Module 3: GIS Data Preparation

3-1 Format Conversion

The default formats of data for various GIS software are usually different from each
other. Therefore, data you obtained from different sources usually have to be converted
to the format that matches your GIS application.

E00 is one of the most popular formats in GIS worldwide. The majority of GIS
software has a function to input and convert the *.E00 file format. Particularly when
you download free GIS data via the Internet, it would be very likely that you would
obtain an E00 file. In this section, you will become familiar with how to convert a
*.E00 file to a shape file.

Exercise 1: Adding E00 Data and Change to Feature Class

1. Start Explorer > browse to the folder EthiopiaArcGIS/Data/CountriesData/ETH

> find the file DNNET.E00.gz > decompress it.

This data was freely downloaded from one of the most well-known GIS websites,
GeoComunity. From GeoComunity, you can download country-wide data such as
administrative boundaries, rivers, roads and others, for any countries throughout the
world. The address of the website is as follows:

http://www.geocomm.com/

2. Start ArcMap > open a new map > set the coordinate system of the data frame to
WGS_1984 >click the Add Data button > browse to the same folder you have
decompressed the
E00 file > find the
decompressed E00
file > click the Add
button.

3-1
 You will find that compressed E00 files will also appear in the Add Data window
to show that they are recognized by ArcGIS, but if you try to add them before
decompression you will encounter an error.

3. Not only one layer, but a layer set will be added to

your data frame. The layer set consists of 4 layers
which you can check one by one to confirm the
extension of each layer corresponds to the data type.

- point: point feature

- arc: polyline feature (in this data set, shows the

river route stretch)

- Text: a combination of labels

- poly: polygon feature (in this data set, shows lakes and very wide part of rivers)

You can use any of these together with other data to create your map. For example,
you can add the feature of Eth to create a river distribution map in Ethiopia.

4. Add the feature of Eth which you created in Assignment 3. Set the symbology as no
color for fill and red for outline. Then you will find the river route stretch and lakes
do not match the boundary of Ethiopia well; in the northern part are rivers and lakes
which lie outside of Ethiopia.

5. To make the river route stretch and lakes match the country’s boundary, open
ArcToolbox > Analysis Tools > Extract > Clip. Then, using the dropdown menu,
set the Input Features to the layer DNNET_arc, and Eth for Clip Features. Next,
set the Output location to your Result folder with the Output name as eth_river.

3-2
6. Using the same procedure as in Step 5, you can clip the DNNET_poly by Eth to
obtain a polygon feature of Eth_lake.

Note: the point layer is useless in this exercise and the text layer can not be
converted to a shape file.

7. If you would like to perform any editing, you may, for example, express all river
routes as only one line no matter how big the river width is. To do this, Start

Editing by clicking the Edit Tool button then click Start Editing in the
dropdown menu for Editor.

Notice at this moment that it is not possible to edit all layers in your data frame,
only Eth and those clipped features can be selected from the Target layer list.
Therefore, after you add E00 data into your data frame, it is recommended to export
the necessary layers to your Result folder so that these data will be more
conveniently used and also be editable.

Exercise 2: Using Global Mapper (optional)

GIS software “Global Mapper” (USGS) has a very powerful format conversion
function. Not only *.E00 files, but many other common GIS data formats can also be
read and then converted to other formats to meet your needs. If you convert a *.E00 file
with ArcGIS, you need to decompress the source data first because, generally, *.E00
data is distributed in a compressed format, especially when distributed over the Internet.
However, if you use Global Mapper to perform the conversion, it’s unnecessary to
decompress the source data, which allows you to keep your database light and simplify
the tasks you must perform.

1. Start “Global Mapper” from

the Start menu > All
Programs. In Global Mapper,
click the File menu > click
Open Data File(s).

You can check the File Type list to find how many formats are supported, and
consequently, all supported formats can be converted by Global Mapper.

3-3
2. In the “Open File(s)” window, select the file DNNET.E00.gz from the following
folder:

C:/EthiopiaArcGIS/Data/CountriesData/Ethiopia.

When you use Global Mapper for format conversion you don’t need to decompress
the original file.

3. If the file you select to open does not have a coordinate system indicator, you will
be requested to select projection/datum. At this point the only option is to simply
click OK button, so do so.

4. In the following window to Select

Projection for Dnnet.e00 Datum,
enter the following settings before
you click the OK button:

- Select Geographic
(Latitude/Longitude) in the
textbox for Projection

- Select WGS84 in the textbox

for Datum

- Check that the default setting

of ARC DEGREES is set for
Planar Units

- Make sure the box is checked for “Use Selected Projection for All Selected
Files”

- Click OK

5. All the contents within the compressed data file DNNET.E00.gz will appear in the
window of Global Mapper.

3-4
6. In order to convert the *.E00 file to a commonly used format, i.e. shape file, you
can use the export function of Global Mapper. Click the File menu > click Export
Vector Data > choose the export type by clicking Export Shapefile.

A "Tip" may appear. Click OK to go to the next window.

7. In the Shapefile Export Options window, check the box for Export Areas and
then in the following Save As window, set the necessary parameters as:

- Location: C:/GIS_Exercise/Database/My_data

- File name: Eth_lake

8. For Export Lines, repeat the

same process as the last step.
Set the output in the same
location, but this time set the
file name as Eth_river.

9. In the lower half of the window,

check the first box (generate
PRJ) and the third (Add
LAYER) box and click the OK
button.

Assignment 4: Create a Base Map for Your Country

1. In this section, you will convert an E00 file to a shapefile or coverage file to get
rivers, lakes roads and railways in your country or region.

Please find the relative E00 file from the folder:

C:/EthiopiaArcGIS/Data/CountriesData/your country name

3-5
 The contents of E00 files are as follows:

DNNET : rivers and lakes

RDLINE : roads

RRLINE : railway

2. Create a background of your country by using the country boundary file that you
created in Assignment 3.
You are free to set the color for fill and outline as you like. However, please avoid
the setting of “no color”. Also, it is suggested that a dark color be set for the outline
and a light color for fill.
Set the width of the outline as 2.
3. Add layers above the background and set their symbology as the following
sequence:
- road (top layer, dark brown, width 1)

- railway (second layer, black, width 1)

- river (third layer, light blue, width 1)

- lake (forth layer, dark blue for fill, the same dark or no color for outline
and the width of the outline not over 1)

4. Set your map to take the full extent of Data View.

5. Export the map in .jpg format to the folder C:/GIS_Exercise/Result and adjust the
resolution setting to make pixels for both width and height greater than 1000. Name
the jpg file to be exported as follows:
Yourlastname_E4 (for example, Lei_E4.jpg).
6. Upload your image result to the folder with your name on the server computer.

3-6
3-2: Coordinate System Conversion
Exercise 1: Understand the Difference of Coordinate Systems

Geographic and UTM

The survey area of the EWTEC project can be expressed by different coordinate
systems according to different purposes. The two kinds of basic coordinate systems are
Geographic (Longitude/Latitude) and UTM (Universal Transverse Mercator). Within
this exercise, you will learn the basic differences between them.

1. Start ArcMap, open a new map, save it as Prj_check in the Maps folder

2. Add the feature SA_prj1 from the GeoDB OtherData

When you add features into the map, a dialog box will appear to inform you if the
feature is missing coordinate system (Spatial reference) information. At this point
just click the OK button and continue with the exercise.

3. Add another feature, SA_UTM from the same GeoDB.

4. Change the symbology of feature SA_UTM to clearly differentiate its expression

from feature SA_prj1.

5. Select the layer SA_prj1, then zoom in to it (right-click the layer and then click
Zoom to Layer). Check the coordinate values in task bar below. You will find
nothing is strange. The coordinates of the study area appear to be properly
expressed in longitude and latitude.

6. Select the layer SA_UTM and also zoom in to it. This time you will find the
coordinate system shows thoroughly strange values. Obviously the values shown in
the task bar are anything but longitude and latitude.

3-7
 Actually the unit in UTM projection is "meter". This unit is necessary in case you
would like to get an accurate calculation in length of lines and areas of polygons.

7. Click the Full Extent tool try to see both layers at the same time. You will find
nothing could be seen in the Data View. That is, you can not express the two layers
properly within the same Data View without processing the necessary coordinate
systems.

Differences between Various Geographic Coordinate Systems

1. Continue the above exercise. Make the layer of SA_UTM invisible.

2. Zoom To Layer SA_prj1, then add all features named SA_prj# from the GeoDB:
C:/GIS_Exercise/Database/OtherData.mdb

3. Select the fill color for all the layers as No Color, and then set a unique outline
color for each layer so that you can easily distinguish between them. All the layers
within the Data View will be seen as shown in the left side of the figure below.

All layers give a similar expression of the study area of EWTEC, but each layer is a
little different in spatial extent from the other layers.

4. To see clearly the difference between layers, you can zoom in to any part in the
Data View, e.g. upper right corner, as shown in the right side of the figure below.

5. Set the coordinate system of the data frame to WGS_1984.

3-8
6. Open the Data Frame Properties window, if it was closed after the coordinate
system setting >
select the
General tab >
set the Units of
Display as
Meters. Then
click the OK
button.

7. Click the Measure tool > measure the difference between layers. You should
find the difference for the same point in different layers will be in the range of
about 50 to 1000 m.

8. Save the map.

Exercise 2: Area Calculation in Different Coordinate Systems

1. Continue Exercise 1. Click the Insert Menu > click Data Frame. A new data frame
is added into the table of contents with the name New Data Frame. At the same
time, you will find that all layers have disappeared from the Data View.

The default setting dictate that when new data is added, the new data frame will be
the active data frame. Thus, from the Data View, you can only see the layers in the
new data frame.

2. Rename the new data frame as UTM.

3. Go back to the old data frame and rename that frame as Geographic.

4. Select the layer SA_UTM from data frame of Geographic > right-click the layer >
click Copy.

5. Right the data frame of UTM > click Paste Layer(s). The layer SA_UTM will be
pasted into the new data frame and is shown in the Data View.

6. Remove the SA_UTM layer from the old data frame (now titled Geographic).

7. Add the original data of shape file SA_UTM from:

3-9
 C:/EthiopiaArcGIS/Data/OtherData/Prj_E

into the data frame of UTM.

The two layers in the Data View are expected to have exactly the same shapes and
spatial extents.

8. Because the two layers have the same name, it is recommended that you change the
expression of the table of contents to Source View to better be able to distinguish
between them.

You can alternatively change the name of one or both layers to make them easily
distinguishable.

