Training Manual - GIS and Remote Sensing

8/8/2019 Training Manual - GIS and Remote Sensing

1/65

Technologies for Conservation and Development, www.t4cd.org, email:[email protected]

Data collection, GIS & remote sensing

An initiative funded by the

Facilitated by
http://www.t4cd.org/http://www.t4cd.org/http://www.t4cd.org/


2/65


Use of this Module

Purpose

The purpose of this module is to offer a basic insight into the meaning, aspects and functions

of GIS. The module is broken down into chapters that will be presented as separate sessions.

Learning Objectives

When you have completed this unit, you should be able to:

- Understanding the functions of GIS

- Defining and understanding the debates around GIS

- Understanding the concepts of Scale in maps, Projection and co-ordinates.

- Understanding Data Models in GIS

- Understanding Data Inputting and analysis in GIS

- Knowing examples of GIS applications in the Southern Africa region and abroad.

- Understand Remote Sensing and its value

- Understanding characteristics of images in GIS

- Understanding Satellite orbits and Swaths.

- Understanding weather satellites and- The reception and transmission of data.

A t4cd Training Manual ii


3/65


Table of ContentsData collection, GIS & remote sensing ............................................................................................ i

.......................................................................................................................................................... iUse of this Module ...........................................................................................................................ii

Table of Contents .............................................................................................................................1Table of Figures ............................................................................................................................... 2

Definition of Study Area .................................................................................................................3

Session 1: Understanding Functions of GIS .................................................................................... 5Session 2: Defining and Understanding the Debates Around GIS .................................................. 7

Session 3: Understanding Concepts of Scale, Projection and Co- Ordinates in GIS......................9

Session 4: Understanding Data Models in GIS ..............................................................................14Session 5: Understanding Data Inputting in GIS ........................................................................... 20

Session 6: Knowing Examples of Applications of GIS in Conservation .......................................26

Session 7: Understanding Remote Sensing in GIS ........................................................................27Session 8: Understanding Characteristics of Images ..................................................................... 38Session 9: Understanding Satellite Characteristics-Orbits and Swaths .........................................41

Session 10: Understanding Weather Satellites ..............................................................................54

Session 11: Understanding Reception, Transmission and Processing of Data .............................. 5714.0 Practical Exercises .................................................................................................................59

15.0 Exercises 1 (Adding and Viewing data) ................................................................................ 60

16.0 Exercises 2 (Viewing and Editing data tables) ......................................................................61References ...................................................................................................................................... 62

A t4cd Training Manual 1


4/65


Table of Figures

Figure 1: The relationship between GIS, computer-aid design, computer cartography, databasemanagement and remote sensing information systems....................................................................8

Figure 2: Conformal Projection.....................................................................................................12

Figure 3: Representation of a projection........................................................................................12Figure 4: Example of a vector file (Burrough, 1986)....................................................................14

Figure 5: Example of a raster file (Burrough, 1986).....................................................................16

Figure 6: Map section showing pixels of raster. (Burrough, 1986)...............................................17Figure 7: The remote sensing process (CCRS/CRT, 2007)...........................................................27

Figure 8: Three wave (CCRS/CRT, 2007)....................................................................................29

Figure 9: The Electromagnetic Spectrum (CCRS/CRT, 2007).....................................................29

Figure 10: Scattering (CCRS/CRT, 2007).....................................................................................31Figure 11: Rayleigh scattering (CCRS/CRT, 2007)......................................................................31

Figure 12: Mie scattering (CCRS/CRT, 2007)..............................................................................32

Figure 13: Nonselective scattering (CCRS/CRT, 2007)................................................................32Figure 14: Absorption (CCRS/CRT, 2007)...................................................................................32

Figure 15: Target Interactions (ccrs, 2007)...................................................................................34

Figure 16: Interactions with vegetation (CCRS, 2007) ................................................................34Figure 17: Interactions with vegetation (CCRS, 2007).................................................................35

Figure 18: Spectral Curve sowing Interactions with vegetation and water (CCRS/CRT, 2007....35

Figure 19: Passive Sensors (CCRS/CRT, 2007)...........................................................................36Figure 20: Active Sensors (CCRS/CRT, 2007).............................................................................36

Figure 21: Characteristics of images (CCRS/CRT, 2007).............................................................38Figure 22: Platforms (CCRS/CRT, 2007)......................................................................................39

