A Gentle Introduction to GIS

Brought to you with Quantum GIS, a Free and Open Source Software GIS Application for everyone.

T. Sutton, O. Dassau, M. Sutton sponsored y! "hief #irectorate! Spatial $lanning % Information, #epartment of &and Affairs, 'astern "ape, South Africa.

in partnership with! Spatial Information (anagement )nit, Office of the $remier, 'astern "ape, South Africa.

Copyright (c)

2009

Chief Directorate: Spatial Planning & Information, Department of Land ffair!, "a!tern Cape#

Permi!!ion i! granted to copy, di!tri$%te and&or modify thi! doc%ment %nder the term! of the '() *ree Doc%mentation Licen!e, +er!ion ,#2 or any later -er!ion p%$li!hed $y the *ree Soft.are *o%ndation/ .ith no In-ariant Section!, no *ront0Co-er 1e2t!, and no 3ac40Co-er 1e2t!# copy of the licen!e i! incl%ded in the !ection entitled 5'() *ree Doc%mentation Licen!e5#

1he a$o-e copyright notice e2cl%de! the 6'IS )!er 7an%al .hich may $e appended to thi! doc%ment# Con!%lt the 6'IS 7an%al! for f%rther copyright and licen!ing information#

A word from the editor: *his pro+ect was sponsored y the "hief #irectorate! Spatial $lanning % Information, #epartment of &and Affairs ,#&A-, 'astern "ape, in con+unction with the Spatial Information (anagement )nit, Office of the $remier, 'astern "ape, South Africa. GIS is ecoming an increasingly important tool in environmental management, retail, military, police, tourism and many other spheres of our daily lives. If you use a computer or a cell phone, you have pro a ly already used a GIS in some form without even realising it. (ay e it was a map on a we site, Google 'arth, an information ooth or your cell phone telling you where you are. $roprietary GIS software ,software that cannot e freely shared or modified- is availa le that will let you do everything we descri e in these wor.sheets and a lot more. /owever this software is usually very e0pensive or otherwise limits your freedom to copy, share and modify the software. GIS vendors sometimes ma.e an e0ception for educational activities, providing cheaper or free copies of their software. *hey do this .nowing that if teachers and learners get to .now their software, they will e reluctant to learn other pac.ages. 1hen learners leave school they will go into the wor.place and uy the commercial software, never .nowing that there are free alternatives that they could e using. 1ith Quantum GIS, we offer an alternative 2 software that is free of cost and free in a social sense. 3ou can ma.e as many copies as you li.e. 1hen learners leave school one day they can use this software to uild their s.ills, solve pro lems at wor. and ma.e the world a etter place. 1hen you uy commercial software, you limit your options for the future. By learning, using and sharing Free and Open Source Software, you are uilding your own s.ills, freeing money to e spent on important things li.e food and shelter and oosting our own economy. By sponsoring the creation of this resource, the #&A has created a foundation to which young minds can e e0posed. '0citing possi ilities lie ahead when principles of free sharing of .nowledge and data are em raced. For this we give our heartfelt than.s to the #&A4 1e hope you en+oy using and learning QGIS in the spirit of ) untu4

*im Sutton, April 5667

Table of Contents
*opic 8! Introducing GIS..........................................................................5 *opic 5! 9ector #ata..............................................................................86 *opic :! 9ector Attri ute #ata.................................................................58 *opic ;! #ata "apture............................................................................:; *opic <! =aster #ata..............................................................................;> *opic >! *opology..................................................................................<< *opic ?! "oordinate =eference Systems....................................................>6 *opic @! (ap $roduction.........................................................................?5 *opic 7! 9ector Spatial Analysis ,Buffers-.................................................@6 *opic 86! Spatial Analysis ,Interpolation-.................................................@? A out the authors % contri utors!...........................................................7: GA) Free #ocumentation &icense............................................................7< QGIS )ser (anual...............................................................................865

GIS for 'ducators

Topic 1: Introducing GIS O +ectives! )nderstanding what GIS is and what it can e used for. GIS, "omputer, (aps, #ata, Information System, Spatial, Analysis

Beywords!

Overview: Cust as we use a word processor to write documents and deal with words on a computer, we can use a GIS application to deal with spatial information on a computer. GIS stands for Geo!raphical Information S"stem . A GIS consists of!

Di!ital Data D the geographical information that you will view and analyse using computer hardware and software. #omputer $ardware D computers used for storing data, displaying graphics and processing data. #omputer Software D computer programs that run on the computer hardware and allow you to wor. with digital data. A software program that forms part of the GIS is called a GIS Application.

1ith a GIS application you can open digital maps on your computer, create new spatial information to add to a map, create printed maps customised to your needs and perform spatial analysis. &etEs loo. at a little e0ample of how GIS can e useful. Imagine you are a health wor.er and you ma.e a note of the date and place of residence of every patient you treat.

&ongitude 5>.@?6;:> 5>.@>@>@5 5>.@>??6? 5>.@<;76@ 5>.@<<@8? 5>.@<5?>; 5>.@<;??@ 5>.@>76?5

&atitude 2:8.767<87 2:8.7675<7 2:8.786;7; 2:8.756?<7 2:8.758757 2:8.758757 2:8.75<885 2:8.7887@@

#isease (umps (umps (umps (easles (easles (easles (easles (umps

#ate 8:F85F566@ 5;F85F566@ 55F68F5667 88F68F5667 5>F68F5667 86F65F5667 55F65F5667 65F65F5667

5>.@>::<;

2:8.78>;6>

"hic.en $o0

5>F65F5667

If you loo. at the ta le a ove you will Guic.ly see that there were a lot of measles cases in Canuary and Fe ruary. Our health wor.er recorded the location of each patientEs house y noting its latitude and longitude in the ta le. )sing this data in a GIS Application, we can Guic.ly understand a lot more a out the patterns of illness!

Illustration 1: Example showing disease records in a GIS application. It is easy to see that the mumps patients all live close to each other. More a%out GIS: GIS is a relatively new field 2 it started in the 87?6Es. It used to e that computerised GIS was only availa le to companies and universities that had e0pensive computer eGuipment. *hese days, anyone with a personal computer or laptop can use GIS software. Over time GIS Applications have also ecome easier to use D it used to reGuire a lot of training to use a GIS Application, ut now it is much easier to get started in GIS even for amateurs and casual users. As we descri ed a ove, GIS is more than +ust software, it refers to all aspects of managing and using digital geographical data. In the tutorials that follow we :

will e focusing on GIS Software.

&hat is GIS Software ' a GIS Application(: 3ou can see an e0ample of what a GIS Application loo.s li.e in Illustration 8 a ove. GIS Applications are normally programs with a graphical user interface that can e manipulated using the mouse and .ey oard. *he application provides menus near to the top of the window ,File, 'dit etc.- which, when clic.ed using the mouse, show a panel of actions. *hese actions provide a way for you to tell the GIS Application what you want to do. For e0ample you may use the menus to tell the GIS Application to add a new layer to the display output.

Illustration 2: Application menus, when clic ed with the mouse, expand to show a list o! actions that can "e carried out. Tool%ars ,rows of small pictures that can e clic.ed with the mouse- normally sit +ust elow the menus and provide a Guic.er way to use freGuently needed actions.

Illustration #: $ool"ars provide %uic access to commonly used !unctions. &olding your mouse over a picture will usually tell you what will happen when you clic on it. A common function of GIS Applications is to display map la"ers. (ap layers are stored as files on a dis. or as records in a data ase. Aormally each map layer will represent something in the real world D a roads layer for e0ample will have data a out the street networ.. 1hen you open a layer in the GIS Application it will appear in the map view. ;

*he map view shows a graphic representing your layer. 1hen you add more than one layer to a map view, the layers are overlaid on top of each other. Illustrations ; to ? elow show a map view that has several layers eing added to it. An important function of the map view is to allow you to Hoom in to magnify, Hoom out to see a greater area and move around ,panning- in the map.

Illustration (: A towns layer added to the map view.

Illustration ': A schools layer added to the map view.

Illustration ): A railways layer added to the map view.

Illustration *: A rivers layer added to the map view.

)nli.e paper maps, the maps displayed in GIS Applications can e changed after they have een created. 3ou can change the s"m%olo!" of the map layers to ma.e them appear in different colours or sym ols. For e0ample, if we ta.e the map in Illustration ? and change the sym ology, we can completely change how it loo.s D as shown in Illustration @ elow. Sym ology plays an important role in how we interpret maps, and GIS Applications are very good at letting you change sym ology Guic.ly and easily.

Illustration +: GIS So!tware let you easily change sym"ology , the way in!ormation is displayed. <

Another common feature of GIS Applications is the map le!end. *he map legend provides a list of layers that have een loaded in the GIS Application. )nli.e a paper map legend, the map legend or Elayers listE in the GIS Application provides a way to re2order, hide, show and group layers. "hanging the layer order is done y clic.ing on a layer in the legend, holding the mouse utton down and then dragging the layer to a new position. In Illustrations 7 and 86 elow, the map legend is shown as the area to the left of the GIS Application window. By changing the layer order, the way that layers are drawn can e ad+usted D in this case so that rivers are drawn over the roads instead of elow them.

Illustration -: .e!ore changing the layer order, rivers are drawn underneath roads.

Illustration 1/: A!ter changing the layer order, rivers are drawn on top o! roads.

Gettin! a GIS Application for "our own computer)s*: *here are many different GIS Applications availa le. Some have many sophisticated features and cost tens of thousands of =ands for each copy. In other cases, you can o tain a GIS Application for free. #eciding which GIS Application to use is a Guestion of how much money you can afford and personal preference. For these tutorials, we will e using the Quantum GIS Application, also .nown as QGIS. Quantum GIS is completely free and you can copy it and share it with your friends as much as you li.e. If you received this tutorial in printed form, you should have received a copy of QGIS with it. If not, you can always visit http!FFGgis.org to download your free copy if you have access to the internet. GIS Data: Aow that we .now what a GIS is and what a GIS Application can do, letEs tal. a out GIS data. #ata is another word for information. *he information we >

use in a GIS normally has a geographical aspect to it. *hin. of our e0ample a ove, a out the health care wor.er. She created a ta le to record diseases that loo.ed li.e this!

&ongitude 5>.@?6;:>

&atitude 2:8.767<87

#isease (umps

#ate 8:F85F566@

*he longitude and latitude columns hold !eo!raphical data. *he disease and date columns hold non+!eo!raphical data. A common feature of GIS is that they allow you to associate information ,non2geographical data- with places ,geographical data-. In fact, the GIS Application can store many pieces of information which are associated with each place D something that paper maps are not very good at. For e0ample, our health care wor.er could store the personEs age and gender on her ta le. 1hen the GIS Application draws the layer, you can tell it to draw the layer ased on gender, or ased on disease type, and so on. So, with a GIS Application we have a way to easily change the appearance of the maps we created ased on the non2geographical data associated with places. GIS Systems wor. with many different types of data. ,ector data is stored as a series of I,3 coordinate pairs inside the computerEs memory. 9ector data is used to represent points, lines and areas. Illustration 88 elow shows different types of vector data eing viewed in a GIS application. In the tutorials that follow we will e e0ploring vector data in more detail.

Illustration 11: 0ector data is used to represent points 1e.g. towns2, lines 1e.g. rivers2 and polygons 1e.g. municipal "oundaries2. -aster data are stored as a grid of values. *here are many satellites circling ?

the earth and the photographs they ta.e are a .ind of raster data that can e viewed in a GIS. One important difference etween raster and vector data is that if you Hoom in too much on a raster image, it will start to appear E loc.yE ,see illustrations 85 and 8: elow-. In fact these loc.s are the individual cells of the data grid that ma.es up the raster image. 1e will e loo.ing at raster data in greater detail in later tutorials.

Illustration 12: 3aster data are o!ten images ta en "y satellites. &ere we can see mountains in the Eastern 4ape. &hat have we learned(

Illustration 1#: $he same raster data, "ut this time 5oomed in. $he grid nature o! the data can "e seen.

&etEs wrap up what we covered in this wor.sheet! A GIS is a system of computer hardware, computer software and geographical data. A GIS Application allows you to view geographical data and is an important part of the GIS. A GIS Application normally consists of a menu %ar, tool%ars, a map view and a le!end. ,ector and raster data are geographical data used in a GIS application. Geo!raphical data can have associated non+!eo!raphical data. .ow "ou tr"/ /ere are some ideas for you to try with your learners!

Geo!raph": #escri e the concept of GIS to your learners as outlined in this tutorial. As. them to try to thin. of : reasons why it might e handy to use a GIS instead of paper maps. /ere are some that we could thin. of! GIS Applications allow you to create many different types of maps from the same data. GIS is a great visualisation tool that can show you things a out your data and how they are related in space ,e.g. those disease out rea.s @

we saw earlier-. $aper maps need to e filed and are time consuming to view. *he GIS can hold a very large amount of map data and ma.e it Guic. and easy to find a place you are interested in.

Geo!raph": "an you and your learners thin. of how raster data from satellites could e usefulJ /ere are some ideas we had! #uring natural disasters, raster data can e useful to show where the impacted areas are. For e0ample a recent satellite image ta.en during a flood can help to show where people may need rescuing. Sometimes people do ad things to the the environment, li.e dumping dangerous chemicals that .ill plants and animals. )sing raster data from satellites can help us to monitor for these type of pro lems. *own planners can use raster data from satellites to see where informal settlements are and to help in planning infrastructure.

Somethin! to thin0 a%out: If you donEt have a computer availa le, many of the topics we cover in this tutorial can e reproduced using an overhead and transparency as it uses the same techniGue of layering information. /owever, to properly understand GIS it is always etter to learn it using a computer. 1urther readin!: 2oo0: #es.top GIS! (apping the $lanet with Open Source *ools. Author: Gary Sherman. IS2.: 7?@87:;:<>6>? &e%site: http!FFwww.gisdevelopment.netFtutorialsFtuman66>.htm *he QGIS )ser Guide also has more detailed information on wor.ing with QGIS. &hat s ne3t( In the sections that follow we are going to go into more detail, showing you how to use a GIS Application. All of the tutorials will e done using QGIS. Ae0t up, letEs loo. at vectors4

GIS for 'ducators

Topic 4: Vector Data O +ectives! Beywords! )nderstanding of vector data models as used in GIS. 9ector, $oint, $olyline, $olygon, 9erte0, Geometry, Scale, #ata Quality, Sym ology, #ata Sources

Overview: ,ector data provide a way to represent real world features within the GIS environment. A feature is anything you can see on the landscape. Imagine you are standing on the top of a hill. &oo.ing down you can see houses, roads, trees, rivers, and so on ,see Illustration 8; elow-. 'ach one of these things would e a feature when we represent them in a GIS Application. 9ector features have attri%utes, which consist of te0t or numerical information that descri%e the features.

Illustration 1(: 6oo ing over a landscape you can see the main !eatures, such as roads, houses and trees. A vector feature has its shape represented using !eometr". *he geometry is made up of one or more interconnected vertices. A verte0 descri es a position in space using an 3, " and optionally 5 a0is. Geometries which have vertices with a H a0is are often referred to as 4.6D since they descri e height or depth at each verte0, ut not oth. 86

1hen a featureEs geometry consists of only a single verte0, it is referred to as a point feature ,see Illustration 8< elow-. 1here the geometry consists of two or more vertices and the first and last verte0 are not eGual, a pol"line feature is formed ,see Illustration 8> elow-. 1here four or more vertices are present, and the last verte0 is eGual to the first, an enclosed pol"!on feature is formed ,see Illustration 8? elow-.

Illustration 1': A point !eature is descri"ed "y its 7, 8 and optionally 9 coordinate. $he point attri"utes descri"e the point e.g. i! it is a tree or a lamp post.

Illustration 1): A polyline is a se%uence o! :oined vertices. Each vertex has an 7, 8 1and optionally 92 coordinate. Attri"utes descri"e the polyline.

Illustration 1*: A polygon, li e a polyline, is a se%uence o! vertices. &owever in a polygon, the !irst and last vertices are always at the same position.

&oo.ing ac. at the picture of a landscape we showed you further up, you should e a le to see the different types of features in the way that a GIS represents them now ,see Illustration 8@ elow-. 7oint features in detail: *he first thing we need to realise when tal.ing a out point features is that what we descri e as a point in GIS is a matter of opinion, and often dependent on scale. letEs loo. at cities for e0ample. If you have a small scale map ,which covers a large area-, it may ma.e sense to represent a city using a point feature. /owever as you Hoom in to the map, moving towards a larger scale, it ma.es more sense to show the city limits as a polygon. 1hen you choose to use points to represent a feature is mostly a matter of scale ,how far away are you from the feature-, convenience ,it ta.es less time and effort to create point features than polygon features-, and the type of 88

feature ,some things li.e telephone poles +ust donEt ma.e sense to e stored as polygons-.

Illustration 1+: 6andscape !eatures as we would represent them in a GIS. 3ivers 1"lue2 and roads 1green2 can "e represented as lines, trees as points 1red2 and houses as polygons 1white2. As we show in Illustration 8<, a point feature has an I,3 and optionally, K value. *he I and 3 values will depend on the #oordinate -eference S"stem ,"=S- eing used. 1e are going to go into more detail a out "oordinate =eference Systems in a later tutorial. For now letEs simply say that a "=S is a way to accurately descri e where a particular place is on the earthEs surface. One of the most common reference systems is 8on!itude and 8atitude. &ines of &ongitude run from the Aorth $ole to the South $ole. &ines of &atitude run from the 'ast to 1est. 3ou can descri e precisely where you are at any place on the earth y giving someone your &ongitude ,I- and &atitude ,3-. If you ma.e a similar measurement for a tree or a telephone pole and mar.ed it on a map, you will have created a point feature. Since we .now the earth is not flat, it is often useful to add a K value to a point feature. *his descri es how high a ove sea level you are. 7ol"line features in detail: 1here a point feature is a single verte0, a pol"line has two or more vertices. *he polyline is a continuous path drawn through each verte0, as shown in Illustration 8> a ove-. 1hen two vertices are +oined, a line is created. 1hen more than two are +oined, they form a Eline of linesE, or pol"line. 85

A polyline is used to show the geometry of linear features such as roads, rivers, contours, footpaths, flight paths and so on. Sometimes we have special rules for polylines in addition to their asic geometry. For e0ample contour lines may touch ,e.g. at a cliff face- ut should never cross over each other. Similarly, polylines used to store a road networ. should e connected at intersections. In some GIS applications you can set these special rules for a feature type ,e.g. roads- and the GIS will ensure that these polylines always comply to these rules. If a curved polyline has very large distances etween vertices, it may appear an!ular or +agged, depending on the scale at which it is viewed ,see Illustration 87 elow-. Because of this it is important that polylines are digitised ,captured into the computer- with distances etween vertices that are small enough for the scale at which you want to use the data. *he attri%utes of a polyline decri e its properties or characteristics. For e0ample a road polyline may have attri utes that descri e whether it is surfaced with gravel or tar, how many lanes it has, whether it is a one way street, and so on. *he GIS can use these attri utes to sym olise the polyline feature with a suita le colour or line style.