9. Open the attribute table of the original data SA_UTM, which is from the folder
C:/EthiopiaArcGIS/Data/OtherData/Prj_E. You will find there are three (3)
fields: FID, Shape, and NAME.

You will also find an asterisk attached to the Shape field, this asterisk indicates that
the value in this field can not be edited.

10. Open the attribute table of the other layer that is from your GeoDB OtherData.
You will find that five (5) fields are stored there. The first three fields correspond to
the three fields of the original data. The last 2 fields, Shape_length and
Shape_Area, give the length and area in the coordinate system unit and are
automatically added into the attribute table when you import the feature into the
GeoDB.

Therefore, if you find the default method of area calculation is troublesome, you
can simply import the feature into a GeoDB, after you perform the appropriate
coordinate system conversion.

11. You may decide the length of the polygon is not so important, but the area would be
very important data in many cases for analysis, so a reliable method is necessary for
area calculation.

In this exercise you can see from the attribute data that the area of the study area of
EWTEC is 6,759,590,300m2.

12. To calculate area by the default method of ArcGIS, you must follow the steps
below:

3-10
- Open the attribute table of a polygon feature layer for which you would
like to calculate the area. In this case it is SA_UTM from

C:/EthiopiaArcGIS/Data/OtherData/Prj_E.

- Click the Options button > click Add Field.... The Add Field dialog box
will appear.

- Set the Name of the new field as Area, Type as Float, Precision as 20 and
Scale as 2, then click the OK button.

- Right-click the title of the new field Area > click Calculate Values... and
click Yes in the following Field Calculator message box. The Field
Calculator window will appear

- Check the box for Advanced

- Input the following statements to the textbox of Pre-Logic VBS Script

Code

Dim dblArea as double

Dim pArea as IArea
Set pArea = [shape]
dblArea = pArea.area

- Input the following statement to the test box of Area=

dblArea

- Click the OK button.

- The calculation result is shown in the attribute table and the value is also
6,759,590,300m2.

If you find it difficult to remember the above statements for area calculation,
you can access them through the following procedure:

• After you get to the Field Calculator window, go to ArcCatalog and

click the Help menu.

• Click ArcGIS Desktop Help > click the Key Word tab > input area
as the key word > select calculating for polygons > click the Display
button > scroll to the section of Updating area for a Shapefile

3-11
 • You will find all the statements there. You can use the help menu as a
reference to input the statements yourself, or you can even select the
statements > right-click the selection area > click Copy > go back to
the Field Calculator window > put your curser in the corresponding
text box > right-click > click Paste.

Now go back to the table of contents and make sure it is in expression of Display.
Then go to the data frame of Geographic.

13. Select one of the layers with the geographic coordinate system e.g. SA_prj1. Open
its attribute table. You can find the two fields Shape_length and Shape_Area have
also been added into the attribute table when you imported them into GeoDB.
However, the values of both length and area are nonsense due to the fact that the
calculation is based on the unit of the coordinate system, that is, the decimal
expression of longitude and latitude.

14. You can try to use the procedure given in Step 12 to calculate the area once again,
and you will find the result is identical to that shown in the current attribute table.

In performing the above exercise, you are requested to keep in mind the following
very essential concepts of coordinate system:

- Maps or GIS data are created in different coordinate systems according to

their different purposes.

- You can not make GIS data match each other by simply adding them into
the same data frame without performing the necessary coordinate system
processing.

3-12
 - Coordinate systems can be generally classified into two basic types:
Geographic and Projected.

- The geographic coordinate system uses longitude and latitude as a map

unit, and the most widely used geographic projection is WGS1984 (World
Geographic System 1984).

- The projected coordinate system generally uses meters as map unit. Also,
the most widely used projected coordinate system is UTM with differing
zones for countries or even areas.

- Geographic coordinate systems are suitable to express data covering a vast

extent, for example, the African continent. But the geographic coordinate
systems are not suitable for area calculation because the units are longitude
and latitude. Therefore, if area information for polygon features is required,
you have to convert the coordinate system of your GIS data from a
geographic to a projected system, such as UTM.

- For more information about coordinate systems, refer to the reference of

this text book.

Exercise 3: Check and Define Coordinate Systems

When you obtain GIS data, you have to check the projection (coordinate system) by
which the data has been created. In most cases the projection data is stored in metadata,
that is, the information about the GIS data.

In the above exercises (Exercise 1 and 2), 7 features were used to express the same
location of the EWTEC study area. These 7 features are created in different projections
as follows:

- SA_prj1.shp : WGS1984 (world wide)

- SA_prj2.shp : D.Abidjan 1987 (Ethiopia)

- SA_prj3.shp : Leidon (Ghana)

- SA_prj4.shp : Minna (Nigeria)

- SA_prj5.shp: Schwarzeek (Namibia)

- SA_prj6.shp: Point Noire (Congo)

3-13
 - SA_UTM.shp : UTM Zone37N (world wide)

Set a WGS 1984 UTM Coordinate System Using ArcCatalog

1. Close ArcMap if it is still open.

2. Open ArcCatalog and browse to GeoDB C:/GIS_Exercise/Database/OtherData

where you stored the data for the previous exercises.

3. Right-click the feature class of SA_UTM, and then click Properties.... The
Feature Class Properties window will appear.

4. Click the Fields tab > click the Data Type of Geometry corresponding to the Field
Name of Shape. The Field Properties will appear in the corresponding list.

5. You will find the property is Unknown from the bottom row of Spatial Reference.

6. Because the coordinate system of the feature SA_UTM is UTM WGS1984 zone
37N, you can set the coordinate system by clicking the button next to the
property item. The Spatial Reference Property window will appear. The
Coordinate System tab will be selected as the default setting.

7. Now the Name of the coordinate system is Unknown, so you have to set it by
clicking one of the buttons in the lower part of the window. Because most

3-14
 coordinate systems have been saved in the coordinate system list of ArcGIS, the
most convenient method to set the coordinate system is by clicking the Select...
button.

8. The Browse for Coordinate System window will appear, which you have
probably encountered several times in previous exercises. Expand the folder of
Projected Coordinate System > expand the folder of Utm > expand the folder of
Wgs 1984. The UTM WGS1984 zone list will appear.

9. Slide the scroll button to find the destination zone WGS 1984 UTM Zone 37N.prj
and then click the Add button.

10. When you go back to the Fields tab of the Feature Class Properties window you
can check again if you have correctly selected the coordinate system from the Field
Properties partition, and then click the OK button.

11. When you go back to the Coordinate System tab of the Spatial Reference
Property window you still have a chance to check if you have correctly selected
the coordinate system from the Name textbox, and then click the OK button.

12. Open ArcMap and open the map of prj_check that you created in the previous
exercise.

13. If the data frame of Geographic is inactive, make it active (so it is current showing
in Data View) by right-clicking on it and then clicking Activate.

14. Make all layers invisible except the layer SA_prj1, because you already know the
feature is created in the same coordinate system of WGS1984 that you set for the
dataset in Exercise1.

3-15
15. Adjust the symbology of the SA_prj1 layer to no color for fill color and red for
outline color.

16. Add the feature AS_UTM from GeoDB C:/GIS_Exercise/Database/OtherData

for which you set the coordinate system as UTM WGS1984 zone 37N in the prior
steps of this exercise.

17. This time you will find the layer AS_UTM can be shown in the same Data View
and exactly matches the layer of AS_prj1.

In this exercise you will have learned that if your GIS data are created in widely used
coordinate systems such as WGS84 and WGS84 UTM, and you have set the
coordinate system of those GIS data and the data frame, then the data may be
automatically matched to each other in the coordinate system of the data frame.

Set a Geographic (Longitude/ Latitude) Projection Using ArcMap

1. Start ArcMap, open the map of prj_check. Make the data frame of Geographic
active if it is inactive.

2. Display the Arc Toolbox window if it is hidden.

3. Expand the menu of Data Management Tools > expand Projections And
Transformations > open the Define Projection tool. The Define Projection
window will appear.

4. Select the SA_prj1 layer into the textbox of Input Dataset or Feature Class

5. Click the coordinate system selection button next to the Coordinate System

textbox , and the same Spatial Reference Properties window will appear.

6. Refer to the procedure you performed in the previous exercise, select Geographic
Coordinate System > World > WGS 1984.

3-16
7. Activate the data frame UTM.

8. Remove the layer that is from C:/EthiopiaArcGIS/Data/OtherData/Prj_E. Then

you will have only one layer of SA_UTM in the data frame UTM and the layer is
from the GeoDB C:/GIS_Exercise/Database/OtherData.

9. Adjust the symbology of the layer SA_UTM as no color for Fill color and red for
Outline.

10. Set the coordinate system of the data frame as WGS_1984_UTM_Zone_37N,

which is the same projection as the layer of SA_UTM.

11. Add the feature AS_prj1 from GeoDB C:/GIS_Exercise/Database/OtherData for

which you set the coordinate system in the previous exercise.

12. This time you will find the layer AS_prj1 can be shown in the same Data View
and exactly matches the layer of AS_UTM in UTM projection and map unit of
meters.

13. Right-click the layer of AS_prj1 > click Data > click Export Data. The Export
Data dialog box will appear.

14. Select the radio button of the data

frame under the Use the same
coordinate system as > set the
Output shape file or feature class
as your Temp folder > click the OK
button.

3-17
15. When the exported feature Temp is added into the same data frame UTM, you can
check its coordinate system by opening the Properties > click the Source tab. You
will find the coordinate system has been changed to
WGS_1984_UTM_Zone_37N.

In this exercise you will have learned that ArcGIS provides several methods for
coordinate systems conversion and one of them is to add a feature to a data frame
with a different data frame, then export the feature with the coordinate system of
the data frame. But you will also learn in later exercises that this method is not
always effective. For many coordinate systems which are not as common as
WGS1984 or UTM, this method is often ineffective.

Set Geographic System for Country Wide Projection

1. Repeat Steps 1 to 6 in the last exercise. However, this time use SA_prj2 as the
target feature class.

2. After you select the folder Geographic Coordinate System > double click the
Africa folder. The list of projections generally used in Africa will appear.

3. Since you already know the feature SA_prj2 is created in the coordinate system of
Abidjan 1987, you can find the corresponding prj file from the list, which will be
the first one. Select the Abidjan 1987.prj and then click the Add button.

4. If the data frame of Geographic is not active, the activate it.

5. Make all layers invisible except the layers SA_prj1 and SA_prj2 since you have

3-18
 finished defining the coordinate systems for these two features. However, you will
find that the two layers still do not match each other even after their coordinate
system has been defined. In order to make SA_prj2 (in the coordinate system of
Abidjan 1987.prj) match SA_prj1 (in the coordinate system of WGS84) you have
to perform a project conversion process.