Figure 23: Satellite orbits (CCRS/CRT, 2007).............................................................................41Figure 24: Spatial Resolution, Pixel Size, and Scale (CCRS/CRT, 2007)....................................43

Figure 25: Spectral Resolution (CCRS/CRT, 2007)......................................................................45

Figure 26: Radiometric Resolution (CCRS/CRT, 2007)...............................................................46Figure 27: Temporal Resolution (CCRS/CRT, 2007)...................................................................46

Figure 28: Cameras and Aerial Photography (CCRS/CRT, 2007)...............................................47

Figure 29: Across-track scanning, (CCRS/CRT, 2007).................................................................50Figure 30: Along-track scanners (CCRS/CRT, 2007)...................................................................51

Figure 31: Geometric distortion (CCRS/CRT, 2007)....................................................................52

Figure 32: Hemispheric images (CCRS/CRT, 2007)....................................................................54Figure 33: Landsat (CCRS/CRT, 2007).........................................................................................54Figure 34: Reception, Transmission, and Processing (CCRS/CRT, 2007)...................................57



5/65


Definition of Study Area

2.1 History and Development of GISGIS is relatively a science and has developed through a number of well-defined steps,normally paralleling computer technological development.1. Pioneer/Research Period: 1950s - 1975 in USA and UK and showed Strength of

individuals, with limited international links, little data and Ambition was greater thantechnological capability.

2. Formal Experimental Research: 1973 Early 1980s Role on individual diminishedat international level, still strong at local level (e.g. head of mapping agencies)

3. Commercial Phase: Started in 1982 and was characterized by Strong competitionbetween vendors

4. User Dominance and Control: End to the beginning of GIS

2.2 What Is GIS Remote Sensing and how can they help you?

A geographic information system (GIS) is a software tool for mapping and analyzing justabout any object, you can think of on earth - from forestland to urban landscape,earthquake faults to tennis courts, oilrigs to restaurants. GIS technology integratespowerful database capabilities with the unique visual perspective of a map. This makesGIS unique among information systems. Its analyses can be used in a wide range ofpublic and private enterprises, helping in planning, cost reduction, and better-informeddecision-making (ESRI, 2007).

2.3 Definition of GIS

GIS (Geographical Information Systems) are systems which deal with geographicalinformation represented as a series of geographical features. This constitutes two dataelements that are Geographical data (locational) and Attribute data (statistical) (ccrs,2007). It is this geographical data that differentiates GIS from other information systemssuch as Land Information Systems (LIS), Land and Resource Information Systems(LRIS), Urban Information Systems (URIS), Environmental Information Systems (ERIS)and Cadastral Information Systems (CAIS). Diversity in the field of GIS makes it difficultto define and most of the debate centres on GIS as software, hardware and as aninformation processor. From the above discussion, GIS thus can be best defined as:

1. A system for capturing, storing, checking, integrating, manipulating, analysingand displaying data which are spatially referenced to the earth.2. This includes data input, data storage, data analysis and data reporting sub-systems.

2.4 The Concept of Remote Sensing



6/65


If you have heard the term "remote sensing" before you may have asked, "what does itmean?" It' is a rather simple, familiar activity that we all do in our daily life but that getscomplicated when we increase the scale. As you view the screen of your computermonitor, you are actively engaged in remote sensing.

A physical quantity (light) emanates from that screen, which is a source of radiation.The radiated light passes over a distance, and thus is "remote" to some extent, until itencounters and is captured by a sensor (your eyes). Each eye sends a signal to aprocessor (your brain), which records the data and interprets this into information.Several of the human senses gather their awareness of the external world almostentirely by perceiving a variety of signals, either emitted or reflected, actively orpassively, from objects that transmit this information in waves or pulses.

Thus, one hears disturbances in the atmosphere carried as sound waves, experiencessensations such as heat (either through direct contact or as radiant energy), reacts tochemical signals from food through taste and smell, and is cognizant of certain materialproperties such as roughness through touch. In addition, one recognizes shapes,colours, and relative positions of exterior objects and classes of materials by means ofseeing visible light issuing from them. In the previous sentence, all sensations that arenot received through direct contact are remotely sensed.