Illustration 1-: ;olylines viewed at a smaller scale 11:2/ /// to the le!t2 may appear smooth and curved. <hen 5oomed in to a larger scale 11:'// to the right2 polylines may loo very angular. 7ol"!on features in detail: $olygon features are enclosed areas li.e dams, islands, country oundaries and so on. &i.e polyline features, polygons are created from a series of vertices that are connected with a continuous line. /owever ecause a polygon always descri es an enclosed area, the first and last vertices should always e at the same place4 $olygons often have shared !eometr" D oundaries that are in common with a neigh ouring polygon. (any GIS applications have the capa ility to ensure that the oundaries of neigh ouring polygons e0actly 8:

coincide. 1e will e0plore this in the topolo!" topic later in this tutorial. As with points and polylines, polygons have attri%utes. *he attri utes descri e each polygon. For e0ample a dam may have attri utes for depth and water Guality. ,ector data in la"ers: Aow that we have descri ed what vector data is, letEs loo. at how vector data is managed and used in a GIS environment. (ost GIS applications group vector features into la"ers. Features in a layer have the the same geometry type ,e.g. they will all e points- and the same .inds of attri utes ,e.g. information a out what species a tree is for a trees layer-. For e0ample if you have recorded the positions of all the footpaths in your school, they will usually e stored together on the computer hard dis. and shown in the GIS as a single layer. *his is convenient ecause it allows you to hide or show all of the features for that layer in your GIS application with a single mouse clic.. 9ditin! vector data: *he GIS application will allow you to create and modify the geometry data in a layer D a process called di!itisin! D which we will loo. at more closely in a later tutorial. If a layer contains polygons ,e.g. farm dams-, the GIS application will only allow you to create new polygons in that layer. Similarly if you want to change the shape of a feature, the application will only allow you to do it if the changed shape is correct. For e0ample it wonEt allow you to edit a line in such a way that it has only one verte0 D remem er in our discussion of lines a ove that all lines must have at least two vertices. "reating and editing vector data is an important function of a GIS since it is one of the main ways in which you can create personal data for things you are interested in. Say for e0ample you are monitoring pollution in a river. 3ou could use the GIS to digitise all outfalls for storm water drains ,as point features-. 3ou could also digitise the river itself ,as a polyline feature-. Finally you could ta.e readings of p/ levels along the course of the river and digitise the places where you made these readings ,as a point layer-. As well as creating your own data, there is a lot of free vector data that you can o tain and use. For e0ample, you can o tain vector data that appears on the 8!<6 666 map sheets from the "hief #irectorate ! Surveys and (apping. Scale and vector data: (ap scale is an important issue to consider when wor.ing with vector data in a GIS. 1hen data is captured, it is usually digitised from e0isting maps, or y ta.ing information from surveyor records and glo al positioning system devices. (aps have different scales, so if you import vector data from a map 8;

into a GIS environment ,for e0ample y digitising paper maps-, the digital vector data will have the same scale issues as the original map. *his effect can e seen in Illustrations 56 and 58 elow. (any issues can arise from ma.ing a poor choice of map scale. For e0ample using the vector data in Illustration Illustration 56 elow- to plan a wetland conservation area could result in important parts of the wetland eing left out of the reserve4 On the other hand if you are trying to create a regional map, using data captured at 8!8666 666 might e +ust fine and will save you a lot of time and effort capturing the data.

Illustration 2/: 0ector data 1red lines2 that was digitised !rom a small scale 11:1/// ///2 map. S"m%olo!":

Illustration 21: 0ector data 1green lines2 that was digitised !rom a large scale 11:'/ ///2 map.

1hen you add vector layers to the map view in a GIS application, they will e drawn with random colours and asic sym ols. One of the great advantages of using a GIS is that you can create personalised maps very easily. *he GIS program will let you choose colours to suite the feature type ,e.g. you can tell it to draw a water odies vector layer in lue-. *he GIS will also let you ad+ust the sym ol used. So if you have a trees point layer, you can show each tree position with a small picture of a tree, rather than the asic circle mar.er that the GIS uses when you first load the layer ,see Illustrations 55,5: % 5; elow-.

8<

Illustration 22: <hen a layer 1!or example the trees layer a"ove2 is !irst loaded, a GIS application will give it a generic sym"ol.

Illustration 2#: In the GIS, you can use a panel 1li e the one a"ove2 to ad:ust how !eatures in your layer should "e drawn.

8>

Illustration 2(: A!ter ma ing our ad:ustments it is much easier to see that our points represent trees. Sym ology is a powerful feature, ma.ing maps come to life and the data in your GIS easier to understand. In the topic that follows ,wor.ing with attri ute data- we will e0plore more deeply how sym ology can help the user to understand vector data. &hat can we do with vector data in a GIS(: At the simplest level we can use vector data in a GIS Application in much the same way you would use a normal topographic map. *he real power of GIS starts to show itself when you start to as. Guestions li.e Ewhich houses are within the 866 year flood level of a riverJEL Ewhere is the est place to put a hospital so that it is easily accessi le to as many people as possi leJEL Ewhich learners live in a particular su ur JE. A GIS is a great tool for answering these types of Guestions with the help of vector data. Generally we refer to the process of answering these types of Guestions as spatial anal"sis. In later topics of this tutorial we will loo. at spatial analysis in more detail. #ommon pro%lems with vector data: 1or.ing with vector data does have some pro lems. 1e already mentioned the issues that can arise with vectors captured at different scales. 9ector data also needs a lot of wor. and maintenance to ensure that it is accurate and relia le. Inaccurate vector data can occur when the instruments used to capture the data are not properly set up, when the people capturing the data arenEt eing careful, when time or money donEt allow for enough detail in the collection process, and so on. If you have poor Guality vector data, you can often detect 8?

this when viewing the data in a GIS. For e0ample slivers can occur when the edges of two polygon areas donEt meet properly ,see Illustration 5< elow-. Overshoots can occur when a line feature such as a road does not meet another road e0actly at an intersection. :ndershoots can occur when a line feature ,e.g. a river- does not e0actly meet another feature to which it should e connected. Illustration 5> elow demonstrates what undershoots and overshoots loo. li.e. Because of these types of errors, it is very important to digitise data carefully and accurately. In the upcoming topic on topolo!", we will e0amine some of these types of errors in more detail.

Illustration 2': Slivers occur when the vertices o! two polygons do not match up on their "orders. At a small scale 1e.g. 1 on le!t2 you may not "e a"le to see these errors. At a large scale they are visi"le as thin strips "etween two polygons 12 on right2.

8@

Illustration 2): =ndershoots 112 occur when digitised vector lines that should connect to each other don>t %uite touch. ?vershoots 122 happen i! a line ends "eyond the line it should connect to.

&hat have we learned( &etEs wrap up what we covered in this wor.sheet! ,ector data is used to represent real world features in a GIS. A vector feature can have a !eometr" type of point, line or a pol"!on. 'ach vector feature has attri%ute data that descri es it. Feature geometry is descri ed in terms of vertices. $oint geometries are made up of a sin!le verte3 ,I,3 and optionally K-. $olyline geometries are made up of two or more vertices forming a connected line. $olygon geometries are made up of at least four vertices forming an enclosed area. *he first and last vertices are always in the same place. "hoosing which geometry type to use depends on scale, convenience and what you want to do with the data in the GIS. (ost GIS applications do not allow you to mi0 more than one geometry type in a single layer. #igitising is the process of creating digital vector data y drawing it in a GIS application. 9ector data can have Guality issues such as undershoots, overshoots and slivers which you need to e aware of. 87

9ector data can e used for spatial anal"sis in a GIS application, for e0ample to find the nearest hospital to a school. 1e have summarised the GIS 9ector #ata concept in Illustration 5? elow.

Illustration 2*: $his diagram shows how GIS applications deal with vector data. .ow "ou tr"/ /ere are some ideas for you to try with your learners!

)sing a copy of a toposheet map for your local area ,li.e the one shown in Illustration 5@ elow-, see if your learners can identify e0amples of the different types of vector data y highlighting them on the map. *hin. of how you would create vector features in a GIS to represent real world features on your school grounds. "reate a ta le of different features in and around your school and then tas. your learners to decide 56

whether they would e est represented in the GIS as a point, line or polygon. See *a le 8 elow for an e0ample.

Illustration 2+: 4an you identi!y two point !eatures, !our line !eatures and one polygon !eature on this map@

=eal world feature *he school flagpole *he soccer field *he footpaths in and around the school $laces where taps are located 'tc.

Suita le Geometry *ype

$a"le 1: 4reate a ta"le li e this 1leaving the geometry type column empty2 and as your learners to decide on suita"le geometry types. Somethin! to thin0 a%out: If you donEt have a computer availa le, you can use a toposheet and transparency sheets to show your learners a out vector data. 1urther readin!: *he QGIS )ser Guide also has more detailed information on wor.ing with vector data in QGIS. &hat s ne3t( In the section that follows we will ta.e a closer loo. at attri%ute data to see 58

how it can e used to descri e vector features.

55

GIS for 'ducators

Topic ;: Vector Attribute Data O +ectives! In this topic we descri e how attri ute data are associated with vector features and can e used to sym olise data. Attri ute, data ase, fields, data, vector, sym ology

Beywords!

Overview: If every line on a map was the same colour, width, thic.ness, and had the same la el, it would e very hard to ma.e out what was going on. *he map would also give us very little information. *a.e a loo. at Illustration 57 elow for e0ample.

Illustration 2-: Aaps come to li!e when colour and di!!erent sym"ols are used to help you to tell one type o! !eature !rom the next. 4an you tell the di!!erence "etween rivers, roads and contours using the map on the le!t@ =sing the map on the right it is much easier to see the di!!erent !eatures. In this topic we will loo. at how attri ute data can help us to ma.e interesting and informative maps. In the previous topic on vector data, we riefly e0plained that attri%ute data are used to descri%e vector features. *a.e a loo. at the house pictures in Illustration :6 elow. *he geometry of these house features is a polygon , ased on the floor plan of the house-, the attri utes we have recorded are roof colour, whether there is a alcony, and the year the house was uilt. Aote that attri utes donEt have to e visi le things D they can descri e things we .now a out the feature such as the year it was uilt. In a GIS Application, we can represent this feature type in a houses polygon layer, and the attri utes in an attri ute ta le ,see Illustration :8 elow-.

5:

Illustration #/: Every !eature has characteristics that we can descri"e. $hese can "e visi"le things, or things we now a"out the !eature 1e.g. year "uilt2.

5;

Illustration #1: A houses layer. &ouse !eatures have attri"utes that descri"e the houses> roo! colour and other properties. $he attri"ute ta"le 1lower image2 lists the attri"utes !or the house areas shown on the map. <hen a !eature is highlighted in the ta"le, it will appear as a yellow polygon on the map. *he fact that features have attri utes as well geometry in a GIS Application opens up many possi ilities. For e0ample we can use the attri ute values to tell the GIS what colours and style to use when drawing features ,see Illustration :5 elow-. *he process of setting colours and drawing styles is often referred to as setting feature s"m%olo!". Attri ute data can also e useful when creating map la%els. (ost GIS Applications will have a facility to select an attri ute that should e used to la el each feature.

5<

Illustration #2: In a GIS Application, we can draw !eatures di!!erently depending on their attri"utes. ?n the le!t we have drawn house polygons with the same colour as the roo! attri"ute. ?n the right we colour coded houses according to whether they have a "alcony or not. If you have ever searched a map for a place name or a specific feature, you will .now how time consuming it can e. /aving attri ute data can ma.e searching for a specific feature Guic. and easy. In Illustration :: elow you can see an e0ample of an attri ute search in a GIS. Finally, attri ute data can e very useful in carrying out spatial anal"sis. Spatial analysis com ines the spatial information stored in the geometry of features with their attri ute information. *his allows us to study features and how they relate to each other. *here are many types of spatial analysis that can e carried out, for e0ample, you could use GIS to find out how many red roofed houses occur in a particular area. If you have tree features, you could use GIS to try to find out which species might e affected if a piece of land is developed. 1e can use the attri utes stored for water samples along a river course to understand where pollution is entering into the stream. *he possi ilities are endless4 In a later topic we will e e0ploring spatial analysis in more detail. Before we move on to attri ute data in more detail, letEs ta.e a Guic. recap! Features are real world things such as roads, property oundaries, electrical su station sites and so on. A feature has a !eometr" ,which determines if it is a point, pol"line or pol"!on- and attri%utes ,which descri e the feature-. *his is shown in Illustration :; elow.

5>

Illustration ##: In a GIS Application, we can also search !or !eatures "ased on their attri"utes. &ere we see a search !or houses with "lac roo!s. 3esults are shown in yellow in the map, tur%uoise on the ta"le.

Illustration #(: 0ector !eatures at a glance.

5?

Attri%utes in detail: Attri utes for a vector feature are stored in a ta%le. A ta le is li.e a spreadsheet. 'ach column in the ta le is called a field. 'ach row in the ta le is a record. *a le 5 elow Shows a simple e0ample of how an attri ute ta le loo.s in a GIS. *he records in the attri ute ta le in a GIS each correspond to one feature. )sually the information in the attri ute ta le is stored in some .ind of data ase. *he GIS application lin.s the attri ute records with the feature geometry so that you can find records in the ta le y selecting features on the map, and find features on the map y selecting features in the ta le. Attri ute *a le =ecord 8 =ecord 5 =ecord : Field 8 ! 3earBuilt 877@ 5666 5668 Field 5! =oof"olour =ed Blac. Silver Field :! Balcony 3es Ao 3es

$a"le 2: An attri"ute ta"le has !ields 1columns2 and records 1in rows2. 'ach field in the attri ute ta le contains contains a specific type of data D te0t, numeric or date. #eciding what attri utes to use for a feature reGuires some thought and planning. In our house e0ample earlier on in this topic, we chose roof colour, presence of a alcony and month of construction as attri utes of interest. 1e could +ust as easily have chosen other aspects of a house such as!

num er of levels num er of rooms num er of occupants type of dwelling ,=#$ /ouse, loc. of flats, shac., ric. house etcyear the house was uilt area of floor space in the house and so on....

1ith so many options, how do we ma.e a good choice as to what attri utes are needed for a featureJ It usually oils down to what you plan to do with the data. If you want to produce a colour coded map showing houses y age, it will ma.e sense to have a E3ear BuiltE attri ute for your feature. If you .now for sure you will never use this type of map, it is etter to not store the information. "ollecting and storing unneeded information is a ad idea ecause of the cost and time reGuired to research and capture the information. 9ery often we o tain vector data from companies, friends or the government. In these cases it is usually not possi le to reGuest specific attri utes and we have to ma.e do with what we get. Sin!le S"m%ols:

5@

If a feature is sym olised without using any attri ute ta le data, it can only e drawn in a simple way. For e0ample with point features you can set the colour and mar0er ,circle, sGuare, star etc.- ut that is all. 3ou cannot tell the GIS to draw the features ased on one of its properties in the attri ute ta le. In order to do that, you need to use either a !raduated, continuous or uni<ue value sym ol. *hese are descri ed in detail in the sections that follow. A GIS application will normally allow you to set the sym ology of a layer using a dialo! %o3 such as the one shown in in Illustration :< elow. In this dialog o0 you can choose colours and sym ol styles. #epending on the geometry type of a layer, different options may e shown. For e0ample with point layers you can choose a mar0er st"le. 1ith line and polygon layers there is no mar.er style option, ut instead you can select a line st"le and colour such as dashed orange for gravel roads, solid orange for minor roads, and so on ,as shown in Illustration :> elow-. 1ith polygon layers you also have the option of setting a fill st"le and colour.

Illustration #': <hen using simple sym"ols, the !eature is drawn without using an attri"ute to control how it loo s. $his is the dialog !or point !eatures. Graduated S"m%ols:

Illustration #): $here are di!!erent options when de!ining simple sym"ols !or polyline and polygon !eatures.

Sometimes vector features represent things with a changing numerical value. "ontour lines are a good e0ample of this. 'ach contour usually has an attri ute value called EheightE that contains information a out what height that contour represents. In Illustration :: earlier in this topic we showed contours all drawn with the same colour. Adding colour to the contours can help us to interpret the meanings of contours. For e0ample we can draw low lying areas with one colour, mid2altitude areas with another and high2altitude areas with a third.

57

Illustration #*: $he height attri"ute o! contours can "e used to separate the contours into # classes. 4ontours "etween -+/m and 112/m will "e drawn in "rown, those "etween 112/m and 12(/m in green and those "etween 12(/m and 1'//m in purple.

Illustration #+: ?ur map a!ter setting graduated colours !or our contours. Setting colours ased on discrete groups of attri ute values is called Graduated Sym ology in QGIS. *he process is shown in Illustrations :? and :@ a ove. Graduated s"m%ols are most useful when "ou want to show clear differences %etween features with attri%ute values in different value ran!es. *he GIS Application will analyse the attri ute data ,e.g. height- and, ased on the num er of classes you reGuest, create groupings for you. *his process is illustrated in *a le : elow.

:6

Attri ute 9alue 8 5 : ; < > ? @ 7

"lass and "olour "lass 8 "lass 8 "lass 8 "lass 5 "lass 5 "lass 5 "lass : "lass : "lass :

$a"le #: Graduated colour "rea s up the attri"ute value ranges into the num"er o! classes you select. Each class is represented "y a di!!erent colour. #ontinuous #olour S"m%ols: In the previous section on Graduated "olour sym ols we saw that we can draw features in discrete groups or classes. Sometimes it is useful to draw features in a colour ran!e from one colour to another. *he GIS Application will use a numerical attri ute value from a feature ,e.g. contour heights or pollution levels in a stream- to decide which colour to use. *a le ; elow shows how the attri ute value is used to define a continuous range of colours. Attri ute 9alue 8 5 : ; < > ? @ 7 $a"le (: 4ontinuous colour sym"ology uses a start colour 1e.g. light orange shown here2 and an end colour 1e.g. dar "rown shown here2 and creates a series o! shades "etween those colours. )sing the same contours e0ample we used in the previous section, letEs see how a map with continuous colour sym ology is defined and loo.s. *he process starts y setting the layers properties to continuous colour using a dialog li.e the one shown in Illustration :7 elow. :8 "olour ,no classes or grouping-

Illustration #-: Setting up continuous colour sym"ology. $he contour height attri"ute is used to determine colour values. 4olours are de!ined !or the minimum and maximum values. $he GIS Application will then create a gradient o! colours !or drawing the !eatures "ased on their heights. After defining the minimum and ma0imum colours in the colour ran!e, the colour features are drawn in will depend on where the attri ute lies in the range etween minimum and ma0imum. For e0ample if you have contour features with values starting at 8666m and ending at 8;66m, the value range is 8666 to 8;66. If the colour set for the minimum value is set to orange and the colour for the ma0imum value is lac., contours with a value of close to 8;66m will e drawn close to lac.. On the other hand contours with a value near to 8666m will e drawn close to orange. See Illustration ;6 elow

Illustration (/: A contour map drawn using continuous colour sym"ology. :5

:ni<ue ,alue S"m%ols: Sometimes the attri utes of features are not numeric, ut instead strin!s are used. EStringE is a computer term meaning a group of letters, num ers and other writing sym ols. Strings attri utes are often used to classify things y name. 1e can tell the GIS Application to give each uniGue string or num er its own colour and sym ol. =oad features may have different classes ,e.g. EstreetE, Esecondary roadE, Emain roadE etc.-, each drawn in the map view of the GIS with different colours or sym ols. *his is illustrated in *a le < elow. Attri ute 9alue Arterial route (ain road Secondary road Street $a"le ': =ni%ue attri"ute values !or a !eature type 1e.g. roads2 can each have their own sym"ol. 1ithin the GIS Application we can open Fchoose to use )niGue 9alue sym ology for a layer. *he GIS will scan through all the different string values in the attri ute field and uild a list of uniGue strings or num ers. 'ach uniGue value can then e assigned a colour and style. *his is shown in Illustration ;8 elow. "olour class and sym ol

Illustration (1: Be!ining uni%ue value sym"ology !or roads "ased on the road type.

