changes on the map itself.
24. Return to the ―Classification Wizard‖ screen. Choose first the ―Quantiles‖ scheme; next change
the number of classes to 7.
25. In the ―Break Values‖ box to the right hand side of the wizard, set the
break values to the following numbers, by simply typing over the
existing values.

26. When done, click OK. Now you are back at the ―Symbology‖ window. The ―Label‖ side of the
menu will reflect the changes that you make to the ―Range‖ side, but you may also use text in
your labels.
27. There are a number of Zones with ―Null‖ values. (You should always account for these in your
mapping). Given that we have only done a Table Join, and we haven‘t exported our data as a
new shapefile, our values reflect ―Null‖, we will show these by adding another layer.

88 Lab 12 GIS Manual – December 2011

28. Copy and paste the ―Zone_Cereal_Area‖ layer. Organize your ―Table of Contents‖ so they look
like the graphic below. Choose the underlying Zone_Cereal_Area layer to be a grey color to
reflect ―Null‖ data. Also, go to the table of contents and double click on this layer you just copied
and rename it ―Null‖.

29. Now return back to your

―Zone_Cereal_Area‖ layer with
the table join. Double click on
the layer and return to the
Symbology tab.
30. Modify the label options with
some additional text, and
commas by clicking on the value
under the Label column.

31. For a softer more subtle style, you will remove the
boundaries from between the individual zones. Click on the
word ―Symbol‖ above the colored category symbols, and in
the pop-up menu, choose ―Properties for all symbols‖. (see
right)

32. In the ―Symbol Selector‖ dialog, change the outline color to ―No color‖.
33. Click OK.
34. To distinguish the boundaries of the higher order administrative units, add the ―Regions‖ layer
from the Lab01 folder, and symbolize as hollow with an appropriate outline thickness.
Remember how to do this? (hint: double click on colored box symbol below the layer name in
the Table of Contents)

GIS Manual – December 2011 Lab 12 89

35. Switch to the layout view, by clicking on the layout symbol in the bottom left and corner of the map

window.

 Save your map if you haven’t recently!

36. In the layout window, resize the map, so that it is roughly one-fourth the size of the page by
clicking once on the map and then dragging one of the corner sizing squares to diminish the size
of the map.
37. Insert three additional data frames. Resize all the frames so that your layout window should look
something like this:

38. Right click on a data frame and go to

Properties. Under the Frame tab, set
Border to <None> as indicated in the figure
below. Do this for all four data frames. This
would remove the line around the data
frame.

90 Lab 12 GIS Manual – December 2011

39. Copy all three layers (named ―Region‖, ―Zone_Cereal_Area‖ and ―No data‖) under the top most
data frame and paste into the remaining three data frames (namely,
New Data Frame,
New Data Frame 2, and
New Data Frame 3.

40. Now in all four data frames you should have map portraying the total area covered by cereals in
2007, shown in the figure below.

41. To show which year‘s cereals coverage data is shown in which map
window, go to ―Insert‖ menu and click on ―Text‖ as indicated beside.

GIS Manual – December 2011 Lab 12 91

42. A small box that have a ―Text‖ inside will appear somewhere in your map. Overwrite ―Text‖ with
2007 and place the text box in the top-left side of the first map window. Insert three additional
―Text‖ boxes and write 2008, 2009, and 2011 inside these boxes. Place each boxes as indicated
in the figure below. By looking at the years, now we can tell which map belongs to which year
cereal coverage data.

43. What we want to do is to integrate and show four years (2007, 2008, 2009, & 2011) of zonal
tabular data about the area covered by cereals. However, all four maps in the above figure show
the total area covered by cereals in 2007. To compare change in area covered by cereals over
the years, you need to join 2008, 2009, & 2011 zonal tabular data of area covered by cereals to
―Zone_Cereal_Area‖ shapefile.
44. In the Lab12 folder you would find ―Zonal_Cereal_2008‖, ―Zonal_Cereal_2009‖, &
―Zonal_Cereal_2011‖ Excel files. These Excels contain cereal coverage data for 2008, 2009, &
2011, respectively.
45. Following the above practiced steps you followed to join 2007 cereal area coverage data, save
each of these Excels in .csv format and join each table with ―Zone_Cereal_Area‖ in each of the
three data frames.
―Zonal_Cereal_2008.csv‖ should be joined with ―Zone_Cereal_Area‖ beside 2008;
―Zonal_Cereal_2009.csv‖ should be joined with ―Zone_Cereal_Area‖ beside 2009;
―Zonal_Cereal_2011.csv‖ should be joined with ―Zone_Cereal_Area‖ beside 2011.
46. Using the ―Insert‖ menu, add a legend, and other appropriate cartographic elements to your map
(North Arrow, and simple scale bar).
47. Based on your expertise, did the maps show recognizable patterns? Did the zones you know to
have large cereal coverage stand out in these maps? What about zones you know to have low
area covered by cereals? Are there zones for which you think the data needs re-check?

92 Lab 12 GIS Manual – December 2011