However, in practice we do not usually think of our bodily senses as remote sensors inthe way we us the term technically. A formal and comprehensive definition of appliedremote sensing, as it is customarily formulated to include determination of geophysicalparameters, is:

The acquisition and measurement of data/information on some property (ies) of aphenomenon, object, or material by a recording device not in physical, intimatecontact with the feature(s) under surveillance. Techniques involve amassingknowledge pertinent to environments by measuring force fields, electromagneticradiation, or acoustic energy employing cameras, radiometers and scanners,lasers, radio frequency receivers, radar systems, sonar, thermal devices,seismographs, magnetometers, gravimeters, scintillometers, and otherinstruments. National, Provincial and local Government, parastatals, NGOs andprivate companies all make use of or could make use of Remote sensing.



7/65


Session 1: Understanding Functions of GIS

3.1 Why Use GIS?

3.1.1 Improve Organizational Integration

One of the main benefits of GIS is improved management of an organization andresources. A GIS can link data sets together by common location data, such asaddresses, which helps departments and agencies share their data. By creating ashared database, one department can benefit from the work of another; data can becollected once and used many times (ESRI, 2007).

3.1.2 Make Better Decisions

The old adage "better information leads to better decisions" is true for GIS. A GIS is notjust an automated decision making system but a tool to query, analyze, and map data insupport of the decision making process. For example, GIS can be used to help reach adecision about the location of a new housing development that has minimalenvironmental impact, is located in a low risk area, and is close to a population center.The information can be presented succinctly and clearly in the form of a map andaccompanying report, allowing decision makers to focus on the real issues rather thantrying to understand the data. Because GIS products can be produced quickly, multiplescenarios can be evaluated efficiently and effectively.

3.1.2 Make Maps

For simplicity's sake, we often call GIS "mapping software." We most often associatemaps with physical geography, but GIS is flexible enough to map any kind of terrain,even the human body. GIS can map any data you wish. Making maps with GIS is muchmore flexible than traditional manual or automated cartography approaches. A GIScreates maps from data pulled from databases. Existing paper maps can be digitizedand translated into the GIS as well. The GIS based cartographic database can be bothcontinuous and scale free. Map products can then be created centered on any location,at any scale, and showing selected information symbolized effectively to highlightspecific characteristics. A map can be created anytime to any scale for anyone, as long

as you have the data. This is important because often we say, "I see" to mean, "Iunderstand." Pattern recognition is something human beings excel at. There is a vastdifference between seeing data in a table of rows and columns and seeing it presentedin the form of a map. The difference is not simply aesthetic, it is conceptual; it turns outthat the way you see your data has a profound effect on the connections you make andthe conclusions you draw from it. GIS gives you the layout and drawing tools that helppresent facts with clear, compelling documents.



8/65


3.1.3 The Future of GIS

Many disciplines can benefit from GIS techniques. An active GIS market has resulted inlower costs and continual improvements in the hardware and software components ofGIS. These developments will, in turn, result in a much wider application of thetechnology throughout government, business, and industry.

Through a function known as visualization, a GIS can be used to produce images - notjust maps, but drawings, animations, and other cartographic products. These imagesallow researchers to view their subjects in new and different ways than before. Theimages often are equally helpful in conveying the technical concepts of GIS studysubjects to non-scientists.

GIS and related technology will help greatly in the management and analysis of theselarge volumes of data, allowing for better understanding of terrestrial processes andbetter management of human activities to maintain world economic vitality andenvironmental quality.



9/65


Session 2: Defining and Understanding the Debates AroundGIS

4.1 Differing Definitions for Geographical Information Systems (GIS)

There are a number of reasons that makes it difficult to define a GIS. Although therehas been some debate about the origin of the term and the date of initiation of work inthe field, it is clear the GIS are relatively recent phenomena. The last 35 years has seena rapid rate of development in the GIS field, leading to a period of intense activity in thelast 10 years or so.

The recent commercial orientation of much GIS activity has led to a great deal of

rhetoric. An increase of new computer systems intended for GIS, many of which areexisting systems that are re-packaged and re-labeled thus leading to marketexploitation. In conjunction to this, there has been a rise in the number of GISconsultants, many of whom offer conflicting advice.

The GIS field is further characterized by a great diversity of application. GIS is anintegrated system, which brings together ideas developed in many areas including fieldsof agriculture, botany, computing, economics, mathematics, surveying, geology,hydrology, environmental sciences and geography. Inevitably, it is difficult to distinguishbetween the competing claims of different organizations and individuals all of whichwish to be represented in a vibrant and profitable sector.