::

1hen the GIS draws the layer, it will loo. at the attri utes of each feature efore drawing it to the screen. Based on the value in the chosen field in the attri ute ta le, the road line will e drawn with suita le colour and line style ,and fill style if its a polygon feature-. *his is shown in Illustration ;5 elow.

Illustration (2: A roads vector layer sym"olised using a uni%ue value per road type. Thin!s to %e aware of: #eciding which attri utes and sym ology to use reGuires some planning. Before you start collecting any GeoSpatial data, you should ensure you .now what attri utes are needed and how it will e sym olised. It is very difficult to go ac. and re2collect data if you plan poorly the first time around. =emem er also that the goal of collecting attri ute data is to allow you to analyse and interpret spatial information. /ow you do this depends on the Guestions you are trying to answer. Sym ology is a visual language that allows people to see and understand your attri ute data ased on the colours and sym ols you use. Because of this you should put a lot of thought into how you sym olise your maps in order to ma.e them easy to understand. &hat have we learned( &etEs wrap up what we covered in this wor.sheet! 9ector features have attri%utes Attri utes descri%e the properties of the feature *he attri utes are stored in a ta%le =ows in the ta le are called records *here is one record per feature in the vector layer "olumns in the ta le are called fields :;

Fields represent properties of the feature e.g. height, roof colour etc. Fields can contain numerical, strin! ,any te0t- and date information *he attri ute data for a feature can e used to determine how it is s"m%olised Graduated colour sym ology groups the data into discrete classes #ontinuous colour sym ology assigns colours from a colour range to the features ased on their attri utes :ni<ue value sym ology associates each different value in the chosen attri ute column with a different sym ol ,colour and styleIf the attri ute of a vector layer is not used to determine its sym ology, it is drawn using a sin!le s"m%ol only

.ow "ou tr"/ /ere are some ideas for you to try with your learners!

)sing the ta le that you created in the last topic, add a new column for the sym ology type you would use for each feature type and have the learners identify which sym ology type they would use ,see *a le > elow for an e0ample-. *ry to identify which sym ology types you would use for the following types of vector features! points showing p/ level of soil samples ta.en around your school lines showing a road networ. in a city polygons for houses with an attri ute that shows whether it is made of ric., wood or EotherE material.

:<

=eal world feature

Geometry Sym ology *ype *ype Sin!le S"m%ol Sin!le S"m%ol /ave your learners count the num er of learners using each footpath in the hour efore school and then use !raduated s"m%ols to show the popularity of each footpath Sin!le s"m%ol :ni<ue value ased on the grade of the learners in the classroom /ave your learners rate the condition of the fence around your school y separating it into sections and grading each section on a scale of 827 ased on its condition. )se !raduated s"m%ols to classify the condition attri ute. "ount the num er of learners in each classroom and use a continuous colour s"m%ol to define a range of colours from red to lue. $olygon $olyline

*he school flagpole $oint *he soccer field *he footpaths in and around the school $laces where taps are located "lassrooms Fence

$oint $olygon $olyline

"lassrooms

$olygon

$a"le ): An example o! a ta"le that de!ines the !eature types and the ind o! sym"ology you would use !or each. Somethin! to thin0 a%out: If you donEt have a computer availa le, you can use transparency sheets and a 8!<6 666 map sheet to e0periment with different sym ology types. For e0ample place a transparency sheet over the map and using different coloured .o.i pens, draw in red all contour lines elow 766m ,or similar- and in green all lines a ove or eGual to 766m. "an you thin. of how to reproduce other sym ology types using the same techniGueJ 1urther readin!: &e%site: http!FFen.wi.ipedia.orgFwi.iF"artographyM(apNsym ology *he QGIS )ser Guide also has more detailed information on wor.ing with attri ute data and sym ology in QGIS. &hat s ne3t( In the section that follows we will ta.e a closer loo. at data capture. 1e will put the things we have learned a out vector data and attri utes into practice :>

y creating new data.

:?

GIS for 'ducators

Topic =: Data Capture O +ectives! Beywords! &earn how to create and edit vector and attri ute data. 'diting, data capture, heads2up, ta le, data ase.

Overview: In the previous two topics we loo.ed at vector data. 1e saw that there are two .ey concepts to vector data, namely! !eometr" and attri%utes. *he geometry of a vector feature descri es its shape and position, while the attri%utes of a vector feature descri e its properties ,colour, siHe, age etc.-. In this section we will loo. more closely at the process of creating and editing vector data D oth the geometry and attri utes of vector features. $ow does GIS di!ital data !et stored(: 1ord processors, spreadsheets and graphics pac.ages are all programs that let you create and edit digital data. 'ach type of application saves its data into a particular file format. For e0ample, a graphics program will let you save your drawing as a E.+pgE C$'G image, word processors let you save your document as an E.odtE Open#ocument or E.docE 1ord #ocument, and so on. Cust li.e these other applications, GIS Applications can store their data in files on the computer hard dis.. *here are a num er of different file formats for GIS data, ut the most common one is pro a ly the Eshape fileE. *he name is a little odd in that although we call it a shape file ,singular-, it actually consists of at least three different files that wor. together to store your digital vector data, as shown in *a le ? elow. '0tension .shp .d f .sh0 #escription *he geometry of vector features are stored in this file *he attri utes of vector features are stored in this file *his file is an inde0 that helps the GIS Application to find features more Guic.ly.

$a"le *: $he "asic !iles that together ma e up a >shape!ile>. 1hen you loo. at the files that ma.e up a shapefile on the computer hard dis., you will see something li.e Illustration ;: elow. If you want to share vector data stored in shapefiles with another person, it is important to give them all of the files for that layer. So in the case of the trees layer shown in Illustration ;: :@

elow, you would need to give the person trees.shp, trees.sh0, trees.d f, trees.pr+ and trees.Gml.

Illustration (#: $he !iles that ma e up a >trees> shape!ile as seen in the computer>s !ile manager. (any GIS Applications are also a le to store digital data inside a data%ase. In general storing GIS data in a data ase is a good solution ecause the data ase can store lar!e amounts of data efficientl" and can provide data to the GIS Application Guic.ly. )sing a data ase also allows many people to wor. with the same vector data layers at the same time. Setting up a data ase to store GIS data is more complicated than using shapefiles, so for this topic we will focus on creating and editing shapefiles. 7lannin! %efore "ou %e!in: Before you can create a new vector layer ,which will e stored in a shapefile-, you need .now what the geometry of that layer will e ,point, polyline or polygon-, and you need to .now what the attri utes of that layer will e. &etEs loo. at a few e0amples and it will ecome clearer how to go a out doing this. 93ample 1! "reating a tourism map Imagine that you want to create a nice tourism map for your local area. 3our vision of the final map is a 8!<6 666 toposheet with mar.ers overlaid for sites of interest to tourists. First, letEs thin. a out the geometry. 1e .now that we can represent a vector layer using point, polyline or polygon features. 1hich :7

one ma.es the most sense for our tourism mapJ 1e could use points if we wanted to mar. specific locations such as loo. out points, memorials, attle sites and so on. If we wanted to ta.e tourists along a route, such as a scenic route through a mountain pass, it might ma.e sense to use polylines. If we have whole areas that are of tourism interest, such as a nature reserve or a cultural village, polygons might ma.e a good choice. As you can see itEs often not easy to .now what type of geometry you will need. One common approach to this pro lem is to ma.e one layer for each geometry type you need. So, for e0ample, if you loo. at digital data provided y the "hief #irectorate ! Surveys and (apping, South Africa, they provide a river areas ,polygons- layer and a rivers polyline layer. *hey use the river areas ,polygons- to represent river stretches that are wide, and they use river polylines to represent narrow stretches of river. In Illustration ;; elow we can see how our tourism layers might loo. on a map if we used all three geometry types.

Illustration ((: A map with tourism layers. <e have used three di!!erent geometry types !or tourism data so that we can properly represent the di!!erent inds o! !eatures needed !or our visitors, giving them all the in!ormation they need. 93ample 4! "reating a map of pollution levels along a river ;6

If you wanted to measure pollution levels along the course of a river you would typically travel along the river in a oat or wal. along its an.s. At regular intervals you would stop and ta.e various measurements such as #issolved O0ygen ,#O- levels, "oliform Bacteria ,"B- counts, *ur idity levels and p/. 3ou would also need to ma.e a map reading of your position or o tain your position using a G$S receiver. *o store the data collected from an e0ercise li.e this in a GIS Application, you would pro a ly create a GIS layer with a point geometry. )sing point geometry ma.es sense here ecause each sample ta.en represents the conditions at a very specific place. For the attri utes we would want a field for each thing that descri es the sample site. So we may end up with an attri ute ta le that loo.s something li.e *a le @ elow. SampleAo p/ 8 5 : ? >.@ >.7 #O > < > "B A 3 3 *ur idity &ow (edium /igh "ollector $atience *ha o 9ictor #ate 85F68F5667 85F68F5667 85F68F5667

$a"le +: Brawing a ta"le li e this "e!ore you create your vector layer will let you decide what attri"ute !ields 1columns2 you will need. Cote that the geometry 1positions where samples were ta en2 is not shown in the attri"ute ta"le D the GIS Application stores it separatelyE #reatin! an empt" shapefile: Once you have planned what features you want to capture into the GIS, and the geometry type and attri utes that each feature should have, you can move on to the ne0t step of creating an empty shapefile. *he process usually starts with choosing the Enew vector layerE option in your GIS Application and then selecting a !eometr" t"pe ,see Illustration ;< elow-. As we covered in an earlier topic, this means choosing either point, polyline or polygon for the geometry.

;8 Illustration (': 4reating a new vector layer is as simple as !illing in a !ew details in a !orm. First you choose the geometry type, and then you add the attri"ute !ields.

Ae0t you will add fields to the attri ute ta le. Aormally we give field names that are short, have no spaces and indicate what type of information is eing stored in that field. '0ample field names may e Ep/E, E=oof"olourE, E=oad*ypeE and so on. As well as choosing a name for each field, you need to indicate how the information should e stored in that field D i.e. is it a num er, a word or a sentence, or a dateJ "omputer programs usually call information that is made up of words or sentences Estrin!sE, so if you need to store something li.e a street name or the name of a river, you should use string for the field type. *he shapefile format allows you to store the numeric field information as either a whole num er ,inte!er- or a decimal num er ,floatin! point- D so you need to thin. efore hand whether the numeric data you are going to capture will have decimal places or not. *he final step ,as shown in Illustration ;> elow- for creating a shapefile is to give it a name and a place on the computer hard dis. where it should e created. Once again it is a good idea to give the shapefile a short and meaningful name. Good e0amples are EriversE, EwatersamplesE and so on.

Illustration (): A!ter de!ining our new layer>s geometry and attri"utes, we need to save it to dis . It is important to give a short "ut meaning!ul name to your shape!ile.

;5

&etEs recap the process again Guic.ly. *o create a shapefile you first say what .ind of geometry it will hold, then you create one or more fields for the attri ute ta le, and then you save the shapefile to the hard dis. using an easy to recognise name. 'asy as 8252:4 Addin! data to "our shapefile So far we have only created an empty shapefile. Aow we need to ena%le editin! in the shapefile using the Eena le editingE menu option or tool ar icon in the GIS Application. Shapefiles are not ena led for editing y default to prevent accidentally changing or deleting the data they contain. Ae0t we need to start adding data. *here are two steps we need to complete for each record we add to the shapefile! 8."apturing geometry 5.'ntering attri utes *he process of capturing geometry is different for points, polylines and polygons. *o capture a point, you first use the map pan and Hoom tools to get to the correct geographical area that you are going to e recording data for. Ae0t you will need to ena le the point capture tool. /aving done that, the ne0t place you clic. with the left mouse %utton in the map view, is where you want your new point !eometr" to appear. After you clic. on the map, a window will appear and you can enter all of the attri%ute data for that point ,see Illustration ;? elow-. If you are unsure of the data for a given field you can usually leave it lan., ut e aware that if you leave a lot of fields lan. it will e hard to ma.e a useful map from your data4

;:

Illustration (*: A!ter you have captured the point geometry, you will "e as ed to descri"e its attri"utes. $he attri"ute !orm is "ased on the !ields you speci!ied when you created the vector layer. *o capture a pol"line the process is similar to that of a point, in that you need to first use the pan and Hoom tools to move the map in the map view to the correct geographical area. 3ou should e Hoomed in enough so that your

Illustration (+: 4apturing lines !or a tourism map. <hen editing a line layer, the vertices are shown with circular mar ers which you can move a"out with the mouse to ad:ust the line>s geometry. <hen adding a new line 1shown in red2, each clic o! the mouse will add a new vertex.

;;

new vector polyline feature will have an appropriate scale ,see *opic 5! 1or.ing with 9ector #ata for more details on scale issues-. 1hen you are ready, you can clic. the polyline capture icon in the tool ar and then start drawing your line y clic.ing on the map. After you ma.e your first clic., you will notice that the line stretches li.e an elastic and to follow the mouse cursor around as you move it. 'ach time you clic. with the left mouse %utton, a new verte0 will e added to the map. *his process is shown in Illustration ;@ a ove. 1hen you have finished defining your line, use the ri!ht mouse %utton to tell the GIS Application that you have completed your edits. As with the procedure for capturing a point feature, you will then e as.ed to enter in the attri ute data for your new polyline feature. *he process for capturin! a pol"!on is almost the same as capturing a polyline e0cept that you need to use the polygon capture tool in the tool ar. Also, you will notice that when you draw your geometry on the screen, the GIS Application always creates an enclosed area. *o add a new feature after you have created your first one, you can simply clic. again on the map with the point, polyline or polygon capture tool active and start to draw your ne0t feature. 1hen you have no more features to add, always e sure to clic. the Eallow editingE icon to toggle it off. *he GIS Application will then save your newly created layer to the hard dis.. $eads+up di!itisin! As you have pro a ly discovered y now if you followed the steps a ove, it is pretty hard to draw the features so that they are spatiall" correct if you do not have other features that you can use as a point of reference. One common solution to this pro lem is to use a raster layer ,such as an aerial photograph or a satellite image- as a ac.drop layer. 3ou can then use this layer as a reference map, or even trace the features off the raster layer into your vector layer if they are visi le. *his process is .nown as Eheads2up digitisingE and is shown in Illustration ;7 elow.

;<

Illustration (-: &eads,up digitising using a satellite image as a "ac drop. $he image is used as a re!erence !or capturing polyline !eatures "y tracing over them. Di!itisin! usin! a di!itisin! ta%le Another method of capturing vector data is to use a digitising ta le. *his approach is less commonly used e0cept y GIS professionals, and it reGuires e0pensive eGuipment. *he process of using a digitising ta le, is to place a paper map on the ta le. *he paper map is held securely in place using clips. *hen a special device called a Epuc.E is used to trace features from the map. *iny cross2hairs in the puc. are used to ensure that lines and points are drawn accurately. *he puc. is connected to a computer and each feature that is captured using the puc. gets stored in the computerEs memory. 3ou can see what a digitising puc. loo.s li.e in Illustration <6 elow.

;>

Illustration '/: A digitising ta"le and puc are used "y GIS pro!essionals when they want to digitise !eatures !rom existing maps. After "our features are di!itised...: Once your features are digitised, you can use the techniGues you learned in the previous *opic to set the sym ology for your layer. "hoosing an appropriate sym ology will allow you to etter understand the data you have captured when you loo. at the map. #ommon pro%lems ' thin!s to %e aware of: If you are digitising using a ac.drop raster layer such as an aerial photograph or satellite image, it is very important that the raster layer is properly georeferenced. A layer that is georeferenced properly displays in the correct position in the map view ased on the GIS ApplicationEs internal model of the earth. 1e can see the effect of a poorly georeferenced image in Illustration <8 elow. ;?

Illustration '1: $he importance o! using properly geore!erenced raster images !or heads,up digitising. ?n the le!t we can see the image is properly georegistered and the road !eatures 1in orange2 overlap per!ectly. I! the image is poorly geore!erenced 1as shown on the right2 the !eatures will not "e well aligned. <orse still, i! the image on the right is used as a re!erence when capturing new !eatures, the newly captured data will "e inaccurateE Also remem er that it is important that you are Hoomed in to an appropriate scale so that the vector features you create are useful. As we saw in the previous topic on vector geometry, it is a ad idea to digitise your data when you are Hoomed out to a scale of 8!8666 666 if you intend to use the data you capture at a scale of 8!<6 666 later. &hat have we learned( &etEs wrap up what we covered in this wor.sheet! Di!itisin! is the process of capturing .nowledge of a featureEs !eometr" and attri%utes into a di!ital format stored on the computerEs dis.. GIS #ata can e stored in a data%ase or as files. One commonly used file format is the shapefile which is actually a group of three or more files ,.shp, .d f and .sh0-. Before you create a new vector layer you need to plan oth what !eometr" type and attri%ute fields it will contain. Geometry can e point, polyline or polygon. Attri utes can e inte!ers ,whole num ers-, floatin! points ,decimal num ers-, strin!s ,words- or dates. *he digitising process consists of drawin! the geometry in the map view and then entering its attri utes. *his is repeated for each feature. $eads+up di!itisin! is often used to provide orientation during digitising y using a raster image in the ac.ground. $rofessional GIS users sometimes use a di!itisin! ta%le to capture information from paper maps. .ow "ou tr"/

;@

/ere are some ideas for you to try with your learners!