In this exercise you will practice converting the coordinate system of SA_prj2 from
Abidjan 1987.prj to WGS84.

6. Display the Arc Toolbox window if it is hidden.

7. Expand the menu for Data Management Tools > expand Projections And
Transformations > expand Feature > select and open the Project tool. The
Project tool window will appear.

8. Select the layer SA_prj2 in the textbox for Input Dataset or Feature Class

- set the Output Dataset or Feature Class as:

C:/GIS_Exercise/Database/OtherData.mdb/SA_prj2wgs84
- set the Output Coordinate System as WGS_1984, which is the same as
the feature of SA_prj1
- The geographic transformation (optional) will be marked by a green circle,
which indicates that you can not omit this parameter specification. Click
the dropdown arrow next to the corresponding textbox. You will find only
one selection for this parameter. Click the selection
“Abidjan_1987_To_WGS_1984”, and it will be added into the
transformation list.
- Click the OK button.
9. After the conversion is complete, the converted feature SA_prj2wgs84 will be
added to the current data frame of Geographic. Then you can find the two layers
exactly match each other within the coordinate system of WGS 1984.

If you would like to practices more to master coordinate systems processing, including
define and conversion, you can practice by converting the other 4 features referring to
the coordinate system information given in the beginning of Exercise 3.

When you perform the coordinate system conversion, you might become confused how
to select the parameter for geographic transformation (optional). Here are some hints
for that step:

3-19
 - SA_prj4 : Minna_To_WGS_1984_2

- SA_prj5 : Schwarzeck_To_WGS_1984

- SA_prj6 : Point_Noire_To_WGS_1984_2

Reference: Fundamentals of Map Projection

Surface of Earth

The world we live in is something of a globe with a

radius of about 6,378 km. Small areas on the Earth
are nearly flat, but for drawing a map of the whole
world, or even a whole country, you have to decide
how to represent the spherical-like relief world on flat
paper.

Mean Sea Level

In order to create a map we have to carry out computations of positions, distances,

directions, etc. on the Earth's surface, and therefore we need to have a mathematical
reference frame. However, the real earth surface can not be accurately described with
any reasonably simple mathematical model because the surface of the Earth is irregular,
complex and constantly changing.

What a geodesist can do is to find a substitute surface or a mathematical model that can
approximate the Earth's surface accurately enough to create a map. The simplest model
of the Earth is a sphere, but the accuracy of this model is usually not high enough
except when making a small scale map, such as 1/1,000,000. A more complicated model
is the geoid, an idealization by expanding the oceans to stretch below the landmasses
and making the assumption that the water there can also flow freely.

If we then neglect tidal and current effects on this "global ocean", a geoid was expected
to be a smooth, regular shape, however, it isn't. Water surface is affected by gravity,
while the direction of gravity depends on the mass distribution inside the Earth. Due to
irregularities or mass variation in distribution, the "global ocean" is forced to be an
undulated surface as shown in the following concept diagram. Consequently, the geoid
is not an analytic surface and it is thereby not suitable as a reference surface to

3-20
determine locations or create maps.

Perspective view of the Geoid

The Earth Model (Ellipsoid)

On the other hand, it is known that if the Earth was of uniform density and the Earth's
topography didn't exist, the geoid would have the shape of an ellipsoid.

The following figure offers the concept of the real Earth surface, geoid surface and
ellipsoid surface. The geoid surface is far smoother than the real surface; however, the
deviation of the geoid to the ellipsoid surface can be as large as +/- 100 meters. The
deviation between the geoid and an ellipsoid is called the Geoid undulation.

Because the ellipsoid exists only in theory and not in real life, it is completely smooth
and does not take into account any irregularities, such as mountains or valleys.

3-21
The shape of an ellipsoid is usually defined by its semi-major axis a and semi-minor
axis b; or semi-major axis a and flattening f. Flattening f is defined by the following
formula.

f = (a - b) / a
Datums

As mentioned above, the ellipsoid model of the Earth is based on the geoid and would
deviate from the geoid in various locations. Therefore, the reference ellipsoid used for
various regions of the Earth are usually determined by the local geoid as shown in the
following figure.

A national reference ellipsoid is defined by a Geodetic Datum, which consists of two

parts:
a) A set of ellipsoid parameters (a, b or a, f).

3-22
 b) A defined orientation, position and scale of the geodetic system.
Datums are the basis for all geodetic survey work. They act as reference points in the
same way that starting points do when you give someone directions. For instance, when
you want to tell someone how to get to your house, you give them a starting point that
they know, like a road or a building.

Main ellipsoids

Name Date a(m) b(m) Use

Everest 1830 6377276 6356079 India, Burma, Sri Lanka
Besset 1841 6377397 6356079 Central Europe, Indonesia
Airy 1849 6377563 6356257 Great Britain
Clarke 1866 6378206 6356584 North America
Clarke 1880 6378249 6356515 France, Africa (parts)
Helmert 1907 6378200 6356818 Africa (part)
Krasovsky 1940 6378245 6356863 Russia, Eastern Europe
GRS80 1980 6378137 6356863 North America
WGS84 1984 6377137 6356752 World (GPS measurements)

The coordinates of a point in the Earth will differ when different datums are used to
define the location of the point. If we plot these coordinates for the same point, for
example, the location of the Texas Capital Dome, as shown in the following figure, the
position shift from datum differences would be as large as about 1km.

3-23
The following figure gives the difference expressions of the Ethiopian boundary from
different datums. A base map uses the datum of WGS84, while other datums are either
used in Ethiopia or neighboring countries. The location shift between different datums
can be as much as 300m or more.

Geographic Coordinates

Based on an ellipse model, locations

on the Earth can be specified using a
spherical coordinate system. The
geographic (“earth-mapping”)
coordinate system is aligned with the
spin axis of the Earth. It defines two
angles measured from the center of
the Earth. One angle, called the
Latitude, measures the angle
between any point and the Equator.
The other angle, called the
Longitude, measures the angle
between the point and the base line
meridian, Greenwich, England.

3-24
Projection

The term 'projection' comes from the notion of placing a light source inside a
transparent globe and projecting shadows of the meridians (longitude), parallels
(latitude) and other geographic features onto a sheet of paper. Mathematically speaking,
map projections are
transformations of geographic
coordinates (latitude, longitude)
into the Cartesian (x, y)
coordinates of a map.

Light Source

Different projections can be produced by changing the position of the light source.
Gnomonic projections place the light source at the centre of the globe, Stereographic
projections place the light source at the antipode of the point of tangency, and
Orthographic projections place the light source an infinite distance from the point of
tangency, resulting in parallel light rays.

3-25
Projection Surface

Map projections can be thought of as being drawn on one of three developable surfaces
- plane, cylinder and cone. Although a cylinder and a cone are not flat surfaces, either
can be flattened by cutting along its length and then unrolling the surface.

Aspect

Aspect refers to the orientation of the projection surface relative to the generating globe.
Polar Aspect (Normal aspect) aligns the axis of rotation of the projection surface with
the rotation axis of the Earth. Equatorial Aspect (Transverse aspect) rotates the
projection surface 90 degrees relative to the Earth’s rotation axis. The result is a plane
tangent at some point on the Equator, a cylinder tangent along an opposing pair of
meridians of longitude, or a cone whose axis of rotation lies in the plane of the Equator.
Oblique Aspect refers to any other aspects and are frequently used to generate maps
centred on particular locations.

The following figure gives the basic combinations of surfaces and aspects.

3-26
Distortion of Projections

It is important to recognize that the approximately spherical geometry of the Earth

cannot be represented on a flat map without introducing distortion. The distortion of a
map will occur in one or more properties of distance, area and angle.

Even though it is not possible to represent all distances or all angles correctly on a
single map, the map can be created to keep one of the selected geometric properties.
Consideration of basic properties results in four classes of map projections: equivalent,
equidistant, azimuthal and conformal.

Equivalent Projections

Maps that maintain constant areal scale over the entire surface.

Equidistant Projections

Equidistant projections preserve correct linear scale over some portion of the map. It is
not possible to represent all distances correctly.

Azimuthal Projections

Azimuthal (or zenithal) projections correctly represent selected angular relationships.

Conformal Projections

The defining property of conformal projections is that at any point on the map, the scale
is the same in all directions about the point. The implication of this property is that
angles about the point are shown correctly and one would therefore expect shapes of
areas to be correctly represented. However, because map scale varies from point to point,
the property of correct representation of angles applies only to angles having infinitely
short sides. Shapes of infinitely small areas are preserved on the map but shapes of
larger areas are distorted.

3-27
Compromise Projections

While many map projections seek to optimize one of the above four geometric
properties, some projections do not attempt to maintain any of the above properties but
instead attempt to seek a balance between different properties.

Method of Construction

Projections can also be classified based on their method of construction. This involves
consideration of the type of projection surface used as well as the positioning and
orientation of the projection surface relative to the Earth.

The type of projection surface determines the basic pattern of the map. Positioning the
projection surface tangent or secant to the generating globe modifies the pattern of
distortion in the map. Re-orienting the projection surface can be used to achieve
minimal distortion within the region of greatest interest.

Although an infinite number of map projections are theoretically possible,

approximately 400 projections have been described in the literature and only a few
dozen of them are widely used.

Global Reference Ellipsoids

With increasing demands for global surveying activities, several worldwide useful
datum and projections have been developed. The World Geodetic System 1984
(WGS84) is the most famous and widely used of the world wide datums.

Basic parameters of WGS84 are shown in the following figure. WGS84 provides the
basic reference frame for GPS (Global Positioning System) measurements and almost
all open source GIS data are created on the basis of the GWS84.
Two kinds of worldwide coordinate systems Geographic and UTM have been
developed on the basis of the WGS84.

3-28
Geographic Coordinates

Geographic coordinates are expressed in terms of latitude and longitude as have been
mentioned in the previous part of this section.

UTM Coordinates

UTM (Universal Transverse Mercator) is a kind of Cartesian coordinate system.

UTM utilizes the transverse Mercator projection, which results from wrapping the
cylinder around the poles instead of around the equator as for the ordinary Mercator
projection. The Earth is subdivided into sixty UTM zones, numbered from west to east,
starting with zone 1 at 180'W. Each zone is thus six degrees of longitude wide, and
extends from 84'N to 80'S. Displacements in the x and y directions are called UTM
eastings and UTM northings, respectively. The UTM is an excellent system for
regions covered by maps at scales of 1:250,000 and larger. At smaller scales, distortions
generally become unacceptable. Another point for the utility of the UTM is its
coordinates unit. Different from geographic coordinates, The UTM use meter as
coordinates unit, therefore it become convenient to calculate length and area.