It is also difficult to define a GIS, as there are many ways in defining and classifyingobjects and subjects. Not surprising there are many different methods that are appliedto define a GIS. The main reason for to this definitional difficulties come from academicdebate about the central focus of current GIS activity. Some believe that the hardwareand software are the central focus; others argue that information processing orapplications are the key elements.

This has lead to GIS being seen by many as a special case of information systems. Theinterpretation of data leads to valuable information, which is a symbolic representationof features. The value of this information depends on many factors such as time scale,

context, cost of collection, storage, analysis or manipulation and availability.



10/65


Figure 1: The relationship between GIS, computer-aid design, computer cartography,

database management and remote sensing information systems.

4.1.2 What distinguishes GIS from other information systems?

GIS integrates spatial and other kinds of information within one system, it offers aconsistent framework for analyzing space. GIS makes connections between activities based on spatial proximity. GIS provides the mechanisms for undertaking the manipulation, analysis anddisplay of geographic knowledge.


ComputerCartograph

y

Remote

Sensing

Database

Management

Computer-aided

Design

GIS


11/65


Session 3: Understanding Concepts of Scale, Projection andCo- Ordinates in GIS.

5.1 Map Scale

The scale of a map is the ratio between distances on the map and correspondingdistances on the ground. If a map has a scale of 1:50 000, then 1 cm on the map equals50 000 cm or 0.5 km on the Earth's surface. The use of small scale and large scale isoften confusing. A large-scale map shows small areas, for example 1 20 or 1:1 000 andit is possible to identify small features from the map - for example individual buildings,streets. A small-scale map shows large areas, for example 1:50 000 or 1: 1 000 000. Atthese scales, a town is represented as an individual symbol. Thus, scale of the mapcontrols not only how features are represented, but also what features are displayed.This links to generalization of features. In other words, Scale determines resolution of

data, for example simplification of reality. It is used as a compromise to attain astandard for many different users due to time and cost?

SCALE: Scale of a map represents the relationship between the size of the map andthe size of the actual area that the map represents and can be represented in threeways:-

A statement or word scale 2cm to 1Km A representative fraction 1:50 000. This uses the same unit for the map as for

land. In the age of digital information and copying, one has to be careful of thisscale. You can enlarge on the photocopier or scan something in and this scale iscompletely inadequate and inaccurate.

Line or linear scale Straight line marking off units indicating the equivalentdistance on the land and looks like a ruler on the map.

Key Scale Measurements in SA: 1:50 000 topographic sheets and 1:10 000orthophoto maps

5.2 Map Referencing Systems

Maps are frequently used as a method to determine location in space, thus it isimportant to have an understanding of how real world features are measured andinitially recorded on a map, particularly when that map data may be captured into a

digital format for use in a GIS. Thematic and topographic maps generally representfeatures or some aspect of the real world. They are thus an abstraction of reality and assuch may contain some degree of generalization and distortion in their spatial properties(for example the representation of a road and rail line running parallel to each other).

In general, terms, the overall shape of the Earth needs to be translated into a two-dimensional flat surface in order for it to be represented as a map. A modified spheroidknown as a geoid (not a true spheroid as the Polar Regions are slightly flattened) is



12/65


used to represent the Earth. Positions can be measured relative to the center of thegeoid that is then divided into imaginary lines known as graticules, separating thenorth/south and east/west hemispheres. The graticules are more commonly referred toas latitude or parallels (east west) and longitude or meridians (north south).Measurements using graticules (latitude/longitude) are read in degrees, minutes andseconds (note: need to convert to decimal degrees for a GIS). Often thesemeasurements are known as geographic coordinates. Map scaling and referencing inturn affect such things as the resolution and angle of projection.

5.3 Properties of Resolution

The smallest distinguishable difference between two measurable values The smallest distance over it is possible to record change

Need to consider what is visible to the naked eye and the human operator andthe smallest distance that particular hardware and software can accept Projection helps to identify location on earth. Based on measurement of

displacement from a given location Two types:

5.3.1 Plane:

Cartesian plane, this is very easy to work with Two axes X; Y co-ordinates all points are measured relative to origin (0; 0), axes

at 90 degrees to each other and uses Pythagoras to determine distance. Useful for plotting individual points on the ground and if mapping a small piece of

the earth

5.3.2 Projection:

Use of mathematical formula to take into account the Earth's shape Use of every odd meridian in the South African LO system as centerline of least

distortion. There are some limitations of this approach :- Uses a 2 degree block - 1 degree either side of the central meridian No distortion in the vertical, only in the horizontal Need to quote 'central meridian' used when applying LO system Rest of the World 2 degree distortion, SA stricter at 1 degree Problem with edge of maps, if need an area greater than on the map, then need

to revert to the global systems Rounding off can lead to error - makes surveyors paranoid!