#raw up a list of features in and around your school that you thin. would e interesting to capture. For e0ample! the school oundary, the position of fire assem ly points, the layout of each class room, and so on. *ry to use a mi0 of different geometry types. Aow split your learners into groups and assign each group a few features to capture. /ave them sym olise their layers so that they are more meaningful to loo. at. "om ine the layers from all the groups to create a nice map of your school and its surroundings4 Find a local river and ta.e water samples along its length. (a.e a careful note of the position of each sample using a G$S or y mar.ing it on a toposheet. For each sample ta.e measurements such as p/, dissolved o0ygen etc. "apture the data using the GIS application and ma.e maps that show the samples with a suita le sym ology. "ould you identify any areas of concernJ 1as the GIS Application a le to help you to identify these areasJ

Somethin! to thin0 a%out: If you donEt have a computer availa le, you can follow the same process y using transparency sheets and a note oo.. )se an aerial photo, orthosheet or satellite image printout as your ac.ground layer. #raw columns down the page in your note oo. and write in the column headings for each attri ute field you want to store information a out. Aow trace the geometry of features onto the transparency sheet, writing a num er ne0t to each feature so that it can e identified. Aow write the same num er in the first column in your ta le in your note oo., and then fill in all the additional information you want to record. 1urther readin!: &e%site: http!FFwww..85science.orgFcurriculumFwaterpro+FS66pro+ectFmiami5666Fmiam iriverfinal.html D A school pro+ect to assess water Guality in their local river. *he QGIS )ser Guide also has more detailed information on digitising vector data in QGIS. &hat s ne3t( In the section that follows we will ta.e a closer loo. at raster data to learn all a out how image data can e used in a GIS.

;7

GIS for 'ducators

Topic 6: Raster Data O +ectives! Beywords! )nderstand what raster data is and how it can e used in a GIS. =aster, $i0el, =emote Sensing, Satellite, Image, Georeference

Overview: In the previous topics we have ta.en a closer loo. at vector data. 1hile vector features use geometry ,points, polylines and polygons- to represent the real world, raster data ta.es a different approach. =asters are made up of a matri0 of pi0els ,also called cells-, each containing a value that represents the conditions for the area covered y that cell ,see Illustration <5 elow-. In this topic we are going to ta.e a closer loo. at raster data, when it is useful and when it ma.es more sense to use vector data.

Illustration '2: A raster dataset is composed o! rows 1running across2 and columns 1running down2 o! pixels 1also now as cells2. Each pixel represents a geographical region, and the value in that pixel represents some characteristic o! that region. -aster data in detail: =aster data is used in a GIS application when we want to display information that is continuous across an area and cannot easily e divided into vector features. 1hen we introduced you to vector data we showed you the image in Illustration <: elow. $oint, polyline and polygon features wor. well for representing some features on this landscape, such as trees, roads and uilding footprints. Other features on a landscape can e more difficult to represent using vector features. For e0ample the grasslands shown have many <6

variations in colour and density of cover. It would e easy enough to ma.e a single polygon around each grassland area, ut a lot of the information a out the grassland would e lost in the process of simplifying the features to a single polygon. *his is ecause when you give a vector feature attri ute values, they apply to the whole feature, so vectors arenEt very good at representing features that are not homogeneous ,entirely the same- all over. Another approach you could ta.e is to digitise every small variation of grass colour and cover as a separate polygon. *he pro lem with that approach is that it will ta.e a huge amount of wor. in order to create a good vector dataset.

Illustration '#: Some !eatures on a landscape are easy to represent as points, polylines and polygons 1e.g. trees, roads, houses2. In other cases it can "e di!!icult. For example how would you represent the grasslands@ As polygons@ <hat a"out the variations in colour you can see in the grass@ <hen you are trying to represent large areas with continuously changing values, raster data can "e a "etter choice. )sing raster data is a solution to these pro lems. (any people use raster data as a %ac0drop to e used ehind vector layers in order to provide more meaning to the vector information. *he human eye is very good at interpreting images and so using an image ehind vector layers, results in maps with a lot more meaning. =aster data is not only good for images that depict the real world surface ,e.g. satellite images and aerial photographs-, they are also good for representing more a stract ideas. For e0ample, rasters can e used to show rainfall trends over an area, or to depict the fire ris. on a landscape. In these .inds of applications, each cell in the raster represents a different value. e.g. ris. of fire on a scale of one to ten. An e0ample that shows the difference etween an image o tained from a <8

satellite and one that shows calculated values can e seen in Illustration <; elow.

Illustration '(: $rue colour raster images 1le!t2 are use!ul as they provide a lot o! detail that is hard to capture as vector !eatures "ut easy to see when loo ing at the raster image. 3aster data can also "e non,photographic data such as the raster layer shown on the right which shows the calculated average minimum temperature in the <estern 4ape !or the month o! Aarch. Georeferencin!: Georeferencing is the process of defining e0actly where on the earthEs surface an image or raster dataset was created. *his positional information is stored with the digital version of the aerial photo. 1hen the GIS application opens the photo, it uses the positional information to ensure that the photo appears in the correct place on the map. Aormally this positional information consists of a coordinate for the top left pi0el in the image, the siHe of each pi0el in the I direction, the siHe of each pi0el in the 3 direction, and the amount ,if any- y which the image is rotated. 1ith these few pieces of information, the GIS application can ensure that raster data are displayed in the correct place. *he georeferencing information for a raster is often provided in a small te0t file accompanying the raster. Sources of raster data: =aster data can e o tained in a num er of ways. *wo of the most common ways are aerial photography and satellite imagery. In aerial photography, an aeroplane flies over an area with a camera mounted underneath it. *he photographs are then imported into a computer and georeferenced. Satellite imagery is created when satellites or iting the earth point special digital cameras towards the earth and then ta.e an image of the area on earth they are passing over. Once the image has een ta.en it is sent ac. to earth using radio signals to special receiving stations such as the one shown in Illustration << elow. *he process of capturing raster data from an aeroplane or satellite is called remote sensin!. <5

Illustration '': $he 4SI3 Satellite Applications 4enter at &arte"eeshoe near Gohannes"urg. Special antennae trac satellites as they pass overhead and download images using radio waves. In other cases, raster data can e computed. For e0ample an insurance company may ta.e police crime incident reports and create a country wide raster map showing how high the incidence of crime is li.ely to e in each area. (eteorologists ,people who study weather patterns- might generate a province level raster showing average temperature, rainfall and wind direction using data collected from weather stations ,see Illustration <; a ove-. In these cases, they will often use raster analysis techniGues such as interpolation ,which we descri e in *opic 86-. Sometimes raster data are created from vector data ecause the data owners want to share the data in an easy to use format. For e0ample, a company with road, rail, cadastral and other vector datasets may choose to generate a raster version of these datasets so that employees can view these datasets in a we rowser. *his is normally only useful if the attri utes, that users need to e aware of, can e represented on the map with la els or sym ology. If the user needs to loo. at the attri ute ta le for the data, providing it in raster format could e a ad choice ecause raster layers do not usually have any attri ute data associated with them. Spatial -esolution: 'very raster layer in a GIS has pi0els ,cells- of a fi0ed siHe that determine its spatial resolution. *his ecomes apparent when you loo. at an image at a small scale ,see Illustration <> elow- and then Hoom in to a large scale ,see Illustration <? elow-.

<:

Illustration '): $his satellite image loo s good when using a small scale...

Illustration '*: ..."ut when viewed at a large scale you can see the individual pixels that the image is composed o!. Several factors determine the spatial resolution of an image. For remote sensing data, spatial resolution is usually determined y the capa ilities of the sensor used to ta.e an image. For e0ample S$O*< satellites can ta.e images where each pi0el is 86m 0 86m. Other satellites, for e0ample (O#IS ta.e images only at <66m 0 <66m per pi0el. In aerial photography, pi0el siHes of <6cm 0 <6cm are not uncommon. Images with a pi0el siHe covering a small area are called Ehi!h resolutionE images ecause it is possi le to ma.e out a high degree of detail in the image. Images with a pi0el siHe covering a large area are called Elow resolutionE images ecause the amount of detail the images show is low. In raster data that is computed y spatial analysis ,such as the rainfall map we mentioned earlier-, the spatial density of information used to create the raster will usually determine the spatial resolution. For e0ample if you want to create a high resolution average rainfall map, you would ideally need many weather stations in close pro0imity to each other. One of the main things to e aware of with rasters captured at a high spatial resolution is storage reGuirements. *hin. of a raster that is :0: pi0els, each of which contains a num er representing average rainfall. *o store all the information contained in the raster, you will need to store 7 num ers in the computerEs memory. Aow imagine you want to have a raster layer for the <;

whole of South Africa with pi0els of 8.m 0 8.m. South Africa is around 8,587,676 .m5. 1hich means your computer would need to store over a million num ers on its hard dis. in order to hold all of the information. (a.ing the pi0el siHe smaller would greatly increase the amount of storage needed. Sometimes using a low spatial resolution is useful when you want to wor. with a large area and are not interested in loo.ing at any one area in a lot of detail. *he cloud maps you see on the weather report, are an e0ample of this D itEs useful to see the clouds across the whole country. Kooming in to one particular cloud in high resolution will not tell you very much a out the upcoming weather4 On the other hand, using low resolution raster data can e pro lematic if you are interested in a small region ecause you pro a ly wonEt e a le to ma.e out any individual features from the image. Spectral resolution: If you ta.e a colour photograph with a digital camera or camera on a cellphone, the camera uses electronic sensors to detect red, green and lue light. 1hen the picture is displayed on a screen or printed out, the red, green and lue ,=GB- information is com ined to show you an image that your eyes can interpret. 1hile the information is still in digital format though, this =GB information is stored in separate colour %ands. 1hilst our eyes can only see =GB wavelengths, the electronic sensors in cameras are a le to detect wavelengths that our eyes cannot. Of course in a hand held camera it pro a ly doesnEt ma.e sense to record information from the non+visi%le parts of the spectrum since most people +ust want to loo. at pictures of their dog or what have you. =aster images that include data for non2visi le parts of the light spectrum are often referred to as multi2spectral images. In GIS recording the non2visi le parts of the spectrum can e very useful. For e0ample, measuring infra2red light can e useful in identifying water odies. Because having images containing multiple ands of light is so useful in GIS, raster data are often provided as multi2 and images. 'ach and in the image is li.e a separate layer. *he GIS will com ine three of the ands and show them as red, green and lue so that the human eye can see them. *he num er of ands in a raster image is referred to as its spectral resolution. If an image consists of only one and, it is often called a !ra"scale image. 1ith grayscale images, you can apply false colouring to ma.e the differences in values in the pi0els more o vious. Images with false colouring applied are often referred to as pseudocolour ima!es.

<<

-aster to vector conversion: In our discussion of vector data, we e0plained that often raster data are used as a ac.drop layer, which is then used as a ase from which vector features can e digitised. Another approach is to use advanced computer programs to automatically e0tract vector features from images. Some features such as roads show in an image as a sudden change of colour from neigh ouring pi0els. *he computer program loo.s for such colour changes and creates vector features as a result. *his .ind of functionality is normally only availa le in very specialised ,and often e0pensive- GIS software. ,ector to raster conversion: Sometimes it is useful to convert vector data into raster data. One side effect of this is that attri ute data ,that is attri utes associated with the original vector data- will e lost when the conversion ta.es place. /aving vectors converted to raster format can e useful though when you want to give GIS data to non GIS users. 1ith the simpler raster formats, the person you give the raster image to can simply view it as an image on their computer without needing any special GIS software. -aster anal"sis: *here are a great many analytical tools that can e run on raster data which cannot e used with vector data. For e0ample, rasters can e used to model water flow over the land surface. *his information can e used to calculate where watersheds and stream networ.s e0ist, ased on the terrain. =aster data are also often used in agriculture and forestry to manage crop production. For e0ample with a satellite image of a farmerEs lands, you can identify areas where the plants are growing poorly and then use that information to apply more fertiliHer on the affected areas only. Foresters use raster data to estimate how much tim er can e harvested from an area. =aster data is also very important for disaster management. Analysis of #igital 'levation (odels ,a .ind of raster where each pi0el contains the height a ove sea level- can then e used to identify areas that are li.ely to e flooded. *his can then e used to target rescue and relief efforts to areas where it is needed the most. #ommon pro%lems ' thin!s to %e aware of: As we have already mentioned, high resolution raster data can reGuire large amounts of computer storage.

<>

&hat have we learned( &etEs wrap up what we covered in this wor.sheet! =aster data are a grid of regularly siHed pi3els. =aster data are good for showing continuall" var"in! information. *he siHe of pi0els in a raster determines its spatial resolution. =aster images can contain one or more %ands, each covering the same spatial area, ut containing different information. 1hen raster data contains ands from different parts of the electromagnetic spectrum, they are called multi+spectral ima!es. *hree of the ands of a multi2spectral image can e shown in the colours =ed, Green and Blue so that we can see them. Images with a single and are called grayscale images. Single and, grayscale images can e shown in pseudocolour y the GIS. =aster images can consume a large amount of storage space. .ow "ou tr"/ /ere are some ideas for you to try with your learners!

#iscuss with your learners in which situations you would use raster data and in which you would use vector data. Get your learners to create a raster map of your school y using A; transparency sheets with grid lines drawn on them. Overlay the transparencies onto a toposheet or aerial photograph of your school. Aow let each learner or group of learners colour in cells that represent a certain type of feature. e.g. uilding, playground, sports field, trees, footpaths etc. 1hen they are all finished, overlay all the sheets together and see if it ma.es a good raster map representation of your school. 1hich types of features wor.ed well when represented as rastersJ /ow did your choice in cell siHe affect your a ility to represent different feature typesJ

Somethin! to thin0 a%out: If you donEt have a computer availa le, you can understand raster data using pen and paper. #raw a grid of sGuares onto a sheet of paper to represent your soccer field. Fill the grid in with num ers representing values for grass cover on your soccer field. If a patch is are give the cell a value of 6. If the patch is mi0ed are and covered, give it a value of 8. If an area is completely covered with grass, give it a value of 5. Aow use pencil crayons to colour the cells ased on their values. "olour cells with value 5 dar. green. 9alue 8 should get coloured light green, and value 6 coloured in rown. 1hen you finish, you should have a raster map of your soccer field4

<?

1urther readin!: 2oo0:

"hang, Bang2*sung ,566>-! Introduction to Geographic Information Systems. :rd 'dition. (cGraw /ill. ,ISBA 66?6><@7@>#e(ers, (ichael A. ,566<-! Fundamentals of Geographic Information Systems. :rd 'dition. 1iley. ,ISBA 7@8;85>87<-

&e%site: http!FFen.wi.ipedia.orgFwi.iFGISM=aster *he QGIS )ser Guide also has more detailed information on wor.ing with raster data in QGIS. &hat s ne3t( In the section that follows we will ta.e a closer loo. at topolo!" to see how the relationship etween vector features can e used to ensure the est data Guality.

<@

GIS for 'ducators

Topic >: Topology O +ectives! Beywords! )nderstanding topology in vector data 9ector, topology, topology rules, topology errors, search radius, snapping distance, simple feature

Overview: Topolo!" e0presses the spatial relationships etween connecting or ad+acent vector features ,points, polylines and polygons- in a GIS. *opological or topology2 ased data are useful for detecting and correcting digitising errors ,e.g. two lines in a roads vector layer that do not meet perfectly at an intersection-. *opology is necessary for carrying out some types of spatial analysis, such as networ. analysis. Imagine you travel to &ondon. On a sightseeing tour you plan to visit St. $aulEs "athedral first and in the afternoon "ovent Garden (ar.et for some souvenirs. &oo.ing at the )nderground map of &ondon ,see Illustration <@ elow- you have to find connecting trains to get from "ovent Garden to St. $aulEs. *his reGuires topological information ,data- a out where it is possi le to change trains. &oo.ing at a map of the underground, the topological relationships are illustrated y circles that show connectivity. "hanging trains at stations allows you to move from one connected part of the networ. to another.

Illustration '+: $opology o! 6ondon =nderground Cetwor .

<7

Topolo!" errors *here are different types of topological errors and they can e grouped according to whether the vector feature types are polygons or polylines. *opological errors with pol"!on features can include unclosed polygons, gaps etween polygon orders or overlapping polygon orders. A common topological error with pol"line features is that they do not meet perfectly at a point ,node-. *his type of error is called an undershoot if a gap e0ists etween the lines, and an overshoot if a line ends eyond the line it should connect to ,see Illustration <7 elow-.

Illustration '-: =ndershoots 112 occur when digitised vector lines that should connect to each other don>t %uite touch. ?vershoots 122 happen i! a line ends "eyond the line it should connect to. Slivers 1#2 occur when the vertices o! two polygons do not match up on their "orders. *he result of overshoot and undershoot errors are so2called Edangling nodesE at the end of the lines. #angling nodes are accepta le in special cases, for e0ample if they are attached to dead2end streets. *opological errors rea. the relationship etween features. *hese errors need to e fi0ed in order to e a le to analyse vector data with procedures li.e networ. analysis ,e.g. finding the est route across a road networ.- or measurement ,e.g. finding out the length of a river-. In addition to topology eing useful for networ. analysis and measurement, there are other reasons why it is important and useful to create or have vector data with correct topology. Cust imagine you digitise a municipal oundaries map for your province and the polygons overlap or show slivers. If such errors were present, you would e a le to use the measurement tools, ut the results you get will e incorrect. 3ou will not .now the correct area for any municipality and you will not e a le to define e0actly, where the orders etween the municipalities are. It is not only important for your own analysis to create and have topologically correct data, ut also for people who you pass data on to. *hey will e e0pecting your data and analysis results to e correct4 >6

Topolo!" rules Fortunately, many common errors that can occur when digitising vector features can e prevented y topology rules that are implemented in many GIS applications. '0cept for some special GIS data formats, topology is usually not enforced y default. (any common GIS, li.e QGIS, define topology as relationship rules and let the user choose the rules, if any, to e implemented in a vector layer. *he following list shows some e0amples of where topology rules can e defined for real world features in a vector map.

Area edges of a municipality map must not overlap. Area edges of a municipality map must not have gaps ,slivers-. $olygons showing property oundaries must e closed. )ndershoots or overshoots of the order lines are not allowed. "ontour lines in a vector line layer must not intersect ,cross each other-.

Topolo!ical tools (any GIS applications provide tools for topological editing. For e0ample in QGIS you can ena%le topolo!ical editin! to improve editing and maintaining common oundaries in polygon layers. A GIS such as QGIS EdetectsE a shared oundary in a polygon map so you only have to move the edge verte0 of one polygon oundary and QGIS will ensure the updating of the other polygon oundaries as shown in Illustration >6,8- elow. Another topological option allows you to prevent pol"!on overlaps during digitising ,see Illustration >6,5- elow-. If you already have one polygon, it is possi le with this option to digitise a second ad+acent polygon so that oth polygons overlap and QGIS then clips the second polygon to the common oundary.