3-29
3-3 Data Integrating (Image Georeference)
Add Raster Dataset and Reference Data

1. Start ArcMap. Open a new map and save it as Georeference into the Maps folder.

2. Set the coordinate of the data frame to WGS_1984.

3. Add the polygon feature of Eth from your GeoDB of Ethiopia/ Geographic_C.

4. Set the symbology of the layer Eth as no color for Fill and red color for Outline.

5. Add the image feature Ethiopia.jpg into the map from the GeoDB Image.

6. Set the full extent of the Data View.

You will find that the two layers don’t
match each other at all, the raster data
layer Ethiopia.jpg takes almost the
full extent of the view and the feature
class layer Eth is located at the upper
left corner as shown in the figure to
the right.

3-30
 You already know that the coordinate system for feature class Eth is defined as
WGS1984, while the image Ethiopia.jpg does not contain a proper coordinate
system. The information of the coordinate system has to be added otherwise the
image layer can not match the feature layer of Eth. The process of defining a
coordinate system for an image requires a reference with a defined coordinate
system. In this exercise, you can take the layer of Eth as a reference. The process
used in this type of matching is usually referred to as georeference.

7. Right-click anywhere on the toolbar and a dropdown menu of ArcMap will appear.

8. Click Georeferencing to open the Georeferencing toolbar. Drag and drop the
Georeferencing toolbar somewhere it will be convenient to use.

9. In the Georeferencing toolbar, click the dropdown arrow next to the Layer textbox
and select the raster layer you want to georeference. Here it is ethiopia

Because you have only one raster datum in your map, the raster data of Ethiopia
will be automatically selected as the target layer of georeferencing.

10. Right-click the reference layer Eth > Zoom to layer

11. Click dropdown arrow next to Georeferencing > click Fit To Display

The image size and location are changed to the full extent View as shown in the
following image. Since the view has been set to the full extent of the reference layer
Eth, the image layer is actually at an extent similar to the reference layer of Eth.

3-31
Add Control Points

To conduct georeferencing the basic method is to find several points in the target image
and their corresponding points in the reference layer. These points are referred to as
control points.

1. Click the Window menu in the toolbar > click Magnifier. A Magnification
window will appear.

2. Drag the magnification window to a location between a pair of control points in the
image and reference layer.

3. Click the Control Points tool in the Georeferencing tool bar. The mouse
pointer turns into a crosshair.

4. Center the crosshair over one control point in the image and click. A green control
point is added to the image.

5. Move the mouse pointer away from the control point but don’t click again. A line
stretches from the control point as you move the mouse. This is the link with which
you’ll connect the other end of the line to the corresponding point in the reference
layer.

6. Center the mouse pointer over the corresponding point on the layer of Eth - you
can see the link stretch as you do this - and click.

As soon as you click on the control point in the reference layer the map will move
to make the point in the image layer fit the point in the reference layer.

7. Drag and drop the Magnifier window to the next pair of control points.

3-32
 If necessary you can use the Zoom In button or Zoom Out button to
make it easier to find the next control point.

8. Center the crosshair over the second control point on the image and click. Then
center the crosshair over the corresponding control point on the reference layer and
click to add the second control point.

9. By repeating the steps 4 to 8, you can add enough control points to make the image
fit the reference layer well.

Perform Georeferencing

If you are satisfied with the result you see in the Data
View window, save the result by clicking the
dropdown arrow of Georeferencing and then clicking
Update Georeferencing.

The information of the coordinate system will be saved into the image. You can check it
> remove it from the data frame and then Add it again. This time you will find the
image is located in the Data View just the same as it was before you removed it.

The above exercise is the simplest case of georeferencing because the image was
created by using the same coordinate system as the reference layer. The reason the
image did not match the reference before georeferencing is because when the image was
saved a projection file to store coordinate system information was not created.

Georeferencing by Transformation

On the other hand, sometimes the image is created in a coordinate system different from
the reference layer. Also, it may be seriously deformed as a consequence of the original
paper map being stored poorly or improperly scanned. In such cases you will find it
impossible to make the image neatly match the reference layer no matter how many
control points you add between the image and the reference layer. For this kind of image
georeference you need to conduct a transform process.

1. Remove the raster data layer Ethiopa.jp.

3-33
2. Add another image EthiopiaN. Repeat the procedure in the previous exercise to
make the image fit the display and then add a sufficient number of control points.

3. You will find that the image Ethiopa.jp could not be adjusted to match the
reference simply by adding control points.

4. When you have added enough control points, you must pay special attention to add
control points with homogeneous distribution in the image as much as possible.

Here “enough” is a relative term with the minimum number being 3 points, but
there is no maximum number. How many control points are needed for
georeferencing depends on the condition of the image. The more deformed the
image is, the more control points will be needed. Experience shows a number
between 10-20 is enough to obtain acceptable georeferencing results.

5. Click the dropdown arrow for Georeferencing > click Transformation > select a
method from the three orders of Polynomial.

The minimum number of control points needed for a different polynomials order is:

- Three (3) for a first-order transformation

- Six (6) for a second order, and

- Ten (10) for a third

order

6. Check the results to see if there

is some portion you would still
like to modify. If so, add new
links for that part(s) and then
perform the transformation
again until the georeference
result is acceptable.

7. In the same way you practiced

for Performing Georeferencing, you can save the result by clicking the dropdown
arrow of Georeferencing and then clicking Update Georeferencing.

3-34
Assignment 5: Georeference a Base Map for Your Country
1. Add the base map of your country in ArcMap.

The base map of your country is an image in .jpg format with the name of
OOObase. Find the image from the folder:

C:/EthiopiaArcGIS/Data/CountriesData/OOO

Here, OOO represents an abbreviation of your country’s name.

2. Add the country boundary file into the same data set as the reference layer.

Use the country boundary file that you created in Assignment 3.

3. Georeference the image to match the reference data.

4. When you finish georeferencing, use the image as the background and the country
boundary file as the overlay layer. Set the symbology of the country boundary as no
color for Fill, red (or some other color that stands out) for Outline. The width
setting of the outline depends on your preference.

5. Set your map to the full extent of Data View.

6. Export the map in .jpg format to the folder C:/GIS_Exercise/Result and set the
resolution to make sure pixels for both width and height are greater than 1000.
Name the export file as follows:

Yourlastname_E5. (such as Lei_E5.jpg)

7. Upload the resulting image to the folder labeled with you name on the server
computer.

3-35
3-4: Point Feature Creation

Exercise 1: Create a Point Feature from an Image

When you finish georeferencing an image, the image becomes available in a number of
ways to support map creation or data creation. You can use the image as a background
for your map or create various kinds of vector data based on the image. Extracting point
data from a paper map is one of the essential processes in GIS data preparation (or
creation).

In this exercise you will become familiar with how to extract point data an image (i.e. a
scanned paper map).

Preparation for Point Feature Creation

1. Start ArcMap > open a new map > set the coordinate system of the data frame to
WGS84 > save the map as Temp (Overwrite the old one).

2. Add the image map_ethiopia_small.tif into the map from your GeoDB Image.

Within this image you will find the main cities of Ethiopia. Your task in this
exercise is to extract these cities from the image to create a point feature of the
main cities. The image has been georeferenced in the WGS84 coordinate system.

3. Display the Arc toolbox window if it is hidden.

4. Expand the menu for Data Management Tools > expand Feature Class > open
the Create Feature Class tool. The Create Feature Class window will appear.

5. Set the Output Locating as the Result folder

- Output Feature Class as Maincities

- Geometry Type (optional) as POINT

- Spatial Reference (optional) as WGS84

6. After the above settings have been made click the OK button. A New empty layer
Maincities is added into the map.

3-36
Extract Points from an Image

1. If the Editor toolbar is not showing click the Editor Toolbar button .

2. Start Editing.

3. Make sure the Task is Create New Feature and the Target layer is Maincities.

4. Click the Sketch Tool > move your curser into the Data View. Your curser

will change to a circle with a cross in it.

3-37
5. Put your curser on one point of a city and click. A new point is added into the Data
View and highlighted in blue.

6. Move your curser to a point of a another city, clicking them one by one.

7. Only the last point added will be

highlighted so the previously added
points will not be clearly visible. You
can change the symbology of the
layer Maincities to a cross mark
colored in red and set a proper size of
the point to make them easy to see..

8. After you have extracted all the cities

that can be identified from the image,
click the dropdown arrow of Editor
and then click Save Edits.

9. Zoom in so that it is possible to check if the points were added to the best location.

10. To adjust a point, change the editing Task to Reshape Features > click the Editor

Tool > select a point you would like to move > move it to the best location by

dragging and dropping.

11. After you finish adjustmenting all the points Stop Editing and save the Edits.

Add Attribute Data

Your task in this exercise is to create a point feature for maincities; at least one field is
necessary to store the name of the cities.

1. Open the attribute table of the layer Maincities > click the Options button > click
Add Field. The Add Field dialog box will appear.

2. Set in the Add Field dialog box:

- Name : Name
- Type : Text
- Field Properties/Length: 25

3-38
3. Click the OK button. Now you have to input the name of the cities into the new
field you added.

4. Start Editing. Make sure the Task is set to Reshape Feature.

5. Zoom in so that you can clearly see both the

point of a city and its name shown in the
background image.

6. Click the Editor Tool > click one point

of a city to select it.

7. Click the Attributes Tool . The Attribute dialog box will appear. Click the cell

within the Value column that corresponds to the Name field > referring to the
background image input the city name into the cell.

8. Repeat Step 7 to add all the city names into the attribute table.

9. After you finish adding attributes of city names, close the Attributes dialog box.

10. Open the Properties of the layer Maincities > click the Labels tab > check the box
for Label features in this layer > accept the default Method of Label all the
features the same
way > set Name as
Label Field > (if you
like you can also set
the Text symbol)
>click OK to close
the Properties
window.

3-39
11. Right-click the layer of Maincities > click Label
Features. All the data you have added into the Name
field of the attribute table will appear. Check if you have
properly added the names of the cities.