5.3 Map Projections

Many map projections have been developed to represent the Earth in two-dimensions.Transforming the surface of a globe into a two-dimensional sheet of paper or computerscreen means that distortion of features - size, shape, distance and direction - will



13/65


14/65


Figure 2: Conformal Projection

Conformal means showing any small area in its correct shape thus showing truedirection and should therefore be used whenever the direction is important The axis of Mercator's projection is the equator - the area on either side of theequator will therefore reflect accurate direction and shape - but as you go towardsthe poles, areas are grossly exaggerated, for example Greenland looks bigger thanSouth America Gauss Central Meridian - in South Africa to better reflect area and direction weuse a system developed by Karl Gauss, a mathematician who turned the Mercatorprojection onto its side = Transverse Mercator projection. Thus, a line of longitudebecomes the central meridian as opposed to the equator as is the case with thestandard Mercator projection.

Figure 3: Representation of a projection

Clarke 1880 spheroid - the earth is not a perfect sphere - it has slight flattening atthe poles caused by its rotation and gravitation. To compensate for this, Clarke, anEnglish geodesist computed the dimensions of the earth as a spheroid and our 1:50000 series are based on these calculations.

5.6.1.1 Equal Area Projection:

'Lambert Projection', usually at 1:500 000 scale Area is retained, angles and shapes become distorted

5.6.1.2 Equidistant Projection:

Deals with small areas only, distance are kept correct, all else h distorted


Cylinder

The cylinder touches theglobe around theequator

Equator


15/65


5.6.1.3 The Gauss Conformal Projection (Transverse Mercator)

At the bottom of each 1:50, 000 topographic sheet in South Africa is printed thefollowing:

Gauss Conform Projection, Central Meridian 31 (this can vary) East, Clarke 1880Spheroid.



16/65


Session 4: Understanding Data Models in GIS

6.1 Spatial Data

Spatial data is a fundamental component of any GIS. There are a number of differentdata models available that allow for the storage of spatial data within a GIS.

6.1.1 Forms of Spatial Data

There are four types of spatial object, which make up all the features on a map. Theseare the basic geographical entities or objects and are frequently used in GISterminology. On a map symbols, colors and variable size may depict these entities. Interms of storage in a GIS, however, these features may have attributes or descriptorslinked to them. The type of entity used in the GIS varies, depending on the scale of the

map under construction.6.1.2 Vector files

Vectors refer to the storage of information about points, lines and polygons which isstored in terms of their co-ordinates. In fig 4, a vector file would describe a blue squareby locating the corners a, b, c and d and filling the polygon with the colour blue. In avector file, every object has a location. This is very useful for GIS analyses because it ispossible to get information such as how far apart two objects might be or whether twoareas overlap.

Figure 4: Example of a vector file (Burrough, 1986)

A Point: This is the simplest representation of a spatial object but the choice of entitiesthat will be represented as points depends on the scale of the map or study. One of thekey attributes of a point is its geodetic location, often represented as a pair of numbers(such as latitude and longitude).

A Line: On a map lines generally consist of two or more points joined together by astraight line. The closeness of the points dictates the quality of the linear representation,so a convoluted line would require many points to represent it accurately. Lines have nowidth, and thus, a specific location must be on one side of the line or the other, butnever on the line itself.



17/65


Nodes - are special kinds of points, usually indicating the junction between lines or theends of line segments.

Area or Polygon: This type of spatial information is composed of connecting lines thatform a closed area. It may be made up of a single closed loop or a number of lineswhose start and end points join. Simple polygons are undivided areas, while complexpolygons are divided into areas of different characteristics.

Surfaces: Surfaces are a more complex description of a geographical entity and areused to represent continuous data. Examples include terrain relief, and environmentalparameters such as temperature, rainfall and pollution concentrations. Surface datamay be held as point data where a third value represents the height, or isolines(contours) which are lines of equal value. Surface data can be stored in a variety ofdifferent ways, using both the raster and vector data models.