Illustration )/: 112$opological editing to detect shared "oundaries, when moving vertices. <hen moving a vertex, all !eatures that share that vertex are updated. 122 $o avoid polygon overlaps, when a new polygon is digitised 1shown in red2 it is clipped to avoid overlapping neigh"ouring areas. >8

Snappin! distance Snapping distance is the distance a GIS uses to search for the closest verte0 and F or segment you are trying to connect when you digitise. A se!ment is a straight line formed etween two vertices in a polygon or polyline geometry. If you arenEt within the snapping distance, a GIS such as QGIS will leave the verte0 where you release the mouse utton, instead of snapping it to an e0isting verte0 and F or segment ,see Illustration >8 elow-.

Illustration )1: $he snapping distance 1"lac circle2 is de!ined in map units 1e.g. decimal degrees2 !or snapping to either vertices or segments. Search -adius Search radius is the distance a GIS uses to search for the closest verte0 you are trying to move when you clic. on the map. If you arenEt within the search radius, the GIS wonEt find and select any verte0 of a feature for editing. In principle, it is Guite similar to the snapping distance functionality. Snapping distance and search radius are oth set in map units so you may need to e0periment to get the distance value set right. If you specify a value that is too ig, the GIS may snap to a wrong verte0, especially if you are dealing with a large num er of vertices close together. If you specify the search radius too small the GIS application wonEt find any feature or verte0 to move or edit. #ommon pro%lems ' thin!s to %e aware of *opology is a comple0 representation of vector data. *rue topological vector datasets are stored in special file formats that record all the relationships etween features. (ost commonly used vector data formats use something called ESimple FeaturesE which also consists of point, line and polygon features. Simple feature datasets are mainly designed for simplicity and for fast rendering ut not for data analysis that reGuire topology ,such as finding routes across a networ.-. (any GIS applications are a le to show topological and simple feature data together and some can also create, edit and analyse oth.

>5

&hat have we learned( &etEs wrap up what we covered in this wor.sheet!

Topolo!" shows the spatial relation of neigh ouring vector features. *opology in GIS is provided y topolo!ical tools. *opology can e used to detect and correct di!iti5in! errors. For some tools, such as networ0 anal"sis, topological data is essential. Snappin! distance and search radius help us to digitise topologically correct vector data. Simple feature data is not a true topological data format ut it is commonly used y GIS applications.

.ow "ou tr"/ /ere are some ideas for you to try with your learners!

(ar. your local us stops on a toposheet map and then tas. your learners to find the shortest route etween two stops. *hin. of how you would create vector features in a GIS to represent a topological road networ. of your town. 1hat topological rules are important and what tools can your learners use in QGIS to ma.e sure that the new road layer is topologically correctJ

Somethin! to thin0 a%out: If you donEt have a computer availa le, you can use a map of a us or railway networ. and discuss the spatial relationships and topology with your learners. 1urther readin!: 2oo0s:

"hang, Bang2*sung ,566>-! Introduction to Geographic Information Systems. :rd 'dition. (cGraw /ill. ,ISBA 66?6><@7@>#e(ers, (ichael A. ,566<-! Fundamentals of Geographic Information Systems. :rd 'dition. 1iley. ,ISBA 7@8;85>87<-

&e%sites:

http!FFwww.innovativegis.comF asisFprimerFconcepts.html http!FFen.wi.ipedia.orgFwi.iFGeospatialNtopology

*he QGIS )ser Guide also has more detailed information on topological editing provided in QGIS. &hat s ne3t( >:

In the section that follows we will ta.e a closer loo. at #oordinate -eference S"stems to understand how we relate data from our spherical earth onto flat maps4

>;

GIS for 'ducators

Topic ?: Coordinate Reference Systems O +ectives! Beywords! )nderstanding of "oordinate =eference Systems. "oordinate =eference System ,"=S-, (ap $ro+ection, On the Fly $ro+ection, &atitude, &ongitude, Aorthing, 'asting

Overview: Map pro@ections try to portray the surface of the earth or a portion of the earth on a flat piece of paper or computer screen. A coordinate reference s"stem ,"=S- then defines, with the help of coordinates, how the two2 dimensional, pro+ected map in your GIS is related to real places on the earth. *he decision as to which map pro+ection and coordinate reference system to use, depends on the regional e0tent of the area you want to wor. in, on the analysis you want to do and often on the availa ility of data. Map 7ro@ection in detail A traditional method of representing the earthEs shape is the use of glo es. *here is, however, a pro lem with this approach. Although glo es preserve the ma+ority of the earthEs shape and illustrate the spatial configuration of continent2siHed features, they are very difficult to carry in oneEs poc.et. *hey are also only convenient to use at e0tremely small scales ,e.g. 8 ! 866 million-. (ost of the thematic map data commonly used in GIS applications are of considera ly larger scale. *ypical GIS datasets have scales of 8!5<6 666 or greater, depending on the level of detail. A glo e of this siHe would e difficult and e0pensive to produce and even more difficult to carry around. As a result, cartographers have developed a set of techniGues called map pro@ections designed to show, with reasona le accuracy, the spherical earth in two2 dimensions. 1hen viewed at close range the earth appears to e relatively flat. /owever when viewed from space, we can see that the earth is relatively spherical. (aps, as we will see in the upcoming map production topic, are representations of reality. *hey are designed to not only represent features, ut also their shape and spatial arrangement. 'ach map pro+ection has advanta!es and disadvanta!es. *he est pro+ection for a map depends on the scale of the map, and on the purposes for which it will e used. For e0ample, a pro+ection may have unaccepta le distortions if used to map the entire African continent, ut may e an e0cellent choice for a lar!e+scale )detailed* map of your country. *he properties of a map pro+ection may also influence some of the design features of the map. Some pro+ections are good for small areas, some are good for mapping areas with a large 'ast21est ><

e0tent, and some are etter for mapping areas with a large Aorth2South e0tent. The three families of map pro@ections *he process of creating map pro+ections can e visualised y positioning a light source inside a transparent glo e on which opaGue earth features are placed. *hen pro+ect the feature outlines onto a two2dimensional flat piece of paper. #ifferent ways of pro+ecting can e produced y surrounding the glo e in a c"lindrical fashion, as a cone, or even as a flat surface. 'ach of these methods produces what is called a map pro@ection famil". *herefore, there is a family of planar pro@ections, a family of c"lindrical pro@ections, and another called conical pro@ections ,see Illustration >5-

Illustration )2: $he three !amilies o! map pro:ections. $hey can "e represented "y a2 cylindrical pro:ections, "2 conical pro:ections or c2 planar pro:ections. *oday, of course, the process of pro+ecting the spherical earth onto a flat piece of paper is done using the mathematical principles of geometry and trigonometry. *his recreates the physical pro+ection of light through the glo e. Accurac" of map pro@ections (ap pro+ections are never a solutely accurate representations of the spherical earth. As a result of the map pro+ection process, every map shows distortions of an!ular conformit", distance and area. A map pro+ection may com ine >>

several of these characteristics, or may e a compromise that distorts all the properties of area, distance and angular conformity, within some accepta le limit. '0amples of compromise pro+ections are the &in0el Tripel pro@ection and the -o%inson pro@ection ,see Illustration >: elow-, which are often used for world maps.

Illustration )#: $he 3o"inson pro:ection is a compromise where distortions o! area, angular con!ormity and distance are accepta"le. It is usually impossi le to preserve all characteristics at the same time in a map pro+ection. *his means that when you want to carry out accurate analytical operations, you need to use a map pro+ection that provides the est characteristics for your analyses. For e0ample, if you need to measure distances on your map, you should try to use a map pro+ection for your data that provides high accuracy for distances. Map pro@ections with an!ular conformit" 1hen wor.ing with a glo e, the main directions of the compass rose ,Aorth, 'ast, South and 1est- will always occur at 76 degrees to one another. In other words, 'ast will always occur at a 76 degree angle to Aorth. (aintaining correct an!ular properties can e preserved on a map pro+ection as well. A map pro+ection that retains this property of angular conformity is called a conformal or orthomorphic pro@ection.

>?

Illustration )(: $he Aercator pro:ection, !or example, is used where angular relationships are important, "ut the relationship o! areas are distorted. *hese pro+ections are used when the preservation of an!ular relationships is important. *hey are commonly used for navigational or meteorological tas.s. It is important to remem er that maintaining true angles on a map is difficult for large areas and should e attempted only for small portions of the earth. *he conformal type of pro+ection results in distortions of areas, meaning that if area measurements are made on the map, they will e incorrect. *he larger the area the less accurate the area measurements will e. '0amples are the Mercator pro@ection ,as shown in Illustration >; a ove- and the 8am%ert #onformal #onic pro@ection. *he ).S. Geological Survey uses a conformal pro+ection for many of its topographic maps. Map pro@ections with e<ual distance If your goal in pro+ecting a map is to accurately measure distances, you should select a pro+ection that is designed to preserve distances well. Such pro+ections, called e<uidistant pro@ections, reGuire that the scale of the map is 0ept constant. A map is eGuidistant when it correctly represents distances

>@ Illustration )': $he =nited Cations 6ogo uses the A5imuthal E%uidistant pro:ection.

from the centre of the pro+ection to any other place on the map. 9<uidistant pro@ections maintain accurate distances from the centre of the pro+ection or along given lines. *hese pro+ections are used for radio and seismic mapping, and for navigation. *he 7late #arree 9<uidistant #"lindrical ,see Illustration >> elow- and the 9<uirectan!ular pro@ection are two good e0amples of eGuidistant pro+ections. *he A5imuthal 9<uidistant pro@ection is the pro+ection used for the em lem of the )nited Aations ,see Illustration >< elow-.

Illustration )): $he ;late 4arree E%uidistant 4ylindrical pro:ection, !or example, is used when accurate distance measurement is important. 7ro@ections with e<ual areas 1hen a map portrays areas over the entire map, so that all mapped areas have the same proportional relationship to the areas on the 'arth that they represent, the map is an e<ual area map. In practice, general reference and educational maps most often reGuire the use of e<ual area pro@ections. As the name implies, these maps are est used when calculations of area are the dominant calculations you will perform. If, for e0ample, you are trying to analyse a particular area in your town to find out whether it is large enough for a new shopping mall, eGual area pro+ections are the est choice. On the one hand, the larger the area you are analysing, the more precise your area measures will e, if you use an eGual area pro+ection rather than another type. On the other hand, an eGual area pro+ection results in distortions of an!ular conformit" when dealing with large areas. Small areas will e far less prone to having their angles distorted when you use an eGual area pro+ection. Al%er s e<ual area, 8am%ert s e<ual area and Mollweide 9<ual Area #"lindrical pro@ections ,shown in Illustration >? elow- are types of eGual area pro+ections that are often encountered in GIS wor..

>7

Beep in mind that map pro+ection is a very comple0 topic. *here are hundreds of different pro+ections availa le world wide each trying to portray a certain portion of the earthEs surface as faithfully as possi le on a flat piece of paper. In reality, the choice of which pro+ection to use, will often e made for you. (ost countries have commonly used pro+ections and when data is e0changed people will follow the national trend.

Illustration )*: $he Aollweide E%ual Area 4ylindrical pro:ection, !or example, ensures that all mapped areas have the same proportional relationship to the areas on the Earth. #oordinate -eference S"stem )#-S* in detail 1ith the help of coordinate reference systems ,"=S- every place on the earth can e specified y a set of three num ers, called coordinates. In general "=S can e divided into pro@ected coordinate reference s"stems ,also called "artesian or rectangular coordinate reference systems- and !eo!raphic coordinate reference s"stems. Geo!raphic #oordinate S"stems *he use of Geographic "oordinate =eference Systems is very common. *hey use degrees of latitude and longitude and sometimes also a height value to descri e a location on the earthOs surface. *he most popular is called &GS A=. 8ines of latitude run parallel to the eGuator and divide the earth into 8@6 eGually spaced sections from Aorth to South ,or South to Aorth-. *he reference line for latitude is the eGuator and each hemisphere is divided into ninety sections, each representing one degree of latitude. In the northern hemisphere, degrees of latitude are measured from Hero at the eGuator to ninety at the north pole. In the southern hemisphere, degrees of latitude are measured from Hero at the eGuator to ninety degrees at the south pole. *o simplify the digitisation of maps, degrees of latitude in the southern hemisphere are often assigned negative values ,6 to 276P-. 1herever you are on the earthOs surface, the distance etween the lines of latitude is the same ?6

,>6 nautical miles-. See Illustration >@ elow for a pictorial view.

Illustration )+: Geographic coordinate system with lines o! latitude parallel to the e%uator and lines o! longitude with the prime meridian through Greenwich. 8ines of lon!itude, on the other hand, do not stand up so well to the standard of uniformity. &ines of longitude run perpendicular to the eGuator and converge at the poles. *he reference line for longitude ,the prime meridianruns from the Aorth pole to the South pole through Greenwich, 'ngland. Su seGuent lines of longitude are measured from Hero to 8@6 degrees 'ast or 1est of the prime meridian. Aote that values 1est of the prime meridian are assigned negative values for use in digital mapping applications. See Illustration >@ a ove for a pictorial view. At the eGuator, and only at the eGuator, the distance represented y one line of longitude is eGual to the distance represented y one degree of latitude. As you move towards the poles, the distance etween lines of longitude ecomes progressively less, until, at the e0act location of the pole, all :>6P of longitude are represented y a single point that you could put your finger on ,you pro a ly would want to wear gloves though-. )sing the geographic coordinate system, we have a grid of lines dividing the earth into sGuares that cover appro0imately 85:>:.:>< sGuare .ilometres at the eGuatorQa good start, ut not very useful for determining the location of anything within that sGuare. *o e truly useful, a map grid must e divided into small enough sections so that they can e used to descri e ,with an accepta le level of accuracy- the location of a point on the map. *o accomplish this, degrees are divided into minutes ) * and seconds )B*. *here are si0ty minutes in a degree, and si0ty seconds in a minute ,:>66 seconds in a degree-. So, at the eGuator, one second of latitude or longitude R :6.@?>5; meters.

?8

?5

7ro@ected coordinate reference s"stems A two2dimensional coordinate reference system is commonly defined y two a0es. At right angles to each other, they form a so called CD2plane ,see Illustration >7 on the left side-. *he horiHontal a0is is normally la elled C, and the vertical a0is is normally la elled D. In a three2dimensional coordinate reference system, another a0is, normally la elled E, is added. It is also at right angles to the C and D a0es. *he E a0is provides the third dimension of space ,see Illustration >7 on the right side-. 'very point that is e0pressed in spherical coordinates can e e0pressed as an C D E coordinate.

Illustration )-: ;ro:ected coordinate re!erence systems. $wo,dimensional with 7 and 8 coordinates 1le!t2 and three,dimensional with 7, 8 and 9 coordinates 1right2. A pro+ected coordinate reference system in the southern hemisphere ,south of the eGuator- normally has its origin on the eGuator at a specific 8on!itude. *his means that the 32values increase southwards and the I2values increase to the 1est. In the northern hemisphere ,north of the eGuator- the origin is also the eGuator at a specific 8on!itude. /owever, now the 32values increase northwards and the I2values increase to the 'ast. In the following section, we descri e a pro+ected coordinate reference system, called :niversal Transverse Mercator ):TM* often used for South Africa. :niversal Transverse Mercator ):TM* #-S in detail: *he )niversal *ransverse (ercator ,)*(- coordinate reference system has its origin on the e<uator at a specific 8on!itude. Aow the 3+values increase Southwards and the C2values increase to the 1est. *he )*( "=S is a glo al map pro+ection. *his means, it is generally used all over the world. But as already descri ed in the section Saccuracy of map pro+ectionsT a ove, the larger the area ,for e0ample South Africa- the more distortion of angular conformity, distance and area occur. *o avoid too much distortion, the world is divided into >F e<ual 5ones that are all > de!rees wide in longitude from ?:

'ast to 1est. *he :TM 5ones are num ered 1 to >F, starting at the international date line ,5one 1 at 8@6 degrees 1est longitude- and progressing 'ast ac. to the international date line ,5one >F at 8@6 degrees 'ast longitude- as shown in Illustration ?6 elow.

Illustration */: $he =niversal $ransverse Aercator 5ones. For South A!rica =$A 5ones ##S, #(S, #'S, and #)S are used. As you can see in Illustration ?6 a ove and Illustration ?8 elow, South Africa is covered y four :TM 5ones to minimiHe distortion. *he 5ones are called :TM ;;S, :TM ;=S, :TM ;6S and :TM ;>S. *he S after the Hone means that the )*( Hones are located south of the e<uator.

?;

Illustration *1: =$A 5ones ##S, #(S, #'S, and #)S with their central longitudes 1meridians2 used to pro:ect South A!rica with high accuracy. $he red cross shows an Area o! Interest 1A?I2. Say, for e0ample, that we want to define a two2dimensional coordinate within the Area of Interest )AOI* mar.ed with a red cross in Illustration ?8 a ove 3ou can see, that the area is located within the :TM 5one ;6S. *his means, to minimiHe distortion and to get accurate analysis results, we should use :TM 5one ;6S as the coordinate reference system. *he position of a coordinate in )*( south of the eGuator must e indicated with the 5one num%er ,:<- and with its northin! )"* value and eastin! )3* value in meters. *he northin! value is the distance of the position from the e<uator in meters. *he eastin! value is the distance from the central meridian ,longitude- of the used )*( Hone. For )*( Hone :<S it is 4? de!rees 9ast as shown in Illustration ?8 a ove. Furthermore, ecause we are south of the eGuator and negative values are not allowed in the )*( coordinate reference system, we have to add a so called false northin! value of 86,666,666m to the northing ,y- value and a false easting value of <66,666m to the easting ,0- value. *his sounds difficult, so, we will do an e0ample that shows you how to find the correct :TM ;6S coordinate for the Area of Interest. The northin! )"* value *he place we are loo.ing for is :,<<6,666 meters south of the eGuator, so the northing ,y- value gets a ne!ative si!n and is 2:,<<6,666m. According to the )*( definitions we have to add a false northin! value of 86,666,666m. *his means the northing ,y- value of our coordinate is >,;<6,666m ,2:,<<6,666m U 86,666,666m-. The eastin! )3* value

?<

First we have to find the central meridian ,longitude- for the :TM 5one ;6S. As we can see in VVV?8VVV it is 4? de!rees 9ast. *he place we are loo.ing for is A6,FFF meters &est from the central meridian. Cust li.e the northing value, the easting ,0- value gets a negative sign, giving a result of +A6,FFFm. According to the )*( definitions we have to add a false eastin! value of <66,666m. *his means the easting ,0- value of our coordinate is ;8<,666m ,2 @<,666m U <66,666m-. Finally, we have to add the 5one num%er to the easting value to get the correct value. As a result, the coordinate for our 7oint of Interest, pro+ected in :TM 5one ;6S would e written as! ;6 =16,FFFm9 ' >,=6F,FFFm.. In some GIS, when the correct )*( Hone :<S is defined and the units are set to meters within the system, the coordinate could also simply appear as =16,FFF >,=6F,FFF. On+The+1l" 7ro@ection As you can pro a ly imagine, there might e a situation where the data you want to use in a GIS are pro+ected in different coordinate reference systems. For e0ample, you might get a vector layer showing the oundaries of South Africa pro+ected in )*( :<S and another vector layer with point information a out rainfall provided in the geographic coordinate system 1GS @;. In GIS these two vector layers are placed in totally different areas of the map window, ecause they have different pro+ections. *o solve this pro lem, many GIS include a functionality called On+the+fl" pro+ection. It means, that you can define a certain pro+ection when you start the GIS and all layers that you then load, no matter what coordinate reference system they have, will e automatically displayed in the pro+ection you defined. *his functionality allows you to overlay layers within the map window of your GIS, even though they may e in different reference systems. #ommon pro%lems ' thin!s to %e aware of: *he topic map pro@ection is very comple0 and even professionals who have studied geography, geodetics or any other GIS related science, often have pro lems with the correct definition of map pro+ections and coordinate reference systems. )sually when you wor. with GIS, you already have pro+ected data to start with. In most cases these data will e pro+ected in a certain "=S, so you donEt have to create a new "=S or even re pro+ect the data from one "=S to another. *hat said, it is always useful to have an idea a out what map pro+ection and "=S means. &hat have we learned( &etEs wrap up what we covered in this wor.sheet!