Exercise 2: Create a Point Feature from an XY Data Table

Using a table is the most common method to store point data in our work. You might
have tables for wells, water quality survey results, village information, and so on.
Within some tables the location information is generally stored in two columns, X and Y.
Certainly, for different coordinate systems, X and Y can be expressed in different ways
such as longitude and latitude for the Geographic coordinate system or Easting and
Northing for the UTM coordinate system. Given that you have at least these two kinds
of coordinate information, you can create a point feature using a table.

Adjust Coordinate Data Using Excel

One of the most popular formats of tabular data is the data sheet using MS Excel,
therefore a *.xls table is used as a sample for this exercise.

1. Start MS Excel > open WQ_AA.xls from the following folder

C:/EthiopiaArcGIS/Data/OtherData.

WQ_AA.xls is a data table for storing groundwater quality analysis results of many
wells in the Addis Ababa area.

2. Check the components of the data sheet well_AA.

You can find the first 3 columns are initiated by Long followed by 3 columns
initiated by Lati. You might have already made out that the meanings of Long and
Lati are abbreviations of Longitude and Latitude, respectively. These 6 columns are
essential coordinate information to add this table into ArcMap.

However, this coordinate information can not be used to add the table into ArcGIS
because ArcGIS can only recognize decimal format no matter if the coordinate
system is Geographic (longitude/latitude) or UTM (meter). Therefore, you have to
perform data format conversion according to the followed steps.

To the right side of the coordinate information you will find several water quality

3-40
 analyses results are stored, such as TDS, EC and so on.

3. Use the following equation to change the coordinate information from separated
degree, minute and second into decimal degree:

Decimal Degree = D+M/60+S/3600

Here, D means degree; M means minute; and S means second.

- add a new column behind the data set of longitude

- set the column title as Longitude

- input the above equation into the top cell of the column

- copy the equation to the end of the column

- repeat the above procedure for the data set of latitude

- copy the whole data sheet

- paste by value the whole data sheet

- delete the columns in which the dataset for longitude and latitude are
stored.

- select the first two columns (Longitude and Latitude)

- set the format of these two column as Value with 6 decimal places

- select the other columns

- set the format of them as Value with 3 decimal places

The above format setting is needed only to save the data sheet as a *.dbf file. If you
would like to save the data sheet as a *.txt file, you can set the format for all data as
standard.

4. After you finish the data format conversion in Step 3, you should save the data
sheet into text format or *.dbf (dBase) format since those are the only two formats
in which ArcGIS can read table data.

- click the File menu

- click Save As

3-41
 - browse to the folder My_data

- set the name as qw_aa_data

- set the Type as DBF4 or text (space delimit)

- click Save

- click Yes in the following message box

- close the table

Add the XY Coordinate Data into ArcMap

1. Start ArcMap > open a new map > set the coordinate system of the data frame as
WGS1984 > save the map as Temp (Overwrite the old one)

2. Add the feature A_Ababa as a reference layer.

3. Click the Tools menu > click Add XY Data. The Add XY Data window will
appear.

4. Click the browse button next to the first textbox

- browse to the folder My_data

- select WQ_AA_data.dbf that you created in the above exercise

- click Add

You will find that if the columns for coordinate information are titled as longitude
and latitude, they will be automatically selected by the Add XY Data tool.

If the columns for storing coordinate data are titled X and Y, they will also be
recognized automatically. However, if the columns of coordinate data are titled
with other words, you have to select them by clicking the dropdown arrow next to
the textboxes of X Field and Y Field.

5. Click the Edit button to define the coordinate system; in this exercise WGS_1984.

3-42
6. Click OK.

7. A new layer QW_AA_data Event is added to your map.

At the present moment you just finished adding the XY data into ArcMap, but the
layer you can see from your map is an in-memory feature. That is, it can not be
used for other maps or programs.

8. Right-click the layer of QW_AA_data Event. The dropdown menu will appear.

- Click Data. The sub dropdown menu will appear.

- Click Export Data. The Export Data dialog box will appear.

- Set the coordinate system as the data frame

- Set the name of export feature as Wq_aa.shp and the location as

C:/GIS_Exercise/Result

- Click OK

3-43
9. When the exporting is complete you will be asked if you would like to add the
export feature to the map. Click Yes.

Create a Feature from XY Data Using ArcCatalog (optional)

1. Start ArcCatalog > browse to the folder My_data > right-click the text file
WQ_AA_data.txt > click Create Feature Class > click From XY Table. The
corresponding window will appear.

2. The same way you added XY data into ArcMap, the two columns of coordinates are
automatically selected into the text boxes of X field and Y Field.

3. Click the button of “Spatial reference of Input Coordinates...”, Spatial

Reference Properties window will appear. Then set the coordinate system as
WGS_1984.

4. Set the output location as your Temp folder, and set the output name to be the same
as the original text file, WQ_AA_data, then click OK.

5. Browse to the Temp folder, although you might not find the feature WQ_AA_data
that you just created. If so, click the View menu > click Refresh. The feature will
appear in your folder.

6. You can drag and drop it into ArcMap to compare the feature WQ_AA_data with
Wq_aa, and you will find them to be identical.

3-44
Assignment 6: Create a Point File from Place Name Data
1. Add your country’s boundary file in ArcMap.

Use the country boundary file that you created in Assignment 3. Set the symbology
as you like, but avoid using “no color” as a color setting.

2. Find the file of place name for your country.

The file is named as OOO_pn.txt in the folder

C:/EthiopiaArcGIS/Data/CountriesData/OOO

Here. OOO is the abbreviation for your country’s name

This place name file is downloaded from the NGA (National

Geospatial-Intelligence Agency, US) website. The address of the website is:

http://earth-info.nga.mil/gns/html/cntry_files.html

Please refer to the table attached to this section to get the details of the field in the
place name file.

3. Add your country’s place name file into ArcMap as an event layer.

Set the column of LONG for X Field, LAT for Y Field.

4. Export the events layer.

Name of output : OOO_pn

Location of output : C:/GIS_Exercise/Result

5. Add the exported feature into the same dataset. Inactivate the events layer (although
in the meantime the events layer should cover the new layer exactly). Set the
symbology of the new layer as you like.

6. If there are points outside the country’s boundary, use the boundary layer to clip the
point layer.

To clip you can use the Clip tool in the ArcToolbox menu for Analysis Tools/
Extract

7. Use the column of FC in the new layer as the target field to create a selection based
on one value in the field, for example, A.

3-45
 8. Export the map in .jpg format into the folder C:/GIS_Exercise/Result and set the
resolution to make sure pixels for both width and height are greater than 1000, with
the file name as follows:
Yourlastname_E5. (such as Lei_E5.jpg)

9. Upload your resulting image to folder with your name in the server computer.

Description of Names Files for Countries and Territories Format

Field Field Default

Field Description
Name Type Selection
Region Font Code. A code that determines the character mapping
used in the SHORT_FORM, Generic, and Full Name fields:
1 = Americas/Western Europe;
2 = Eastern Europe;
RC number Y
3 = Africa/Middle East;
4 = Russia/ Central Asia;
5 = Asia/Pacific;
6 = Vietnam.
Unique Feature Identifier. A number which uniquely identifies a
UFI number Y
Geoname feature.
Unique Name Identifier. A number which uniquely identifies a
UNI number Y
name.
Latitude of the Geoname feature in ± decimal degrees; DD; (±
dd.dd...):
no sign (+) = North;
LAT number Y
05°03'09''E » 5.0525
negative sign (-) = South;
05°03'09''S » -5.0525
Longitude of the feature in ± decimal degrees; DD; (± dd.dd...):
LONG number Y
no sign (+) = East;

3-46
 Field Field Default
Field Description
Name Type Selection
negative sign (-) = West.
Latitude of the Geoname feature in ±degrees, minutes, and
DMS_ seconds; DMS; (±ddmmss):
number Y
LAT no sign (+) = North;
negative sign (-) = South.
Longitude of the Geoname feature in ±degrees, minutes, and
DMS_ seconds; DMS; (±dddmmss) :
number Y
LONG no sign (+) = East;
negative sign (-) = West.
Universal Transverse Mercator coordinate grid reference. (This
field is derived from the system generated and stored Military
UTM character Y
Grid Reference System (MGRS) coordinates when this file is
generated.)
JOG Joint Operations Graphic reference. character Y
Feature Classification: Nine (9) major Geoname feature categories
into which similar feature designations are grouped.
A = Administrative region type feature
P = Populated place type feature
V = Vegetation type feature
FC L = Locality or area type feature character Y
U = Undersea type feature
R = Streets, highways, roads, or railroad type feature
T = Hypsographic type feature
H = Hydrographic type feature
S = Spot type feature
Feature Designation Code. A two to five-character code used to
DSG character
identify the type of Geoname feature a name is applied to.
Populated Place Classification. A graduated numerical scale
denoting the relative importance of a populated place. The scale
PC ranges from 1, relatively high, to 5, relatively low. The scale number
could also include NULL (no value) as a value for populated
places with unknown or undetermined classification.
Primary Country Code. A two alphabetic character code from the
CC1 FIPS 10-4 Standard that uniquely identify a geopolitical entity character
(countries, dependencies, and areas of special sovereignty).
First-order administrative division code. A two alpha-numeric
code from the FIPS 10-4 Standard describing a primary
ADM1 character
administrative division of a geopolitical entity, such as a state in
the United States.
Second-order administrative division. The name of a subdivision
variable
ADM2 of a first-order administrative division, such as a county in the
character
United States.
POP Population figures. number
ELEV Elevation in meters. number
Secondary Country Code. A two alphabetic character code from
the FIPS 10-4 Standard uniquely identifying the country code of a
CC2 particular Geoname feature name if different than that of the character
primary country code of that Geoname feature (International
features).
Name Type:
NT C = Conventional name character
N = BGN Standard name

3-47
 Field Field Default
Field Description
Name Type Selection
NS = BGN Standard name in non-Roman script
P = Provisional name
PS = Provisional name in non-Roman script
H = Historic name
HS = Historic name in non-Roman script
D = Not verified or daggered name
DS = Not verified name in non-Roman script
V = Variant or alternate name
VS = Variant name in non-Roman script
Language Code. A three alphabetic character code (ISO 639-3)
LC uniquely identifying the language assigned to a Geoname feature character
name.
SHORT_ variable
A part of the full name that could substitute for the full name.
FORM character
GENERI The descriptive part of the full name such as Mount, River, or Sea variable
C (generally does not apply to populated place names). character
A form of the full name that allows for alphabetical sorting of the
file into gazetteer sequence. All character/diacritic combinations
and special characters are substituted with QWERTY (visible U.S.
English keyboard) characters, all characters are upper-cased,
SORT_ variable
numerals are converted to lower-case characters (0-9 = a-j), spaces
NAME character
are stripped out, and commas replaced with space. The
substitution is done according to a character mapping table that is
Region Font Code specific. This field is included for the benefit of
the end user of the data to aid in the sorting of names if required.
The full name is the complete name that identifies a named
FULL_N feature. The full name is output in reading order, "Mount variable
AME Everest", vs. reversed generic, "Everest, Mount", as stored in the character
database.
Same as the full name but the character/diacritic combinations and
special characters are substituted with QWERTY (visible U.S.
FULL_
English keyboard) characters while still maintaining casing and variable
NAME_
spaces. The substitution is done according to a character mapping character
ND
table that is Region Font Code specific ND = No Diacritics /
Stripped Diacritics.
MODIFY The date a new feature was added or any part of an existing
date-time
DATE feature was modified (YYYY-MM-DD).