There are two basic data models: vector and raster. The vector model is based onpoints joined together to form lines and lines joined together to form polygons. Thismodel can precisely replicate map information. The raster model representsgeographical entities by dividing the area of interest into a number of equal grid cells.These models represent an abstraction of the real world and, in the case of spatial data;entities or objects are represented as points, lines, polygons and surfaces. The datamodels used in GIS are the building blocks for the representation and organization ofspatial data within the information system.

For each of these data models there are also a number of data structures that representthe precise way in which the data model is implemented within the GIS. It is easy to

confuse these two terms data model and data structure. The data structure refers to thelogical arrangement of the data within the computer system. Data structures for theraster model include chain coding, run-length encoding and quadtree coding. Datastructures for the vector model include unlinked or spaghetti and topological structuring.

The raster data model divides the entire area into a regular grid of cells, each cellcontains a single value and every location in the study area corresponds to a cell in theraster. The vector model, conversely, uses discrete segments or points to identifylocations and discrete objects. These are formed by connecting line segments. Vectorobjects do not necessarily fill space and not all locations in space need to bereverenced in the model. In general, this makes the vector model a more efficient way

of storing spatial data. Conceptually however, the raster model is the simpler datamodel.

6.1.3 Raster Data

Geographical space is divided into equal sized cells. Each cell has a single value, whichcould relate to a category of information within a theme. The most important componentof a layer is its resolution. The resolution of a map or image is the smallest feature thatcan be represented or is visible. In terms of a raster grid, it is dependent upon the pixel:



18/65


picture element size. Features that are smaller than a grid cell may not be represented.High resolution refers to rasters with a small cell dimension and means lots of detailand., therefore, lots of cells. The converse is true for low resolution. Rasters with lowresolution contain large cell dimensions, there are fewer cells and these cells cover awider geographical area.

6.2 Raster Data Structures

6.2.1 Raster files

In a raster file, the entire study area is divided up into a regular grid. Each grid square orpixel is assigned a value the information is stored pixel by pixel. In fig 5, a raster filewould describe the blue square by listing which pixels were coloured blue (rows 2 to 4and columns b to d). In a raster file, every location has a value. Raster files are veryuseful in GIS for mapping how a continuous variable (such as altitude) varies overspace.

Figure 5: Example of a raster file (Burrough, 1986)

When you scan a paper map and create an image file (such as a *.gif or *.jpg file), the

image is stored in raster format. You can use the image in MapInfo by registering itsexact spatial location. To do this you need to know the precise co-ordinates of at leastfour points on the image. MapInfo will use these control points to locate the imagecorrectly. You can open vector and raster images at the same time. However,remember that you will only be able to select and work with map objects in a vectorimage. Raster images do not contain map objects, just a grid of pixels each with its ownattribute value(s). Fig 6 shows a small section of a raster image, if you look carefully,you can see how it is composed of black, grey and white pixels.



19/65


Figure 6: Map section showing pixels of raster. (Burrough, 1986)

You can create a vector image from a raster image by tracing the outline of objects thatyou need using MapInfo's drawing tools. MapInfo places a cosmetic layer over an openraster image. You can draw on the cosmetic layer without affecting the imageunderneath. You can then save your drawing as a new vector image.

Run Length Coding: Geographical data tends to be spatially auto-correlated, meaningthat objects that are close to each other tend to have similar attributes. Due to thisprinciple, neighboring pixels or cells are expected to have similar values. Thus insteadof repeating pixel values, it is possible to code the raster as groups of values, for

example: run, length and value. By doing this, the amount of storage space required tohold the information is reduced. This technique would not be of use for digital elevationmodel data or any type of data where neighboring pixels almost always have differentvalues.

Chain Coding: This data structure involves the tracing of the shape of each object, andthus recording the outer pixels. To achieve chain coding it is necessary to choose oneobject and a start point for the coding within the matrix. It is necessary to trace theshape of the object, recording the direction in which one is moving as well as thenumber of pixels traced until one changes direction.

Quadtree Encoding: The basis of the quadtree data structure is to recursively dividesthe raster into quarters. Once the raster has been divided into 4, each quarter isexamined to determine whether all pixels have the same value. If all pixels do have thesame value, then this quarter is not sub-divided any further. Where this is not the case,then each of the non-homogenous quarters is divided into 4. Each of these quarters areexamined, and that are not homogenous are subdivided again. This recursivesubdivision is complete once every 'quarter' is homogenous.