?>

Map pro@ections portray the surface of the earth on a two2dimensional, flat piece of paper or computer screen. *here are glo al map pro+ections, ut most map pro+ections are created and optimi5ed to pro@ect smaller areas of the earthEs surface. (ap pro+ections are never a solutely accurate representations of the spherical earth. *hey show distortions of an!ular conformit", distance and area. It is impossi le to preserve all these characteristics at the same time in a map pro+ection. A #oordinate reference s"stem ,"=S- defines, with the help of coordinates, how the two2dimensional, pro+ected map is related to real locations on the earth. *here are two different types of coordinate reference systems! Geo!raphic #oordinate S"stems and 7ro@ected #oordinate S"stems. On the 1l" pro@ection is a functionality in GIS that allows us to overlay layers, even if they are pro+ected in different coordinate reference systems.

.ow "ou tr"/ /ere are some ideas for you to try with your learners!

Start QGIS and load two layers of the same area ut with different pro+ections and let your pupils find the coordinates of several places on the two layers. 3ou can show them that it is not possi le to overlay the two layers. *hen define the coordinate reference system as GeographicF 1GS @; inside the $ro+ect $roperties #ialog and activate the chec. o0 Eena le On2the2fly "=S transformationE. &oad the two layers of the same area again and let your pupils see how On2the2fly pro+ection wor.s. 3ou can open the $ro+ect $roperties #ialog in QGIS and show your pupils the many different "oordinate =eference Systems so they get an idea of the comple0ity of this topic. 1ith EOn2the2fly "=S transformationE ena led you can select different "=S to display the same layer in different pro+ections.

Somethin! to thin0 a%out: If you donEt have a computer availa le, you can show your pupils the principles of the three map pro+ection families. Get a glo e and paper and demonstrate how cylindrical, conical and planar pro+ections wor. in general. 1ith the help of a transparency sheet you can draw a two2dimensional coordinate reference system showing I a0es and 3 a0es. *hen, let your pupils define coordinates ,0 and y values- for different places. 1urther readin!: 2oo0s: ??

"hang, Bang2*sung ,566>-! Introduction to Geographic Information Systems. :rd 'dition. (cGraw /ill. ,ISBA 66?6><@7@>#e(ers, (ichael A. ,566<-! Fundamentals of Geographic Information Systems. :rd 'dition. 1iley. ,ISBA 7@8;85>87<Galati, Stephen =. ,566>-! Geographic Information Systems #emystified. Artech /ouse Inc. ,ISBA 8<@6<:<::I-

&e%sites:

http!FFwww.colorado.eduFgeographyFgcraftFnotesFmappro+Fmappro+Nf.ht ml http!FFgeology.isu.eduFgeostacFFieldN'0erciseFtopomapsFinde0.htm

*he QGIS )ser Guide also has more detailed information on wor.ing with map pro+ections in QGIS. &hat s ne3t( In the section that follows we will ta.e a closer loo. at Map 7roduction.

?@

GIS for 'ducators

Topic A: Map Production O +ectives! Beywords! )nderstanding of map production for spatial data (ap production, map layout, scale ar, north arrow, legend, map ody, map unit

Overview: Map production is the process of arranging map elements on a sheet of paper in a way that, even without many words, the average person can understand what it is all a out. (aps are usually produced for presentations and reports where the audience or reader is a politician, citiHen or a learner with no professional ac.ground in GIS. Because of this, a map has to e effective in communicating spatial information. "ommon elements of a map are the title, map %od", le!end, north arrow, scale %ar, ac0nowled!ement, and map %order ,see Illustration ?5 elow-.

Illustration *2: 4ommon map elements 1la"elled in red2 are the title, map "ody, legend, north arrow, scale "ar, ac nowledgement and map "order. Other elements that might e added are e.g. a !raticule, or name of the ?7

map pro@ection ,"=S-. *ogether, these elements help the map reader to interpret the information shown on the map. *he map ody is, of course, the most important part of the map ecause it contains the map information. *he other elements support the communication process and help the map reader to orientate himself and understand the map topic. For e0ample, the title descri es the su +ect matter and the legend relates map sym ols to the mapped data. Title in detail: *he map title is very important ecause it is usually the first thing a reader will loo. at on a map. It can e compared with a title in a newspaper. It should e short ut give the reader a first idea of what the map is a out. Map 2order in detail: *he map order is a line that defines e0actly the edges of the area shown on the map. 1hen printing a map with a graticule ,which we descri e further down-, you often find the coordinate information of the graticule lines along the order lines, as you can see in Illustration ?: elow. Map 8e!end in detail: A map is a simplified representation of the real world and map s"m%ols are used to represent real o +ects. 1ithout sym ols, we wouldnEt understand maps. *o ensure that a person can correctly read a map, a map legend is used to provide a .ey to all the sym ols used on the map. It is li.e a dictionary that allows you to understand the meaning of what the map shows. A map legend is usually shown as a little o0 in a corner of the map. It contains icons, each of which will represent a type of feature. For e0ample, a house icon will show you how to identify houses on the map ,see Illustration ?: elow-.

Illustration *#: $wo maps !rom the same area, "oth with a water "ody in the "ac ground "ut with di!!erent themes, map sym"ols and colours in the legend. @6

3ou can also use different sym ols and icons in your legend to show different themes. In Illustration ?: a ove you can see a map with a la.e in light lue overlaid with contour lines and spot heights to show information a out the terrain in that area. On the right side you see the same area with the la.e in the ac.ground ut this map is designed to show tourists the location of houses they can rent for their holidays. It uses righter colours, a house icon and more descriptive and inviting words in the legend. .orth arrow in detail: A north arrow ,sometimes also called a compass rose- is a figure displaying the main directions, .orth, South, 9ast and &est. On a map it is used to indicate the direction of Aorth. For e0ample, in GIS this means that a house that is located north from a la.e can e found on top of the la.e on a map. *he road in the east will then e to the right of the water ody on the map, a river in the south will e elow the water ody and if you are searching for a train station to the west of the la.e you will find it on the left side on the map. Scale in detail *he scale of a map, is the value of a single unit of distance on the map, representing distance in the real world. *he values are shown in map units ,meters, feet or degrees-. *he scale can e e0pressed in several ways, for e0ample, in words, as a ratio or as a graphical scale ar ,see Illustration ?; elow-. 93pressin! a scale in words is a commonly used method and has the advantage of eing easily understood y most map users. 3ou can see an e0ample of a word ased scale in Illustration ?;a elow. Another option is the representative fraction )-1* method, where oth the map distance and the ground distance in the real world are given in the same map units, as a ratio. For e0ample, a =F value 8!5<,666 means that any distance on the map is 8F5<,666th of the real distance on the ground ,see Illustration ?; elow-. *he value 5<,666 in the ratio is called the scale denominator. (ore e0perienced users often prefer the representative fraction method, ecause it reduces confusion. 1hen a representative fraction e0presses a very small ratio, for e0ample 8!8666 666, it is called a small scale map. On the other hand if the ratio is very large, for e0ample a 8!<6 666 map, it is called a lar!e scale map. It is handy to remem er that a small scale map covers a lar!e area, and a large scale map covers a small area4 A scale e3pression as a !raphic or %ar scale is another asic method of e0pressing a scale. A ar scale shows measured distances on the map. *he @8

eGuivalent distance in the real world is placed a ove as you can see in Illustration ?;c elow.

Illustration *(: A map scale can "e expressed in words 1a2, as a ratio 1"2, or as graphic or "ar scale 1c2. (aps are usually produced at standard scales of, for e0ample, 8!86 666, 8!5< 666, 8!<6 666, 8!866 666, 8!5<6 666, 8!<66 666. 1hat does this mean to the map readerJ It means that if you multiply the distance measured on the map y the scale denominator, you will .now the distance in the real world. For e0ample, if we want to measure a distance of 866mm on a map with a scale of 8!5<,666 we calculate the real world distance li.e this! 1//mm x 2',/// H 2,'//,/// mm *his means that 866mm on the map is eGuivalent to 5,<66,666mm ,5<6m- in the real word. Another interesting aspect of a map scale, is that the lower the map scale, the more detailed the feature information in the map will e. In Illustration ?< elow, you can see an e0ample of this. Both maps are the same siHe ut have a different scale. *he image on the left side shows more details, for e0ample the houses south2west of the water ody can e clearly identified as separate sGuares. In the image on the right you can only see a lac. clump of rectangles and you are not a le to see each house clearly.

@5

Illustration *': Aaps showing an area in two di!!erent scales. $he map scale on the le!t is 1:2',///. $he map scale on the right is 1:'/,///. Ac0nowled!ment in detail: In the ac.nowledgment area of a map it is possi le to add te0t with important information. For e0ample information a out the Guality of the used data can e useful to give the reader an idea a out details such as how, y whom and when a map was created. If you loo. at a topographical map of your town, it would e useful to .now when the map was created and who did it. If the map is already <6 years old, you will pro a ly find a lot of houses and roads that no longer e0ist or may e never even e0isted. If you .now that the map was created y an official institution, you could contact them and as. if they have a more current version of that map with updated information. Graticule in detail: A graticule is a networ. of lines overlain on a map to ma.e spatial orientation easier for the reader. *he lines can e used as a reference. As an e0ample, the lines of a graticule can represent the earthEs parallels of latitude and meridians of longitude. 1hen you want to refer to a special area on a map during your presentation or in a report you could say! Sthe houses close to latitude 5>.6; F longitude 2:5.88 are often e0posed to flooding during Canuary and Fe ruaryT ,see Illustration ?> elow-.

@:

Illustration *): Graticules 1red lines2 representing the Earth>s parallels o! latitude and meridians o! longitude. $he latitude and longitude values on the map "order can "e used !or "etter orientation on the map.

.ame of the map pro@ection in detail A map pro+ection tries to represent the :2dimensional 'arth with all its features li.e houses, roads or la.es on a flat sheet of paper. *his is very difficult as you can imagine, and even after hundreds of years there is no single pro+ection that is a le to represent the 'arth perfectly for any area in the world. 'very pro+ection has advantages and disadvantages. *o e a le to create maps as precisely as possi le, people have studied, modified, and produced many different .inds of pro+ections. In the end almost every country has developed its own map pro+ection with the goal of improving the map accuracy for their territorial area ,see Illustration ?? elow-.

Illustration **: $he world in di!!erent pro:ections. A Aollweide E%ual Area pro:ection le!t, a ;late 4arree E%uidistant 4ylindrical pro:ection on the right.

@;

1ith this in mind, we can now understand why it ma.es sense to add the name of the pro+ection on a map. It allows the reader to see Guic.ly, if one map can e compared with another. For e0ample, features on a map in a so2 called E%ual Area pro+ection appear very different to features pro+ected in a 4ylindrical E%uidistant pro+ection ,see Illustration ?? a ove-. (ap pro+ection is a very comple0 topic and we cannot cover it completely here. 3ou may want to ta.e a loo. at our previous topic! "oordinate =eference Systems if you want to .now more a out it. #ommon pro%lems ' thin!s to %e aware of: It is sometimes difficult to create a map that is easy to understand and well laid out whilst still showing and e0plaining all the information that the reader needs to .now. *o achieve this, you need to create an ideal arrangement and composition of all the map elements. 3ou should concentrate on what story you want to tell with your map and how the elements, such as the legend, scale ar and ac.nowledgements should e ordered. By doing this, you will have a well designed and educational map, that people would li.e to loo. at and e a le to understand.

&hat have we learned( &etEs wrap up what we covered in this wor.sheet!

Map production means arranging map elements on a sheet of paper. Map elements are the title, map ody, map order, legend, scale, north arrow and the ac.nowledgement. Scale represents the ratio of a distance on the map to the actual distance in the real world. Scale is displayed in map units ,meters, feet or degreesA le!end e0plains all the sym ols on a map. A map should e3plain comple3 information as simpl" as possi%le. (aps are usually always displayed E.orth up .

.ow "ou tr"/ /ere are some ideas for you to try with your learners!

&oad some vector layers in your GIS for your local area. See if your learners can identify e0amples of different types of legend elements such as road types or uildings. "reate a list of legend elements and define what the icons should loo. li.e, so a reader can most easily figure out @<

their meaning in the map. "reate a map layout with your learners on a sheet of paper. #ecide on the title of the map, what GIS layers you want to show and what colours and icons to have on the map. )se the techniGues you learned in *opics 5 and : to ad+ust the sym ology accordingly. 1hen you have a template, open the QGIS (ap "omposer and try to arrange a map layout as planned.

Somethin! to thin0 a%out: If you donEt have a computer availa le, you can use any topographical map and discuss the map design with your learners. Figure out if they understand what the map wants to tell. 1hat can e improvedJ /ow accurately does the map represent the history of the areaJ /ow would a map from 866 years ago differ from the same map todayJ 1urther readin!: 2oo0s:

"hang, Bang2*sung ,566>-! Introduction to Geographic Information Systems. :rd 'dition. (cGraw /ill. ,ISBA 66?6><@7@>#e(ers, (ichael A. ,566<-! Fundamentals of Geographic Information Systems. :rd 'dition. 1iley. ,ISBA 7@8;85>87<-

&e%sites:

http!FFen.wi.ipedia.orgFwi.iFScaleN,maphttp!FFwww.colorado.eduFgeographyFgcraftFnotesFmappro+Fmappro+.html

*he QGIS )ser Guide also has more detailed information on map production provided in QGIS. &hat s ne3t( In the section that follows we will ta.e a closer loo. at vector anal"sis to see how we can use a GIS for more than +ust ma.ing good loo.ing maps4

@>

GIS for 'ducators

Topic G: Vector Spatial Analysis ( uffers! O +ectives! )nderstanding the use of uffering in vector spatial analysis. 9ector, uffer Hone, spatial analysis, uffer distance, dissolve oundary, outward and inward uffer, multiple uffer

Beywords!

Overview: Spatial anal"sis uses spatial information to e0tract new and additional meaning from GIS data. )sually spatial analysis is carried out using a GIS Application. GIS Applications normally have spatial analysis tools for feature statistics ,e.g. how many vertices ma.e up this polylineJ- or geoprocessing such as feature uffering. *he types of spatial analysis that are used vary according to su +ect areas. $eople wor.ing in water management and research ,hydrology- will most li.ely e interested in analysing terrain and modelling water as it moves across it. In wildlife management users are interested in analytical functions that deal with wildlife point locations and their relationship to the environment. In this topic we will discuss uffering as an e0ample of a useful spatial analysis that can e carried out with vector data. 2ufferin! in detail: 2ufferin! usually creates two areas! one area that is within a specified distance to selected real world features and the other area that is %e"ond. *he area that is within the specified distance is called the %uffer 5one. A %uffer 5one is any area that serves the purpose of .eeping real world features distant from one another. Buffer Hones are often set up to protect the environment, protect residential and commercial Hones from industrial accidents or natural disasters, or to prevent violence.

@?

Illustration *+: $he "order "etween the =nited States o! America and Aexico is separated "y a "u!!er 5one. 1;hoto ta en "y SG$ Gim Greenhill 2//)2.

"ommon types of uffer Hones may e green elts etween residential and commercial areas, order Hones etween countries ,see Illustration ?@ a ove-, noise protection Hones around airports, or pollution protection Hones along rivers. In a GIS Application, %uffer 5ones are always represented as vector pol"!ons enclosing other polygon, line or point features ,see Illustrations ?72 @8 elow-.

Illustration *-: A "u!!er 5one around vector points ,ariations in %ufferin!:

Illustration +/: A "u!!er 5one around vector polylines.

Illustration +1: A "u!!er 5one around vector polygons

*here are several variations in uffering. *he %uffer distance or uffer siHe can var" according to numerical values provided in the vector layer attri ute ta le for each feature. *he numerical values have to e defined in map units according to the "oordinate =eference System ,"=S- used with the data. For e0ample, the width of a uffer Hone along the an.s of a river can vary depending on the intensity of the ad+acent land use. For intensive cultivation @@

the uffer distance may e igger than for organic farming ,see Illustration @5 elow and *a le 7 elow-.

=iver Breede =iver Bomati Oran+e *elle river

Illustration +2: .u!!ering rivers with di!!erent "u!!er distances. Ad+acent land use Buffer distance ,metersIntensive vegeta le cultivation Intensive cotton cultivation Organic farming Organic farming 866 8<6 <6 <6

$a"le -: Attri"ute ta"le with di!!erent "u!!er distances to rivers "ased on in!ormation a"out the ad:acent land use. .u!!ers around polyline !eatures, such as rivers or roads, do not have to "e on "oth sides o! the lines. $hey can "e on either the le!t side or the right side o! the line !eature. In these cases the le!t or right side is determined "y the direction !rom the starting point to the end point o! line during digitising. Multiple %uffer 5ones: A feature can also have more than one uffer Hone. A nuclear power plant may e uffered with distances of 86, 8<, 5< and :6 .m, thus forming multiple rings around the plant as part of an evacuation plan ,see Illustration @: elow-.

@7

Illustration +#: .u!!ering a point !eature with distances o! 1/, 1', 2' and #/ m. 2ufferin! with intact or dissolved %oundaries Buffer Hones often have dissolved oundaries so that there are no overlapping areas etween the uffer Hones. In some cases though, it may also e useful for oundaries of uffer Hones to remain intact, so that each uffer Hone is a separate polygon and you can identify the overlapping areas ,see Illustration @; elow-.