3-48
3-5: Polyline and Polygon Feature Creation
Exercise :1 Create a Polyline or Polygon Feature by Tracing an Image (optional)

The method of tracing a polyline or polygon from an image is basically the same as
creating a point feature from an image.

1. Start ArcMap, open a new map > set the coordinate system of the data frame, for
example WGS_1984 UTM Zone 37N > save it as Temp (overwrite the old one).

2. Add the reference image Baseline_primary_data_a3 into the map from your
GeoDB Image.

Your task in this exercise is to delineate the main roads within the study area of
EWTEC.

3. Create a new empty feature with the following procedure

- expand toolbox of Data Management Tools

- expand toolbox of Feature Class

- open the Create Feature Class tool

- set the Output Location to your Result folder

- set the Output Feature Class as SA_Mroad

- set the Geometry Type (optional) as POLYLINE

- set the Spatial Reference (optional) as WGS_1984 UTM Zone 37N

4. After the above settings, click OK. A new empty layer of SA_Mroad is added to
the map.

5. Start Editing > set the Task as

Create New Feature.

6. Click the Sketch tool > click on

the Data View to trace lines of roads.

7. After tracing the road lines, save

Edits.

3-49
8. Edit the traced result (refer to the Exercise 3 in Module 2, Section 5)

The procedure of creating a polygon feature by tracing an image is almost the same as
the above exercise. The difference is:

- Select POLYGON when you cerate a new empty feature.

- Edit the traced result by referring to the procedure in Exercise 6 in

Module 2, Section 5.

Exercise 2: Using ArcScan for Line Vectorization

When your base map image contains a lot of lines or polygons, the tracing method as
shown in the above exercise would be a very tedious and time consuming task.
Therefore, as you might expect, there are some methods to carry out the delineating
process more effectively and more automatically. The following exercise will introduce
some of these methods.

Preparation

1. Scan the paper as an image with the extension *.tif, *.jpg, *.bmp or other formats. In
this exercise you should have already completed scanning and have the file
afoutline.tif.

2. Start ArcMap, open a new map, set the WGS84 as the coordinate system for data
frame, save the map as Temp (overwrite the old one).

3. Add the image afoutline.tif from your GeoDB Image.

You will find the image shows the boundary lines for all African countries. In this
exercise, you will try to delineate these countries’ boundary lines to create a
polygon feature.

4. Start ArcCatalog > browse to and right-click your Temp folder > click New > click
Shape file. The Create New Shape file window will appear.

5. Input the required settings as follows:

- Name: templine
- Feature Type : polyline
- Spatial Reference: WGS_1984.

3-50
6. Click OK. A new empty polyline shape templine file will be added into your Temp
folder.

7. Drag and drop the newly created shape file templine into ArcMap.

8. Referring to Steps 4 to 7, create and add a new empty polygon tempgon.shp into
ArcMap.

9. Set the Symbology of the new layers you added as follows:

- Templine: red for line symbol

- Tempgon: Blue for Outline and yellow for Fill.

Vectorization Using ArcScan

1. Go to ArcMap > open the Layer properties window of the layer afoutline > click
the Symbology tab > set the Show as Classified and then set the classes as 2. Click
OK.

2. Display the Editor tool bar if it is hidden.

3. Start Editing. Make sure the Task is Create New Feature and the Target is
templine that you created and added earlier in this exercise.

4. Click the Tools menu > click Extensions.... make sure the extension ArcScan has
been checked.

5. Right-click anywhere on the tool bar or menu bar to bring up the toolbar list, select
ArcScan, if it is unselected.

6. Click the dropdown arrow next to Vectorization in the ArcScan Toolbar > click
Vectorization Setting.... The Vectorization Settings window will appear.

3-51
7. Specify the values for each parameter as shown in the figure, and then click Apply.

Notice: If you don’t set any parameter in the above window the Apply button will
remain unavailable.

8. Don’t close the Vectorization Settings window. Move it to one side of the display
to allow you to see the Data View.

9. Click the dropdown arrow next to Vectorization and then click Show PreView.

10. Generally the result will be acceptable, but if you are not satisfied with the result
you can go back to the Vectorization Settings window and change the parameter
setting, then regenerate the lines for the layer of templine.

In the full extent of Data View, you might find it difficult to confirm if the
vectorization has been performed properly because in some part the red line appears
to deviate from the line in the background image (refer to the left side of the figure
below). Then you can zoom in to a scale large enough to confirm the result of
vectorization (refer to the right side of the figure below)

3-52
11. After you have checked the result and find it acceptable, close the Vectorization
Settings window.

12. Click the dropdown arrow next to Vectorization and then click Generate Features
to bring up the Generate Features window.

13. Make sure the generated line class will be added into the layer of templine, accept
the default setting for “Generate polygon where the maximum line width setting
is exceeded”, and make sure those polygons will be added into the newly created
empty polygon layer tempgon.

14. Click OK and wait until the vectorization result is added into the layer of templine.

15. Stop Editing and save the vectorization result.

Since you have completed the vectorization, it is suggested you hide the ArcScan

3-53
 toolbar to keep your window uncluttered and easy to work in.

You will find that when you stopped editing, the ArcScan tools of were dimmed. In
other words, the extension of ArcScan only work in editing mode.

Exercise 3: Convert Multiple Polylines to Multiple Polygons

If your task were only to extract lines from an image to create a polyline feature, you
have completed your task with the exercise above. However, the source map in this
exercise is a boundary map of African countries, and generally the country’s boundary is
expressed as a polygon shape. Therefore, the results from the last exercise should be
further processed to convert to a polygon feature.

1. Display the ArcTool Box window if it is hidden.

2. Expand the Data Management Tools menu > expand Features, open the Feature
to Polygon tool. The Feature To Polygon window will appear.

3. In the Feature To Polygon window, enter the following settings:

- Input Features: templine that you created in the last exercise

- Output Feature Class: Af_cntry_temp

- Location of Output: your Result folder

3-54
4. Click OK and wait a moment for the conversion process to complete.

5. The new polygon feature will be added into your map and then you can check and
modify it to complete the vectorization.

You can also use ArcCatalog to perform the feature type conversion from polyline to
polygon with the following steps.

1. Start ArcCatalog.

2. Right-click your GeoDB OtherData > click New > click Feature Dataset > name
the new feature dataset as temp and set the coordinate to WGS_1984.

3. Import the polyline feature templine from your Temp folder to the new dataset
temp.

4. Right-click the new feature dataset of temp > click New > click Polygon Feature
Class From Lines…. The Polygon Feature Class From Line window will appear.

5. In the Enter a name for the feature class textbox input a name such as
Af_cntry_temp1 for the new polygon feature. Then check the only polyline file

3-55
 Templine in the Select the features classes that will contribute lines feature list
box. Click OK.

Adding Attribute Data into Polygon Feature

You have completed the creation of a polygon feature for African countries, but the
feature is not particularly useful at this moment because it doesn't contain any of the
necessary attributes. The feature of Af_cntry_temp will not even tell you the name of
the countries. Therefore, the next task should be adding the necessary attributes to the
polygon feature of Af_cntry_temp. There are many methods to add attributes to a
feature; a common method is to add attribute data by obtaining them from another
related feature.

In this exercise, you will learn the process of adding attributes from a point feature to a
polygon.

1. Add the point feature cntry_pt from your GeoDB OtherData

This point feature gives the location of centroid points of all countries on the
African continent and stores basic attributes like country name, area and so on.

2. Right-click the layer of Af_cntry_temp > click Joins and Relates > click Join.
The Join Data window will appear.

3-56
3. In the top textbox, select “Join data from another layer based on spatial
location”.

4. In the second textbox, select the layer of cntry_pt.

5. Check the second radio button to make all the attributes of the point layer cntry_pt
added into the polygon
layer Af_cntry_temp.

6. In the third section click

the browse icon and

input the settings as

- Location: your Temp

folder

- Name: Temp

- Save as type:
Shapefile

7. Click the Save button to

go back to the Join data
window, and click the
OK button.

8. A new layer Temp is saved in the destination folder and added to your data frame.
This layer is a polygon feature, and has the same spatial extent as the layer of
Af_cntry_temp.

9. Open the attribute table of the layer Temp, you will find the attribute data has been
added from the cntry_pt into Temp.

10. Find the fields of COUNTRY, in which

the abbreviation of country names are
stored. Right-click the heading of the field
> click Freeze/Unfreeze Column.

3-57
 The field will go to the leftmost side of the attribute table and when you move the
button of the slide bar in a horizontal direction so that the COUNTRY field will
not disappear. It will now be easy for you to compare the data in field of
COUNTRY with data in any other field.

11. Sort the COUNTRY field in ascending order, make width of the attribute table
window as narrow as can to show only two fields, and make the last field of
Distance shown together with the field of COUNTRY.

The comparison of these two fields will allow you to find:

- Several countries for which you have created two or more polygons, so
you can to combine (merge) them into one.

- Every country that has a feature with the data value 0 in field of Distance.

This value indicates to you that there is a point from layer cntry_pt
located within the polygon feature.

- Those features with values larger than 0 in field of Distance.

These values indicate to you that there are no points from layer cntry_pt
located within the corresponding polygon features. You will find that all
polygon features with these values correspond to more than one of the
same country name in the COUNTRY field. Therefore, these features
should be modified by combining them with another polygon feature or
deleting them.

12. Based on the indications in the attribute table, it will be easy for you to find and
reshape the polygon layer Af_cntry_temp.

The following is the basic procedure for your reference.

- Inactivate all layers except the background image afoutline and Temp.

- Put the Attributes table window adjacent to the Data View so that you
can easily change between them.

- Start Editing. Set Task to Reshape Feature and the Target layer to
Temp.