20/65


6.2.2 Vector Data Structures

For each data model, different data structures are used to determine the precisemethod by which information is stored in the G1S. Two such types are unlinked and

topology within the vector data model. The latter of the two is recognized as one of themore complex data structures.Unlinked:Vector data stored in an unlinked or spaghetti format can be described as acollection of lines and points with no real connection. The features are stored as stringsof coordinates with no particular relationship to each other. A number of G1S software(for example ARCINFO) adopts unlinked data structures for their data captureprocesses. This allows one to digitize data without defining any relationships. These canbe established later (for example using the command "clean" in ARCINFO) and createtopology. Consequently, the term unlinked or spaghetti data is also associated with rawdigital data. It may be described as non-intelligent because no relationships orconnections have been established. In terms of graphic presentation, unlinked data isacceptable; however, it is not possible to interrogate the data.Topology: In most G1S software, topology is defined in terms of connectivity,adjacency and contiguity or containment. Each of these categories records a particularaspect of the data.Connectivity - Describes the linking of points or polygons to each otherAdjacency - Summarizes the sharing of a common boundary of two regions orPolygonsContiguity - Is a term that describes the touching of spatial entities (usually polygons).The topological data structure relies on each line or arc that forms part of a boundary toan area or polygon having certain information - the point or node at which it starts andthe point or node at which it finishes. This information gives direction to the line and thusthe area to the left and the area to the right can also be defined.

Advantages:

No ambiguity, easy to analysis mathematically, easy to write computerprogrammes to handle the data Tremendous output of data available and covered, e.g. Satellite imagery Create good Digital Elevation Model (DEM)/ Digital Terrain Model (DTM).

Disadvantages:

Location of feature within a single pixel is not accurate Majority rules idea in an individual pixel Large file size, store grids with no data

Some types of analysis not suitable to raster data format e.g. Networkanalysis, discrete data.

6.3 Data Models using the Third Dimension

There are two data models in GIS using the third dimension; namely the digital elevationmodel (DEM-the term digital terrain model [DTM] is synonymous with DEM) and thetriangulated irregular network model (TIN). Both were designed to simulate real world



21/65


entities, for example, valleys, peaks, slopes and stream channels. Most surfaces can berepresented in both models. A DEM is a digital representation of a continuous variableover a two dimensional surface, defined by a regular or array of z values, referenced toa common datum. DEMS are typically used to represent terrain relief. Digital elevationdata normally represents continuous surfaces, such as surface terrain, thoughdiscontinuous elevation data exists. By contrast, a TIN represents a surface derivedfrom irregularly spaced sample points and so-called break line features (relate tosmoothness and continuity). Every sample point has an x, y co-ordinate and a surfaceor z value. These points are connected to edges to form a set of non-overlappingtriangles that can be used to model the surface.

6.3.1 Data Quality and Inputs

There are four ways to determine accuracy:

1) Positional Accuracy: How close does the location = true position in reality. 1:50 000scale digitizing error of 5 meters = good.

2) Attribute Accuracy: Closeness of attribute data (stored characteristics) to their truevalue e.g. Elevation affected by level of precision. Completeness = degree to whichthe data exhausts the universe of possibilities - has everything been included?Affected by rules of selection (what is chosen) and generalization (simplification ofclasses). Course scale = large generalization

3) Lineage: What processing the data has undergone in the computer and the Need tokeep track of the source (original) data = link to metadata

4) Digitizing: Lines defined by points/co-ordinates. Distance between points = shape ofthe line. Digitizing/ tracing are a large source of error. Physiological (twitching) andpsychological (omission of lines, duplication, positional accuracy due tofatigue/boredom). High resolution = small pixel size, good detail, time/costimplications, therefore determine resolution before data capture, otherwise get poorresults. Output resolution cannot be better than input resolution and 'Ecologicalfallacy'. Issue of 'map currency' and problems related to it.



22/65


23/65


that particular; point in time. This also includes the various data sources that wereconsulted to obtain, the present composition of the dataset.

Accuracy: This relates to how "true" the data is to the real world. This also relates totextual data. Spatial data accuracy can be a direct result of both the collection andcapture processes. Consideration should also be given to: how well does a GISrepresent the real world and how well do GIS analyses compute the accurate values?"