Illustration +(: .u!!er 5ones with dissolved 1le!t2 and with intact "oundaries 1right2 showing overlapping areas. 76

2ufferin! outward and inward Buffer Hones around polygon features are usually e0tended outward from a polygon oundary ut it is also possi le to create a uffer Hone inward from a polygon oundary. Say, for e0ample, the #epartment of *ourism wants to plan a new road around =o en Island and environmental laws reGuire that the road is at least 566 meters inward from the coast line. *hey could use an inward uffer to find the 566m line inland and then plan their road not to go eyond that line. #ommon pro%lems ' thin!s to %e aware of: (ost GIS Applications offer uffer creation as an analysis tool, ut the options for creating uffers can vary. For e0ample, not all GIS Applications allow you to uffer on either the left side or the right side of a line feature, to dissolve the oundaries of uffer Hones or to uffer inward from a polygon oundary. A uffer distance always has to e defined as a whole num er ,inte!er- or a decimal num er ,floatin! point value*. *his value is defined in map units ,meters, feet, decimal degrees- according to the "oordinate =eference System ,"=S- of the vector layer. More spatial anal"sis tools Buffering is a an important and often used spatial analysis tool ut there are many others that can e used in a GIS and e0plored y the user. Spatial overla" is a process that allows you to identify the relationships etween two polygon features that share all or part of the same area. *he output vector layer is a com ination of the input features information ,see Illustration @< elow -. *ypical spatial overlay e0amples are!

Intersection! *he output layer contains all areas where oth layers overlap ,intersect-. :nion! the output layer contains all areas of the two input layers com ined. S"mmetrical difference! *he output layer contains all areas of the input layers e0cept those areas where the two layers overlap ,intersect-. Difference! *he output layer contains all areas of the first input layer that do not overlap ,intersect- with the second input layer.

78

Illustration +': Spatial overlay with two input vector layers 1aIinput H rectangle, "Iinput H circle2. $he resulting vector layer is displayed green. &hat have we learned( &etEs wrap up what we covered in this wor.sheet!

2uffer 5ones descri e areas around real world features. Buffer Hones are always vector pol"!ons. A feature can have multiple uffer Hones. *he siHe of a uffer Hone is defined y a %uffer distance. A uffer distance has to e an inte!er or floatin! point value. A uffer distance can e different for each feature within a vector layer. $olygons can e uffered inward or outward from the polygon oundary. Buffer Hones can e created with intact or dissolved oundaries. Besides uffering, a GIS usually provides a variety of vector analysis tools to solve spatial tas.s.

.ow "ou tr"/ /ere are some ideas for you to try with your learners!

Because of dramatic traffic increase, the town planners want to widen the main road and add a second lane. "reate a uffer around the road to find properties that fall within the uffer Hone ,see Illustration @> elow-. For controlling protesting groups, the police want to esta lish a neutral Hone to .eep protesters at least 866 meters from a uilding. "reate a uffer around a uilding and colour it so that event planners can see where the uffer area is. A truc. factory plans to e0pand. *he siting criteria stipulate that a potential site must e within 8 .m of a heavy2duty road. "reate a uffer along a main road so that you can see where potential sites are. Imagine that the city wants to introduce a law stipulating that no ottle stores may e within a 8666 meter uffer Hone of a school or a church. "reate a 8.m uffer around your school and then go and see if there would e any ottle stores too close to your school.

75

Illustration +): .u!!er 5one 1green2 around a roads map 1"rown2. 8ou can see which houses !all within the "u!!er 5one, so now you could contact the owner and tal to him a"out the situation. Somethin! to thin0 a%out: If you donEt have a computer availa le, you can use a toposheet and a compass to create uffer Hones around uildings. (a.e small pencil mar.s at eGual distance all along your feature using the compass, then connect the mar.s using a ruler4 1urther readin!: 2oo0s:

Galati, Stephen =. ,566>-! Geographic Information Systems #emystified. Artech /ouse Inc. ,ISBA 8<@6<:<::I"hang, Bang2*sung ,566>-! Introduction to Geographic Information Systems. :rd 'dition. (cGraw /ill. ,ISBA 66?6><@7@>#e(ers, (ichael A. ,566<-! Fundamentals of Geographic Information Systems. :rd 'dition. 1iley. ,ISBA 7@8;85>87<-

&e%sites:

http!FFwww.manifold.netFdocFtransformN orderN uffers.htm

*he QGIS )ser Guide also has more detailed information on analysing vector data in QGIS. &hat s ne3t( In the section that follows we will ta.e a closer loo. at interpolation as an e0ample of spatial analysis you can do with raster data.

7:

GIS for 'ducators

Topic 1F: Spatial Analysis (Interpolation! O +ectives! Beywords! )nderstanding of interpolation as part of spatial analysis $oint data, interpolation method, Inverse #istance 1eighted, *riangulated Irregular Aetwor.

Overview: Spatial anal"sis is the process of manipulating spatial information to e0tract new information and meaning from the original data. )sually spatial analysis is carried out with a Geographic Information System ,GIS-. A GIS usually provides spatial analysis tools for calculating feature statistics and carrying out geoprocessing activities as data interpolation. In hydrology, users will li.ely emphasiHe the importance of terrain analysis and hydrological modelling ,modelling the movement of water over and in the earth-. In wildlife management, users are interested in analytical functions dealing with wildlife point locations and their relationship to the environment. 'ach user will have different things they are interested in depending on the .ind of wor. they do. Spatial interpolation in detail:

Illustration +*: $emperature map interpolated !rom South A!rican <eather Stations.

Spatial interpolation is the process of using points with .nown values to estimate values at other un.nown points. For e0ample, to ma.e a precipitation ,rainfall- map for your country, you will not find enough evenly spread weather stations to cover the entire region. Spatial interpolation can estimate the 7;

temperatures at locations without recorded data y using .nown temperature readings at near y weather stations ,see Illustration @? a ove-. *his type of interpolated surface is often called a statistical surface. 'levation data, precipitation, snow accumulation, water ta le and population density are other types of data that can e computed using interpolation. Because of high cost and limited resources, data collection is usually conducted only in a limited num er of selected point locations. In GIS, spatial interpolation of these points can e applied to create a raster surface with estimates made for all raster cells. In order to generate a continuous map, for e0ample, a digital elevation map from elevation points measured with a G$S device, a suita le interpolation method has to e used to optimally estimate the values at those locations where no samples or measurements were ta.en. *he results of the interpolation analysis can then e used for analyses that cover the whole area and for modelling. *here are many interpolation methods. In this introduction we will present two widely used interpolation methods called Inverse Distance &ei!htin! ,I#1and Trian!ulated Irre!ular .etwor0s ,*IA-. If you are loo.ing for additional interpolation methods, please refer to the further reading section at the end of this topic. Inverse Distance &ei!hted )ID&* In the I#1 interpolation method, the sample points are weighted during interpolation such that the influence of one point relative to another declines with distance from the un.nown point you want to create ,see Illustration @@ elow-.

7<

Illustration ++: Inverse Bistance <eighted interpolation "ased on weighted sample point distance 1le!t2. Interpolated IB< sur!ace !rom elevation vector points 1right2. Image Source: Aitas, 6., Aitasova, &. 11---22. 1eighting is assigned to sample points through the use of a weighting coefficient that controls how the weighting influence will drop off as the distance from new point increases. *he greater the weighting coefficient, the less the effect points will have if they are far from the un.nown point during the interpolation process. As the coefficient increases, the value of the un.nown point approaches the value of the nearest o servational point. It is important to notice that the I#1 interpolation method also has some disadvantages! *he Guality of the interpolation result can decrease, if the distri ution of sample data points is uneven. Furthermore, ma0imum and minimum values in the interpolated surface can only occur at sample data points. *his often results in small pea.s and pits around the sample data points as shown in Illustration @@ a ove.

7>

In GIS, interpolation results are usually shown as a 5 dimensional raster layer. In Illustration @7 elow, you can see a typical I#1 interpolation result, ased on elevation sample points collected in the field with a G$S device.

Illustration +-: IB< interpolation result !rom irregularly collected elevation sample points 1shown as "lac crosses2. Trian!ulated Irre!ular .etwor0 )TI.* *IA interpolation is another popular tool in GIS. A common *IA algorithm is called Delauna" triangulation. It tries to create a surface formed y triangles of nearest neigh our points. *o do this, circumcircles around selected sample points are created and their intersections are connected to a networ. of non overlapping and as compact as possi le triangles ,see Illustration 76 elow-.

7?

Illustration -/: Belaunay triangulation with circumcircles around the red sample data. $he resulting interpolated $IC sur!ace created !rom elevation vector points is shown on the right. Image Source: Aitas, 6., Aitasova, &. 11---22. *he main disadvantage of the *IA interpolation is that the surfaces are not smooth and may give a +agged appearance. *his is caused y discontinuous slopes at the triangle edges and sample data points. In addition, triangulation is generally not suita le for e0trapolation eyond the area with collected sample data points ,see Illustration 78 elow-.

Illustration -1: Belaunay $IC interpolation result !rom irregularly collected rain!all sample points 1"lue circles2.

7@

#ommon pro%lems ' thin!s to %e aware of: It is important to remem er that there is no single interpolation method that can e applied to all situations. Some are more e0act and useful than others ut ta.e longer to calculate. *hey all have advantages and disadvantages. In practice, selection of a particular interpolation method should depend upon the sample data, the type of surfaces to e generated and tolerance of estimation errors. Generally, a three step procedure is recommended! 8. 'valuate the sample data. #o this to get an idea on how data are distri uted in the area, as this may provide hints on which interpolation method to use. 5. Apply an interpolation method which is most suita le to oth the sample data and the study o +ectives. 1hen you are in dou t, try several methods, if availa le. :. "ompare the results and find the est result and the most suita le method. *his may loo. li.e a time consuming process at the eginning. /owever, as you gain e0perience and .nowledge of different interpolation methods, the time reGuired for generating the most suita le surface will e greatly reduced. Other interpolation methods Although we concentrated on I#1 and *IA interpolation methods in this wor.sheet, there are more spatial interpolation methods provided in GIS, such as =egulariHed Splines with *ension ,=S*-, Briging or *rend Surface interpolation. See the additional reading section elow for a we lin.. &hat have we learned( &etEs wrap up what we covered in this wor.sheet!

Interpolation uses vector points with .nown values to estimate values at un.nown locations to create a raster surface covering an entire area. *he interpolation result is typically a raster layer. It is important to find a suita%le interpolation method to optimally estimate values for un.nown locations. ID& interpolation gives weights to sample points, such that the influence of one point on another declines with distance from the new point eing estimated. TI. interpolation uses sample points to create a surface formed y triangles ased on nearest neigh our point information.

.ow "ou tr"/ 77

/ere are some ideas for you to try with your learners!

*he #epartment of Agriculture plans to cultivate new land in your area ut apart from the character of the soils, they want to .now if the rainfall is sufficient for a good harvest. All the information they have availa le comes from a few weather stations around the area. "reate an interpolated surface with your learners that shows which areas are li.ely to receive the highest rainfall. *he tourist office wants to pu lish information a out the weather conditions in Canuary and Fe ruary. *hey have temperature, rainfall and wind strength data and as. you to interpolate their data to estimate places where tourists will pro a ly have optimal weather conditions with mild temperatures, no rainfall and little wind strength. "an you identify the areas in your region that meet these criteriaJ

Somethin! to thin0 a%out: If you donEt have a computer availa le, you can use a toposheet and a ruler to estimate elevation values etween contour lines or rainfall values etween fictional weather stations. For e0ample, if rainfall at weather station A is <6 mm per month and at weather station B it is 76 mm, you can estimate, that the rainfall at half the distance etween weather station A and B is ?6 mm. 1urther readin!: 2oo0s:

"hang, Bang2*sung ,566>-! Introduction to Geographic Information Systems. :rd 'dition. (cGraw /ill. ,ISBA 66?6><@7@>#e(ers, (ichael A. ,566<-! Fundamentals of Geographic Information Systems. :rd 'dition. 1iley. ,ISBA 7@8;85>87<(itas, &., (itasova, /. ,8777-! Spatial Interpolation. In! $.&ongley, (.F. Goodchild, #.C. (aguire, #.1.=hind ,'ds.-, Geographical Information Systems! $rinciples, *echniGues, (anagement and Applications, 1iley.

&e%sites:

http!FFen.wi.ipedia.orgFwi.iFInterpolation http!FFen.wi.ipedia.orgFwi.iF#elaunayNtriangulation http!FFwww.agt. me.huFpu licNeFfuncintFfuncint.html

*he QGIS )ser Guide also has more detailed information on interpolation tools provided in QGIS. &hat s ne3t(

866

*his is the final wor.sheet in this series. 1e encourage you to e0plore QGIS and use the accompanying QGIS manual to discover all the other things you can do with GIS software4

868

About t"e aut"ors # contributors$ Tim Sutton + 9ditor H 8ead Author *im Sutton is a developer and pro+ect steering committee mem er of the Quantum GIS pro+ect. /e is passionate a out seeing GIS eing Freely availa le to everyone. *im is also a founding mem er of &infiniti "onsulting "". 2 a small usiness set up with the goal of helping people to learn and use open source GIS software. &e%: http:''linfiniti.com 9mail: timIlinfiniti.com Otto Dassau + Assistant Author Otto #assau is the documentation maintainer and pro+ect steering committee mem er of the Quantum GIS pro+ect. Otto has considera le e0perience in using and training people to use Free and Open Source GIS software. &e%: http:''www.nature+consult.de 9mail: otto.dassauI!m3.de Marcelle Sutton + 7ro@ect Mana!er (arcelle Sutton studied english and drama and is a Gualified teacher. (arcelle is also a founding mem er of &infiniti "onsulting "". 2 a small usiness set up with the goal of helping people to learn and use open source GIS software. &e%: http:''linfiniti.com 9mail: marcelleIlinfiniti.com

865

8erato .si%ande J ,ideo 7resenter &erato is a grade 85 scholar living in $retoria. &erato learns Geography at school and has en+oyed learning GIS with us4

Si%on!ile Mthom%eni J ,ideo 7resenter Si ongile lives near Cohannes urg with her young daughter. /er goal is to continue her studies and ecome a nurse. 1or.ing on this pro+ect was the first time Si ongile used a computer.

86:

G%& 'ree Documentation (icense

Version 1.2, November 2002 Copyright (C) 2000,200,,2002 *ree Soft.are *o%ndation, Inc# 8, *ran4lin St, *ifth *loor, 3o!ton, 7 02,,00,90, )S "-eryone i! permitted to copy and di!tri$%te -er$atim copie! of thi! licen!e doc%ment, $%t changing it i! not allo.ed# 0# P:" 73L" 1he p%rpo!e of thi! Licen!e i! to ma4e a man%al, te2t$oo4, or other f%nctional and %!ef%l doc%ment 5free5 in the !en!e of freedom: to a!!%re e-eryone the effecti-e freedom to copy and redi!tri$%te it, .ith or .itho%t modifying it, either commercially or noncommercially# Secondarily, thi! Licen!e pre!er-e! for the a%thor and p%$li!her a .ay to get credit for their .or4, .hile not $eing con!idered re!pon!i$le for modification! made $y other!# 1hi! Licen!e i! a 4ind of 5copyleft5, .hich mean! that deri-ati-e .or4! of the doc%ment m%!t them!el-e! $e free in the !ame !en!e# It complement! the '() 'eneral P%$lic Licen!e, .hich i! a copyleft licen!e de!igned for free !oft.are# ;e ha-e de!igned thi! Licen!e in order to %!e it for man%al! for free !oft.are, $eca%!e free !oft.are need! free doc%mentation: a free program !ho%ld come .ith man%al! pro-iding the !ame freedom! that the !oft.are doe!# 3%t thi! Licen!e i! not limited to !oft.are man%al!/ it can $e %!ed for any te2t%al .or4, regardle!! of !%$<ect matter or .hether it i! p%$li!hed a! a printed $oo4# ;e recommend thi! Licen!e principally for .or4! .ho!e p%rpo!e i! in!tr%ction or reference# ,# PPLIC 3ILI1= (D D"*I(I1I>(S

1hi! Licen!e applie! to any man%al or other .or4, in any medi%m, that contain! a notice placed $y the copyright holder !aying it can $e di!tri$%ted %nder the term! of thi! Licen!e# S%ch a notice grant! a .orld0.ide, royalty0free licen!e, %nlimited in d%ration, to %!e that .or4 %nder the condition! !tated herein# 1he 5Doc%ment5, $elo., refer! to any !%ch man%al or .or4# ny mem$er of the p%$lic i! a licen!ee, and i! addre!!ed a! 5yo%5# =o% accept the licen!e if yo% copy, modify or di!tri$%te the .or4 in a .ay re?%iring permi!!ion %nder copyright la.# 57odified +er!ion5 of the Doc%ment mean! any .or4 containing the Doc%ment or a portion of it, either copied -er$atim, or .ith modification! and&or tran!lated into another lang%age# 5Secondary Section5 i! a named appendi2 or a front0matter !ection of the Doc%ment that deal! e2cl%!i-ely .ith the relation!hip of the p%$li!her! or a%thor! of the Doc%ment to the Doc%ment@! o-erall !%$<ect (or to related matter!) and contain! nothing that co%ld fall directly .ithin that o-erall !%$<ect# (1h%!, if the Doc%ment i! in part a te2t$oo4 of mathematic!, a Secondary Section may not e2plain any mathematic!#) 1he relation!hip co%ld $e a matter of hi!torical connection .ith the !%$<ect or .ith related matter!, or of legal, commercial, philo!ophical, ethical or political po!ition regarding them# 1he 5In-ariant Section!5 are certain Secondary Section! .ho!e title! are de!ignated, a! $eing tho!e of In-ariant Section!, in the notice that !ay! that the Doc%ment i! relea!ed %nder thi! Licen!e# If a !ection doe! not fit the