- Select all features with the same country name.

3-58
 - Zoom in to the selected features to check if all the features should be
combined into one country. If so, merge them; if not, decide how you will
deal with it. In this exercise you have only one option other than merging,
which is deleting. In other cases you might have another option, leaving it
without reshape. In that case you have to add attribute data by another
method.

13. After you finish editing the polygon feature, you can complete your task by Stop
Editing > Save the editing result > export the edited polygon feature to your data
base.

Exercise 4: Create a Polygon Feature from a Color Classified Image

In the above exercises you practiced how to convert a single color image into a line
feature or even a polygon feature. However, in your stocks of paper maps, which will be
used to create GIS data, you may have some color classified maps, such as geology
maps, landuse maps and so on. To make color classified maps available in our GIS, the
method of using ArcScan is not a suitable selection because the method is based on the
two class classification for symbology of the map.

In this exercise, one nonstandard method will be introduced for your reference.

3-59
Preparation

1. Start ArcMap > open a new map > set coordinate system of the data frame to:
WGS_1984 UTM Zone 37N > save the map as Temp (overwrite the old one).

2. Add the background image geology from the GeoDB Image.

The image is a geology map which covers the majority of the study area of EWTEC.
Just glancing at the map, you might imagine how hard it would be to manually trace
all polygons from the map.

Convert Raster to Polygon

1. Display the Arc Toolbox window if it is hidden.

2. Expand the Conversion Tools menu > expand From Raster > open the Raster to
Polygon tool. The corresponding window will appear.

3. Set the necessary factors in the Raster to Polygon window as follows:

- Input raster: layer geology

- Field: Value (the tool will automatically set this factor for you)

- Location of output: your Temp folder

- Name of output: Temp_geology.shp

The converted result of Temp_geology will be added into the data frame and you
can find the boundary line zigzags and is too thick. If you check the attribute data of
the layer, you will also find the number of polygons included in the layer is as

3-60
 much as 450,000 or greater. If you check the size of the shape file from the folder
you saved it to, you will find it is as big as about 88M.

Add a Field and Calculate Area

1. Open the attribute table of the layer Temp_geology > click Options > click Add
Field. The Add Field dialog box will appear.

2. Input the settings as follows and then click OK

- Name: Area
- Type: Float
- Precision: 20
- Scale: 0.
3. Right-click the heading of the new field Area > click Calculate Value > click Yes
in the following message box. The Field Calculator window will appear.

4. Referring to Exercise 2 in Section 2 of this module, conduct area calculation.

5. Sort the Area field in ascending direction. You should find that most polygons have
a very small area, less than 1000. These features consist of boundary and slivers.
The geological entities are those features with relatively large areas.

3-61
Remove Large Area Features

1. Click the Selection menu, make the Temp_geology layer the only selectable layer.

2. Click Select By Attribute. The Select By Attributes window will appear.

3. Confirm that the layer Temp_geology is the target layer, set the selection equation
as: "Area" > 1000000. Then click the Apply button.

4. When the selection complete, close the Select By Attributes window. Check the
selected result.

The result of your selection will be highlighted in the Data View. All the main
geological entities have been selected except a very small one on the upper lefthand
corner.

Also, you can check the attribute table to see that your selection contains about 32
features.

5. Start Editing, set the Task to Reshape Feature and the Target layer as
Temp_geology.

6. Click the Delete button in the toolbar. When deletion is complete, the main
geological entities and the boundaries in left, top and right directions disappear. The
Data View will look like the figure shown below.

7. Stop Editing.

3-62
Add a Frame and Export the Data View

1. Set the Symbology of the layer Temp_gfology as black for Fill and no color for
Outline.

2. Display the Draw tool > select the New Rectangle tool > draw a rectangle in
the Data View.

3. Make the Outline of the rectangle black, 1 for outline width, and no color for Fill.

4. Zoom in to each side of the rectangle and adjust it to make it fit the boundary of the
layer Temp_gfology.

5. After you finish adjusting the rectangle, hide the Draw toolbar.

6. Adjust the extent of your Data View to make it fit the layer of Temp_gfology and
the rectangle you drew. The Data View will look like the figure shown below.

7. Click the File menu > click Export Map.... The Export Map window will appear.
Input the settings within the window as follows:

In the main window:

- Output location: C:/GIS_Exercise/Temp

- Name: Temp_geology

- Type: TIFF(*.tif)

In the General tab:

3-63
 - Resolution: Adjust to make the pixels for both height and width greater than
1500.
- Write World File: checked
In the Format tab:
- Color Mode : 1-bit Monochrome Threshold
- Compression: None
8. Add the exported result of Temp_geology. You will be asked if you would like to
build a Pyramid, click Yes.

Generally, the exported image should have a similar spatial extent as the original
Data View. However, sometimes you might find the spatial extent of the
exported .tif format image is not exactly the same as its original View. If this is the
case, you should conduct georeferencing by the method you practiced in Section 3
of this Module.

After georeferencing, hide the georeference toolbar.

Vectorization Using ArcScan

The procedure of this part is almost the same as you practiced to vectorize the image of
African countries in Exercise 2 of this section. So only a digest is given in this text
book as a reference for you to practice vectorization.

1. Display the ArcTool window if it is hidden.

2. Expand Data Management > expand Feature class > open Create Feature Class

- Output location: Temp folder

- Name of Output Feature Class: Temp_line

- Geometry Type (optional): POLYLINE

- Spatial Reference (optional): WGS_1984_UTM_Zone_37N

3. Refer to the process above to create a new, empty polygon feature Temp_gon.

Both new, empty polyline and polygon features are added into the same data frame
as the image Temp_geology.tif.

3-64
4. Set the symbology of the layer Temp_line as red. And set the symbology of the
layer Temp_gon as pink for outline; no color for fill.

5. Start Editing. Set the Task to Create New Feature and the Target to the layer of
Temp_line.

6. Set the Symbology of the image Temp_geology.tif as classified and the classes as
2.

7. Display the ArcScan toolbar and make sure the Target Raster is
Temp_geology.tif.

8. Click the dropdown arrow next to Vectorization > Vectorization Settings.

9. Set the Maximum Line Width as 50. (Even though the setting might be 50, you
have to set it once again, otherwise the Apply button will not function). Click the
Apply button.

10. Click the dropdown arrow next to Vectorization > Show Previous.

11. Click the dropdown arrow next to Vectorization > Generate Features. Set the
Temp_line and Temp_gon as the two necessary layers feature generation. Click
OK.

12. Stop Editing with the edited result saved. Hide the ArcScan toolbar.

13. When you check the result from Temp_line and Temp_gon, respectively you will
find that the main geological entities are stored in the layer of Temp_line (refer to
the upper side of the figure below), and the layer of Temp_gon contains mainly
slivers (refer to the lower side of the figure below). Maybe you would agree that the
polyline feature would be better than the polygon feature for use as a vectorization
result, even though you still have to edit over the polyline feature.

3-65
Vectorization Result Editing

The main task of polyline feature editing is to delete the extra line features that
correspond to slivers such as labels or defect segments that occurred in all of the above
processes. You can use the method that you practiced in Exercise 3, Section 5, Module
2. You can also utilize the characteristic that most slivers have a relatively short length
to simplify the editing process by the following procedure.

1. Start by converting the temp_line into a polygon feature. Display the ArcTool
window if it is hidden.

2. Expand Data Management Tools > expand Features > open Feature To Polygon.
In the Feature to Polygon window, input the settings as follows:

- Input Features: Temp_line

- Name of Output Feature Class: geo_line2gon

- Location: your Temp folder

3-66
 The new polygon will be added into the same data frame as its source data of
Temp_line. You will find many slivers were eliminated in the conversion process.
Only a few extra polygons representing labels and slivers need to be edited further
to delete them.

Although you can use the method polygon editing feature that you practiced in
Exercise 6, Section 5, Module 2, here your main purpose is to attach small-area
polygons to their adjacent large-area polygon so you can use the Elimination tool
of ArcGIS.

3. Add a new field and calculate the area as you have practiced several times already.

4. Sort the new field and define selection criteria to distinguish small sliver polygons
from real geological entities. You might set 7000m2 as a temporary value of the
criteria.

5. Perform Selection By Attributes to select all polygons with an area less than 7000
in the layer of geo_line2gon.

You will find that even

if you selected as many
as 130 polygons or
more from the 200 or
so polygon features,
the majority of your
selection are not
boundaries but slivers．

3-67
6. Expand of the Data Management Tools menu > expand Generalization > open
the Eliminate tool.

7. In the Eliminate Window, input the settings as follows:

- Input Layer: Geo_line2gon

- Name of Output Feature Class: geo_line2gon_E

- Location: your Temp folder

The eliminate result geo_line2gon_E is added into your data frame. You can check
the result of the elimination and find the there are less than 100 polygons remaining
in the layer geo_line2gon_E.

8. You can eliminate slivers further by selecting the slivers directly from Data View
or setting new selection criteria.

For example, if you set the new selection criteria as area < 1,000,000 and then
perform elimination once again you would get a result as shown in the figure
below.

3-68
 The new eliminated result contains 32 features; to check all of them would not be
too difficult a task. After you set the symbology of the new eliminated result as no
color for fill and red for outline, you will be able to compare the eliminated result
with the original geology map. You will find that you still need to do some manual
editing to complete the polygon feature creation from a color classed image.

In this exercise you practiced creating a polygon feature based on a color classified
image. One thing that you should understand is that right now there is no standard
method for color classified image vectorization. That is why high precision GIS data
requires that you perform digitizing using very expensive machinary and consume much
more time. The method introduced here is only to be used in the following conditions:

- You don’t have a digitizer

- You don’t have enough time to manually digitizing the paper map

- The precision of this semi-automatic method for digitizing is sufficient for your
needs of GIS data base creation.

Exercise 5: Extract a Polygon Feature from a Satellite Photograph

A lot of features such as lakes, ponds, farm land, etc. are base data for water resources
management. For example, farmland distribution is one of the most important factors
for groundwater management because irrigation in many countries is the largest portion
of water use. In many countries excessive groundwater pumpage for irrigation has

3-69
caused serious groundwater head depression, and thereby resulted in a number of
environmental problems such as reduction in natural vegetation or even extinction,
abandonment of existing water use facilities (qanats, wells, etc.), expantion of
contaminated groundwater and so on. Therefore, irrigation water use control has
become a very important issue in groundwater management in these countries. To carry
our irrigation water use control, data concerning the distribution of farmland is
indispensable. But this kind of map is not a common one and might be difficult for you
to find and obtain. Therefore, you have to create such maps by yourselves.
Developments in remote sensing technology enable you to obtain the image data,
including farmland distribution. In this exercise you will become familiar with the basic
concept of satellite photograph processing.

Introducing MrSID Data

Many types of satellite photography have become available for purchase, but most of
them are too expensive. Fortunately, one of the most famous satellite photograph types,
Landsat images from USNASA, have been released for free. You can download Landsat
imagery from the following homepage:
http://glcfapp.umiacs.umd.edu:8080/esdi/index.jsp
Another very useful source of data you should know is MrSID (Multi Resolution
Seamless Image Database). MrSID (pronounced Mister Sid) is one of the most excellent
formats for image compression available and thereby used by NATO, NASA USGS and
many other well-known organizations and agencies. MrSID has also become one of the
standard formats of GIS data and adopted by ESRI, Autodesk, Mapinfo, ERDAS and
other famous GIS applications.
Some of MrSID images have been released by NASA. For general utility, you can
obtain (download) 2 sets of MrSID data from the following Web site:
https://zulu.ssc.nasa.gov/mrsid/

The two sets of MrSID data are MrSID 1990 and MrSID 2000. Both of the data are
based on US satellite image Landsat and cover the whole world.

One scene from MrSID covers a far more vast extent than one scene from Landsat itself.
MrSID extends north-south over 5 degrees of latitude, and spans east-west for the full
width of the UTM zone (6 degrees of longitude). The pixel size of MrSID is 14.25m for
MrSID 2000 and 28.5m for MrSID 1990.

3-70
The spectral bands for MrSID are red, green and blue, which correspond to the Landsat
bands of 7, 4 and 2.

1. Start ArcCatalog > click the Tools menu > click Options > click the General tab >
uncheck the box for Hide File Extensions > click OK.

2. Browse to the folder C:/EthiopiaArcGIS/Data/MrSID /N-37-05_2000 > click the

data N-37-05_2000.sid > click the Preview tab, if you are not already viewing it.

This is a MrSID scene which covers Addis Ababa and a vast area within Ethiopia.

3. Right-click the data in the directory tree > open the Properties of the data > move
the slide button to find the item Spatial Reference. You will find it is currently
Undefined.

4. Click the Edit button and the now familiar Spatial Reference Property window
will appear.

5. Set the coordinate to the UTM_WGS_1984, and the UTM zone, which is just the
same as the data name, here is 37N.

6. Start ArcMap > open a new map > set the coordinate system of the new map to
UTM_WGS_1984_Zone_37N > save the map as SatilleteImage in your Maps
folder.

7. Add the MrSID data N-37-05_2000.sid into your map, then add the features
A_Ababa and Eth as a reference.

3-71
8. Even you don’t carry out any more of the process, the map you have added into your
data set can be used as a background.

Tip: when you add the MrSID data into your map, you can add the individual band
form of red, green and blue. When you reach MrSID, double click it and the three
bands will be shown in your Add Data window; then you can select one of them to
add into your data frame.

Farmland Area Extraction by Simplifying Raster Data

To extract a feature from a satellite image you have to know the area you're interested in
clearly, otherwise you will not be able to select the color range to extract a special
feature. One water resource survey JICA project was carried out in the northwest part of
China, so the color range of farmland has been made clear there. This data will be used
here for the farmland extraction exercise.

1. Open a new map > set the coordinate system of the data frame to UTM_
WGS_1984_Zone45N > save the map as SatilleteImage (overwrite the old one).

2. Add the base MrSID data N-45-40_2000.sid from the folder:

C:/EthiopiaArcGIS/Data/MrSID/N-45-40

Then add a feature class nss_gridbase from your GeoDB of OtherData.

This feature you have used in the exercise of conversion polygon to centroid. The
feature gives the shape and extent of a simplified grid for a groundwater simulation
model. Your task here is to extract the farmland distribution in the simulation area
so that the irrigation amount can be calculated for model establishment.

3. Zoom to the layer of

nss_gridbase > set the
Symbology of the layer
nss_gridbase as no color for
fill, grey for outline.

3-72
4. Confirm the structure of MrSID within the Data View. Right-click the layer of
N-45-40.sid > click Data > click Export Data. The Export Raster Data window
will appear. You can confirm the structure of the data is as follows (don't save):

- Cell Size: 14.25

- Number of Columns: about 712

- Number of Rows: about 506

After confirming the structure, click the Cancel button to close the window.

5. Open the Properties of the layer N-45-40.sid > click the Symbology tab > make
sure the Show type is RGB Composite > select Stretch Type to Standard
Deviations > set the number of n to 0.01 > click OK.

You will find that the layer of N-45-40.sid changes its expression from full color to
something like a color classified map.

6. Make the reference feature of nss_gridbase inactive.

7. The next step is to convert the MrSID expression to an image using the map export
function with which you have become familiarized. However, here you have to
specify some factors that are different from what you set in your assignment.

- Set the location as your Temp folder

3-73
 - Set name and format of the output image to Sp_image.jpg

- Click the Format tab and then select the 8-bit grayscale as Color Mode

- Click the General tab > adjust the Resolution to make the width and
height of the pixels near the numbers of column and row that you checked
in Step 4.

- Don’t forget to put a check in the box for Write World File, otherwise the
exported result will not match other data in your data frame.

8. Add the exported image Sp_image.jpg. Check the value of the area that

corresponds to the farmland by using the Identify Tool . You will find the value

of the farmland area in this layer is 150.

9. Displya the ArcTool box window if it is hidden.

10. Expand the Spatial Analyst menu > expand Extraction > open the Extract by
Attribute tool. The corresponding tool window will appear.

11. In the Extract by Attributes window, set the necessary factors as follows:

- Input raster: Sp_image.jpg

- Output location: your Temp folder
- Output name: Sp_fmland
- Click the SQL (Structured Query Language) button next to the text box of
where clause, and the Query Builder window will appear.
- Based on the value that was obtained using the Identify Tool in Step 8,
input the following equation into the text box of SELECT * FROM
Sp_Image WHERE:.
Value =150

3-74
 - Click the OK button twice to perform the extraction.

12. The extracted result well be added into the data frame.

Farmland Polygon Feature Creation

If you make all layers inactive except the newly created layer of Sp_fmland, you will
find the farmland area has been extracted, but it contains many small slivers. One
method to reduce these slivers is to resample the image.

1. Expand the toolbox menu of Data Management Tool, expand Raster > open the
Resample tool. The corresponding window will appear.

2. In the Resample window, input the following settings:

- Input raster: Sp_fmland

- Location of output raster: your Temp folder

- Name of output raster: Sp_farmland

- Output cell size: 30

After entering the above settings, click the OK button.

3. The resampled image will be added into your data frame.

4. Expand the toolbox menu of Conversion Tools > expand From Raster > open the
Raster to Polygon tool.

5. In the Raster to Polygon window, input the following settings:

3-75
 - Input raster: Sp_farmland

- Field (optional): Value

- Output location: to your Result folder

- Output polygon feature name: Sp_farmland

Notice: Uncheck the box of Simplify polygons (optional)

6. Make the reference layer nss_gridbase active, set the symbology of the feature
layer Sp_farmland as red for fill; no color for outline; set the symbology of
MrSID layer to its default setting. The result of farmland area extraction from
MrSID data will appear as shown in the figure below.

Farmland Area Extraction by Raster Calculation (optional)

1. Make sure the symbol of source raster data N-45-40_2000.sid is set to default.
(RGB Composite for Show and None for Stretch Type)

2. Make sure the Data View extent is the same as in your last exercise.

3. Check the extent of groundwater simulation model area by opening the Properties of
layer nss_gridbase > click Source tab.

3-76
 As shown in the figure below, the limit of the groundwater simulation area is:

759000 for left, 4756000 for top, 769000 for right and 4749000 for bottom

4. Display the Spatial Analyst toolbar.

5. Click the dropdown arrow next to Spatial Analyst > click Options > click the
Extent tab > click the dropdown arrow next to the textbox of Analysis extent and
then select As Specified Below > use the results from checking the extent to define
the analysis extent > click the OK button.

6. Make all layers inactive except the layer N-45-40_2000.sid.

7. Export your map with factors specified as follows:

- Output name: N_45_40_image

- Type: JPEG

- Format: 24-bit true Color

- Resolution: the same as Sp_iamge

- Write World File: checked

3-77
8. Click the Add Data button > browse to the Temp folder > double-click the
newly exported raster data N_45_40_image > expand the image and then select all
three (3) bands > click the Add button. The three bands which represent the colors
red, green and blue are added into your data frame.

9. Check the value correspondence of Band 1.

- Make layers of Band 2 and Band 3 inactive, then you can confirm that the
relatively small value in Band 1 (red) corresponds to the farmland area.

- Open the Properties of Band 1

- Set the symbology of Band 1 to Classified

- Set the Classes to 2

- Set the color of red for Class 1 (class for small value range), no color for
Class 2.

- Adjust the Break Value until almost all farmland areas are covered by
Class 1 of Band 1.

- Make note of the Break Value.

10. Repeat Step 9 for Band 2 and Band 3

11. Click the dropdown arrow next to Spatial Analyst > click the Raster Calculator.
The corresponding window will appear. Referring to the equations in the following
figure, create farmland calculation equations. The values for extraction from each
band should be based on notes that you will have taken in Steps 9 and 10.

3-78
 The basic rules for raster calculation are as follows:

- one equation has to be limited within one line

- in an equation, all marks and values have to be separated by a space

12. All the equations you build in the Raster Calculator will create a raster data layer.
As the result of the calculation, 4 raster data layers of B1, B2, B3 and Farmland
are added into your data frame. B1 to B3 are interim results, which are not so useful
so you can remove them from your data frame.

13. The symbol of Farmland is automatically set as Unique Value. You can set all
values to no color except value of 3. According to your calculation, the value of the
raster data Farmland is as follows:

- Value 1: area selected in only one of the three bands.

- Value 2: area selected in two of the three bands

- Value 3: area selected from all the three bands

The result should look like the figure shown below.

You will find this result obtained by raster calculation is a little better than the
result of the raster data simplifying method.

3-79
14. Referring to the previous exercise, complete the farmland polygon feature creation:

- Extract the farmland raster data by using the tool Extract by Attribute in
the ArcToolbox Spatial Analyst/ Extraction

- Convert the extracted farmland raster data to polygon feature by using the
Raster to Polygon tool in the ArcToolbox Conversion Tools/ From
Raster

3-80