86;

a$o-e definition of Secondary then it i! not allo.ed to $e de!ignated a! In-ariant# 1he Doc%ment may contain Aero In-ariant Section!# If the Doc%ment doe! not identify any In-ariant Section! then there are none# 1he 5Co-er 1e2t!5 are certain !hort pa!!age! of te2t that are li!ted, a! *ront0 Co-er 1e2t! or 3ac40Co-er 1e2t!, in the notice that !ay! that the Doc%ment i! relea!ed %nder thi! Licen!e# *ront0Co-er 1e2t may $e at mo!t 8 .ord!, and a 3ac40Co-er 1e2t may $e at mo!t 28 .ord!# 51ran!parent5 copy of the Doc%ment mean! a machine0reada$le copy, repre!ented in a format .ho!e !pecification i! a-aila$le to the general p%$lic, that i! !%ita$le for re-i!ing the doc%ment !traightfor.ardly .ith generic te2t editor! or (for image! compo!ed of pi2el!) generic paint program! or (for dra.ing!) !ome .idely a-aila$le dra.ing editor, and that i! !%ita$le for inp%t to te2t formatter! or for a%tomatic tran!lation to a -ariety of format! !%ita$le for inp%t to te2t formatter!# copy made in an other.i!e 1ran!parent file format .ho!e mar4%p, or a$!ence of mar4%p, ha! $een arranged to th.art or di!co%rage !%$!e?%ent modification $y reader! i! not 1ran!parent# n image format i! not 1ran!parent if %!ed for any !%$!tantial amo%nt of te2t# copy that i! not 51ran!parent5 i! called 5>pa?%e5# "2ample! of !%ita$le format! for 1ran!parent copie! incl%de plain SCII .itho%t mar4%p, 1e2info inp%t format, La1eB inp%t format, S'7L or B7L %!ing a p%$licly a-aila$le D1D, and !tandard0conforming !imple C17L, Po!tScript or PD* de!igned for h%man modification# "2ample! of tran!parent image format! incl%de P(', BC* and DP'# >pa?%e format! incl%de proprietary format! that can $e read and edited only $y proprietary .ord proce!!or!, S'7L or B7L for .hich the D1D and&or proce!!ing tool! are not generally a-aila$le, and the machine0generated C17L, Po!tScript or PD* prod%ced $y !ome .ord proce!!or! for o%tp%t p%rpo!e! only# 1he 51itle Page5 mean!, for a printed $oo4, the title page it!elf, pl%! !%ch follo.ing page! a! are needed to hold, legi$ly, the material thi! Licen!e re?%ire! to appear in the title page# *or .or4! in format! .hich do not ha-e any title page a! !%ch, 51itle Page5 mean! the te2t near the mo!t prominent appearance of the .or4@! title, preceding the $eginning of the $ody of the te2t# !ection 5"ntitled B=E5 mean! a named !%$%nit of the Doc%ment .ho!e title either i! preci!ely B=E or contain! B=E in parenthe!e! follo.ing te2t that tran!late! B=E in another lang%age# (Cere B=E !tand! for a !pecific !ection name mentioned $elo., !%ch a! 5 c4no.ledgement!5, 5Dedication!5, 5"ndor!ement!5, or 5Ci!tory5#) 1o 5Pre!er-e the 1itle5 of !%ch a !ection .hen yo% modify the Doc%ment mean! that it remain! a !ection 5"ntitled B=E5 according to thi! definition# 1he Doc%ment may incl%de ;arranty Di!claimer! ne2t to the notice .hich !tate! that thi! Licen!e applie! to the Doc%ment# 1he!e ;arranty Di!claimer! are con!idered to $e incl%ded $y reference in thi! Licen!e, $%t only a! regard! di!claiming .arrantie!: any other implication that the!e ;arranty Di!claimer! may ha-e i! -oid and ha! no effect on the meaning of thi! Licen!e# 2# +":3 1I7 C>P=I(' =o% may copy and di!tri$%te the Doc%ment in any medi%m, either commercially or noncommercially, pro-ided that thi! Licen!e, the copyright notice!, and the licen!e notice !aying thi! Licen!e applie! to the Doc%ment are reprod%ced in all copie!, and that yo% add no other condition! .hat!oe-er to tho!e of thi! Licen!e# =o% may not %!e technical mea!%re! to o$!tr%ct or control the reading or f%rther copying of the copie! yo% ma4e or di!tri$%te# Co.e-er, yo% may accept compen!ation in e2change for copie!# If yo% di!tri$%te a large eno%gh

86<

n%m$er of copie! yo% m%!t al!o follo. the condition! in !ection 9# =o% may al!o lend copie!, %nder the !ame condition! !tated a$o-e, and yo% may p%$licly di!play copie!# 9# C>P=I(' I( 6) (1I1= If yo% p%$li!h printed copie! (or copie! in media that commonly ha-e printed co-er!) of the Doc%ment, n%m$ering more than ,00, and the Doc%ment@! licen!e notice re?%ire! Co-er 1e2t!, yo% m%!t enclo!e the copie! in co-er! that carry, clearly and legi$ly, all the!e Co-er 1e2t!: *ront0Co-er 1e2t! on the front co-er, and 3ac40Co-er 1e2t! on the $ac4 co-er# 3oth co-er! m%!t al!o clearly and legi$ly identify yo% a! the p%$li!her of the!e copie!# 1he front co-er m%!t pre!ent the f%ll title .ith all .ord! of the title e?%ally prominent and -i!i$le# =o% may add other material on the co-er! in addition# Copying .ith change! limited to the co-er!, a! long a! they pre!er-e the title of the Doc%ment and !ati!fy the!e condition!, can $e treated a! -er$atim copying in other re!pect!# If the re?%ired te2t! for either co-er are too -ol%mino%! to fit legi$ly, yo% !ho%ld p%t the fir!t one! li!ted (a! many a! fit rea!ona$ly) on the act%al co-er, and contin%e the re!t onto ad<acent page!# If yo% p%$li!h or di!tri$%te >pa?%e copie! of the Doc%ment n%m$ering more than ,00, yo% m%!t either incl%de a machine0reada$le 1ran!parent copy along .ith each >pa?%e copy, or !tate in or .ith each >pa?%e copy a comp%ter0net.or4 location from .hich the general net.or40%!ing p%$lic ha! acce!! to do.nload %!ing p%$lic0 !tandard net.or4 protocol! a complete 1ran!parent copy of the Doc%ment, free of added material# If yo% %!e the latter option, yo% m%!t ta4e rea!ona$ly pr%dent !tep!, .hen yo% $egin di!tri$%tion of >pa?%e copie! in ?%antity, to en!%re that thi! 1ran!parent copy .ill remain th%! acce!!i$le at the !tated location %ntil at lea!t one year after the la!t time yo% di!tri$%te an >pa?%e copy (directly or thro%gh yo%r agent! or retailer!) of that edition to the p%$lic# It i! re?%e!ted, $%t not re?%ired, that yo% contact the a%thor! of the Doc%ment .ell $efore redi!tri$%ting any large n%m$er of copie!, to gi-e them a chance to pro-ide yo% .ith an %pdated -er!ion of the Doc%ment# F# 7>DI*IC 1I>(S =o% may copy and di!tri$%te a 7odified +er!ion of the Doc%ment %nder the condition! of !ection! 2 and 9 a$o-e, pro-ided that yo% relea!e the 7odified +er!ion %nder preci!ely thi! Licen!e, .ith the 7odified +er!ion filling the role of the Doc%ment, th%! licen!ing di!tri$%tion and modification of the 7odified +er!ion to .hoe-er po!!e!!e! a copy of it# In addition, yo% m%!t do the!e thing! in the 7odified +er!ion: # )!e in the 1itle Page (and on the co-er!, if any) a title di!tinct from that of the Doc%ment, and from tho!e of pre-io%! -er!ion! (.hich !ho%ld, if there .ere any, $e li!ted in the Ci!tory !ection of the Doc%ment)# =o% may %!e the !ame title a! a pre-io%! -er!ion if the original p%$li!her of that -er!ion gi-e! permi!!ion# 3# Li!t on the 1itle Page, a! a%thor!, one or more per!on! or entitie! re!pon!i$le for a%thor!hip of the modification! in the 7odified +er!ion, together .ith at lea!t fi-e of the principal a%thor! of the Doc%ment (all of it! principal a%thor!, if it ha! fe.er than fi-e), %nle!! they relea!e yo% from thi! re?%irement# C# State on the 1itle page the name of the p%$li!her of the

86>

7odified +er!ion, a! the p%$li!her# D# Pre!er-e all the copyright notice! of the Doc%ment# "# dd an appropriate copyright notice for yo%r modification! ad<acent to the other copyright notice!# *# Incl%de, immediately after the copyright notice!, a licen!e notice gi-ing the p%$lic permi!!ion to %!e the 7odified +er!ion %nder the term! of thi! Licen!e, in the form !ho.n in the ddend%m $elo.# '# Pre!er-e in that licen!e notice the f%ll li!t! of In-ariant Section! and re?%ired Co-er 1e2t! gi-en in the Doc%ment@! licen!e notice# C# Incl%de an %naltered copy of thi! Licen!e# I# Pre!er-e the !ection "ntitled 5Ci!tory5, Pre!er-e it! 1itle, and add to it an item !tating at lea!t the title, year, ne. a%thor!, and p%$li!her of the 7odified +er!ion a! gi-en on the 1itle Page# If there i! no !ection "ntitled 5Ci!tory5 in the Doc%ment, create one !tating the title, year, a%thor!, and p%$li!her of the Doc%ment a! gi-en on it! 1itle Page, then add an item de!cri$ing the 7odified +er!ion a! !tated in the pre-io%! !entence# D# Pre!er-e the net.or4 location, if any, gi-en in the Doc%ment for p%$lic acce!! to a 1ran!parent copy of the Doc%ment, and li4e.i!e the net.or4 location! gi-en in the Doc%ment for pre-io%! -er!ion! it .a! $a!ed on# 1he!e may $e placed in the 5Ci!tory5 !ection# =o% may omit a net.or4 location for a .or4 that .a! p%$li!hed at lea!t fo%r year! $efore the Doc%ment it!elf, or if the original p%$li!her of the -er!ion it refer! to gi-e! permi!!ion# G# *or any !ection "ntitled 5 c4no.ledgement!5 or 5Dedication!5, Pre!er-e the 1itle of the !ection, and pre!er-e in the !ection all the !%$!tance and tone of each of the contri$%tor ac4no.ledgement! and&or dedication! gi-en therein# L# Pre!er-e all the In-ariant Section! of the Doc%ment, %naltered in their te2t and in their title!# Section n%m$er! or the e?%i-alent are not con!idered part of the !ection title!# 7# Delete any !ection "ntitled 5"ndor!ement!5# S%ch a !ection may not $e incl%ded in the 7odified +er!ion# (# Do not retitle any e2i!ting !ection to $e "ntitled 5"ndor!ement!5 or to conflict in title .ith any In-ariant Section# ># Pre!er-e any ;arranty Di!claimer!# If the 7odified +er!ion incl%de! ne. front0matter !ection! or appendice! that ?%alify a! Secondary Section! and contain no material copied from the Doc%ment, yo% may at yo%r option de!ignate !ome or all of the!e !ection! a! in-ariant# 1o do thi!, add their title! to the li!t of In-ariant Section! in the 7odified +er!ion@! licen!e notice# 1he!e title! m%!t $e di!tinct from any other !ection title!# =o% may add a !ection "ntitled 5"ndor!ement!5, pro-ided it contain! nothing $%t endor!ement! of yo%r 7odified +er!ion $y -ario%! partie!00for e2ample, !tatement! of peer re-ie. or that the te2t ha! $een appro-ed $y an organiAation a! the a%thoritati-e definition of a !tandard# =o% may add a pa!!age of %p to fi-e .ord! a! a *ront0Co-er 1e2t, and a pa!!age of %p to 28 .ord! a! a 3ac40Co-er 1e2t, to the end of the li!t of Co-er 1e2t! in the 7odified +er!ion# >nly one pa!!age of *ront0Co-er 1e2t and one of 3ac40 Co-er 1e2t may $e added $y (or thro%gh arrangement! made $y) any one entity# If the Doc%ment already incl%de! a co-er te2t for the !ame co-er, pre-io%!ly added $y yo% or $y arrangement made $y the !ame entity yo% are acting on $ehalf of, yo% may not add another/ $%t yo% may replace the old one, on e2plicit permi!!ion from the pre-io%! p%$li!her that added the old one# 1he a%thor(!) and p%$li!her(!) of the Doc%ment do not $y thi! Licen!e gi-e permi!!ion to %!e their name! for p%$licity for or to a!!ert or imply

86?

endor!ement of any 7odified +er!ion# 8# C>73I(I(' D>C)7"(1S =o% may com$ine the Doc%ment .ith other doc%ment! relea!ed %nder thi! Licen!e, %nder the term! defined in !ection F a$o-e for modified -er!ion!, pro-ided that yo% incl%de in the com$ination all of the In-ariant Section! of all of the original doc%ment!, %nmodified, and li!t them all a! In-ariant Section! of yo%r com$ined .or4 in it! licen!e notice, and that yo% pre!er-e all their ;arranty Di!claimer!# 1he com$ined .or4 need only contain one copy of thi! Licen!e, and m%ltiple identical In-ariant Section! may $e replaced .ith a !ingle copy# If there are m%ltiple In-ariant Section! .ith the !ame name $%t different content!, ma4e the title of each !%ch !ection %ni?%e $y adding at the end of it, in parenthe!e!, the name of the original a%thor or p%$li!her of that !ection if 4no.n, or el!e a %ni?%e n%m$er# 7a4e the !ame ad<%!tment to the !ection title! in the li!t of In-ariant Section! in the licen!e notice of the com$ined .or4# In the com$ination, yo% m%!t com$ine any !ection! "ntitled 5Ci!tory5 in the -ario%! original doc%ment!, forming one !ection "ntitled 5Ci!tory5/ li4e.i!e com$ine any !ection! "ntitled 5 c4no.ledgement!5, and any !ection! "ntitled 5Dedication!5# =o% m%!t delete all !ection! "ntitled 5"ndor!ement!5# H# C>LL"C1I>(S >* D>C)7"(1S =o% may ma4e a collection con!i!ting of the Doc%ment and other doc%ment! relea!ed %nder thi! Licen!e, and replace the indi-id%al copie! of thi! Licen!e in the -ario%! doc%ment! .ith a !ingle copy that i! incl%ded in the collection, pro-ided that yo% follo. the r%le! of thi! Licen!e for -er$atim copying of each of the doc%ment! in all other re!pect!# =o% may e2tract a !ingle doc%ment from !%ch a collection, and di!tri$%te it indi-id%ally %nder thi! Licen!e, pro-ided yo% in!ert a copy of thi! Licen!e into the e2tracted doc%ment, and follo. thi! Licen!e in all other re!pect! regarding -er$atim copying of that doc%ment# I# '':"' 1I>( ;I1C I(D"P"(D"(1 ;>:GS

compilation of the Doc%ment or it! deri-ati-e! .ith other !eparate and independent doc%ment! or .or4!, in or on a -ol%me of a !torage or di!tri$%tion medi%m, i! called an 5aggregate5 if the copyright re!%lting from the compilation i! not %!ed to limit the legal right! of the compilation@! %!er! $eyond .hat the indi-id%al .or4! permit# ;hen the Doc%ment i! incl%ded in an aggregate, thi! Licen!e doe! not apply to the other .or4! in the aggregate .hich are not them!el-e! deri-ati-e .or4! of the Doc%ment# If the Co-er 1e2t re?%irement of !ection 9 i! applica$le to the!e copie! of the Doc%ment, then if the Doc%ment i! le!! than one half of the entire aggregate, the Doc%ment@! Co-er 1e2t! may $e placed on co-er! that $rac4et the Doc%ment .ithin the aggregate, or the electronic e?%i-alent of co-er! if the Doc%ment i! in electronic form# >ther.i!e they m%!t appear on printed co-er! that $rac4et the .hole aggregate# J# 1: (SL 1I>( 1ran!lation i! con!idered a 4ind of modification, !o yo% may di!tri$%te

86@

tran!lation! of the Doc%ment %nder the term! of !ection F# :eplacing In-ariant Section! .ith tran!lation! re?%ire! !pecial permi!!ion from their copyright holder!, $%t yo% may incl%de tran!lation! of !ome or all In-ariant Section! in addition to the original -er!ion! of the!e In-ariant Section!# =o% may incl%de a tran!lation of thi! Licen!e, and all the licen!e notice! in the Doc%ment, and any ;arranty Di!claimer!, pro-ided that yo% al!o incl%de the original "ngli!h -er!ion of thi! Licen!e and the original -er!ion! of tho!e notice! and di!claimer!# In ca!e of a di!agreement $et.een the tran!lation and the original -er!ion of thi! Licen!e or a notice or di!claimer, the original -er!ion .ill pre-ail# If a !ection in the Doc%ment i! "ntitled 5 c4no.ledgement!5, 5Dedication!5, or 5Ci!tory5, the re?%irement (!ection F) to Pre!er-e it! 1itle (!ection ,) .ill typically re?%ire changing the act%al title# 9# 1":7I( 1I>( =o% may not copy, modify, !%$licen!e, or di!tri$%te the Doc%ment e2cept a! e2pre!!ly pro-ided for %nder thi! Licen!e# ny other attempt to copy, modify, !%$licen!e or di!tri$%te the Doc%ment i! -oid, and .ill a%tomatically terminate yo%r right! %nder thi! Licen!e# Co.e-er, partie! .ho ha-e recei-ed copie!, or right!, from yo% %nder thi! Licen!e .ill not ha-e their licen!e! terminated !o long a! !%ch partie! remain in f%ll compliance# ,0# *)1):" :"+ISI>(S >* 1CIS LIC"(S" 1he *ree Soft.are *o%ndation may p%$li!h ne., re-i!ed -er!ion! of the '() *ree Doc%mentation Licen!e from time to time# S%ch ne. -er!ion! .ill $e !imilar in !pirit to the pre!ent -er!ion, $%t may differ in detail to addre!! ne. pro$lem! or concern!# See http:&&...#gn%#org&copyleft&# "ach -er!ion of the Licen!e i! gi-en a di!ting%i!hing -er!ion n%m$er# If the Doc%ment !pecifie! that a partic%lar n%m$ered -er!ion of thi! Licen!e 5or any later -er!ion5 applie! to it, yo% ha-e the option of follo.ing the term! and condition! either of that !pecified -er!ion or of any later -er!ion that ha! $een p%$li!hed (not a! a draft) $y the *ree Soft.are *o%ndation# If the Doc%ment doe! not !pecify a -er!ion n%m$er of thi! Licen!e, yo% may choo!e any -er!ion e-er p%$li!hed (not a! a draft) $y the *ree Soft.are *o%ndation# DD"(D)7: Co. to %!e thi! Licen!e for yo%r doc%ment! 1o %!e thi! Licen!e in a doc%ment yo% ha-e .ritten, incl%de a copy of the Licen!e in the doc%ment and p%t the follo.ing copyright and licen!e notice! <%!t after the title page: Copyright (c) =" : =>): ( 7"# Permi!!ion i! granted to copy, di!tri$%te and&or modify thi! doc%ment %nder the term! of the '() *ree Doc%mentation Licen!e, +er!ion ,#2 or any later -er!ion p%$li!hed $y the *ree Soft.are *o%ndation/ .ith no In-ariant Section!, no *ront0Co-er 1e2t!, and no 3ac40Co-er 1e2t!# copy of the licen!e i! incl%ded in the !ection entitled 5'() *ree Doc%mentation Licen!e5# If yo% ha-e In-ariant Section!, *ront0Co-er 1e2t! and 3ac40Co-er 1e2t!, replace the 5.ith###1e2t!#5 line .ith thi!: .ith the In-ariant Section! $eing LIS1 1C"I: 1I1L"S, .ith the *ront0Co-er

867

1e2t! $eing LIS1, and .ith the 3ac40Co-er 1e2t! $eing LIS1# If yo% ha-e In-ariant Section! .itho%t Co-er 1e2t!, or !ome other com$ination of the three, merge tho!e t.o alternati-e! to !%it the !it%ation# If yo%r doc%ment contain! nontri-ial e2ample! of program code, .e recommend relea!ing the!e e2ample! in parallel %nder yo%r choice of free !oft.are licen!e, !%ch a! the '() 'eneral P%$lic Licen!e, to permit their %!e in free !oft.are#

886

)GIS &ser Manual

888