Spatial data that is captured can only be as good as the source maps from which it isbeing captured. Another significant issue to consider is that of precision. This is a termused in computing that often refers to the number of decimal places or significant digitsin a measurement. Precision is not, however, the same as accuracy. It measures theexactness with which a value is expressed, whether the value is right or wrong. A largenumber of significant digits therefore, does not necessarily indicate that themeasurement is accurate. It is important to realize that a GIS may work at highprecision, mostly much higher than the accuracy of the data itself.

Positional Accuracy: This term refers to the closeness of Locational information(usually co-ordinates) to their real world position. There are a number of accuracystandards but conventionally, maps are accurate to approximately 0.5mm.This isequivalent to 25 meters on a 1:50 00 scale map. The precision of the co-ordinates,which represent the cartographic elements, will depend on the scale at which they arecaptured. There are a number of standard tests that can be carried out to measurepositional accuracy. It can be measured, for instance, by reference to original or newsurvey data or more detailed cartographic sources. The quality of a 1:50 000-scale mapcould be tested against a number of 1:10 000 scale maps. This type of evaluation is

undertaken statistically using cartometric testing techniques.

Attribute Accuracy: This measures the closeness of attribute values to their true value.Attribute accuracy must be analyzed in different ways depending on the nature of thedata. Two classes can be considered. In the assessment of continuous attributes (suchas surfaces), accuracy can be expressed as a measurement error. For example,elevation accuracy may be recorded as "plus or minus 1 meter". The other is acategorical attribute, such as those used for the classification of polygons. Rules needto exist for so-called non-homogenous areas. For example, if an area is 60% onecategory and 40% another category then a decision on how should it be classified mustbe made. The classification system must remain consistent.

Completeness: Completeness relates to the amount of information recorded within adataset. These can be associated with verification and classification. The amount anddistribution of features, for instance, may vary within the dataset. Furthermore, theclassification system may or may not be a good representation of the data in question.For example, need to avoid overlapping classes or systems open to misinterpretation.



24/65


Logical Consistency: This term is associated with the relationship between dataelements within a dataset. It also relates to the maintenance of those relations. Forexample, does a one always digitize from the center of a road or along which riverbankdoes one measure? It is difficult to measure logical consistency and perhaps, therefore,it is best to consider such aspects of the data before they are entered into the GIS.

7.2 Data Analysis

7.2.1 Query Functions:

Displaying of inputted data, view of attributes linked to a particular feature Both spatial and attribute Spatial = circle, rectangle, polygon, line & point search Attribute = relies on non-geometric datasets linked to spatial data Numeric, textual or alphanumeric matching

Query language available

The basic GIS functions are those that display all or a selection of the objects underinvestigation. This may include graphic elements and selected attributes. Such analysisoperations are known as query functions. They can be divided into two categories-spatial and attribute.

All spatial query functions are based around the GIS features (points, lines, polygons,surfaces) and there are five geometric functions to investigate - circle, rectangle,polygon, and line and point search. These can be developed into a host of morecomplicated queries.

An attribute query relies on the non-geometric datasets linked to spatial data. Thequeries are generally broken down into a series of elements that initially identify the fieldof interest and then the search can be carried out using specific characteristics. Resultsmay be in the form of a table, or displayed as spatial data. Criteria for attribute queriesmay be based on numeric or textual searches or alphanumeric matching. In addition,the database software associated with the GIS may have a query language available.These are designed to enable the user to tailor queries in a particular manner.

7.2.2 Logical Operations:

Use either algebra or logical operations

Searches can be further developed to employ a set of algebra or logical operations.This are-

> greater than < less than = equal to

Alternatively, a combination of the above can be used. All GIS packages should be ableto perform these operations, plus perform Arithmetic (+; -; *) and Boolean operations(and, or, not, xor).



25/65


7.2.3 Reclassification Functions:

Used to dissolve or merge information, or simple classify Number of options included changing attribute values without altering the geo-

metric properties of the spatial data or removal of boundaries between mergedregions Issue of what happens to attribute data - rules embedded/available in the chosensoftware

Within GIS data, one is often required to dissolve and merge information, in other wordsreclassify. In addition, it may be necessary for simple classification operations to beaccomplished. GIS software should also be able to carry out most of these functions.How the classification or reclassification is achieved, however, varies according to thetype of software being used. Classification operations may involve changing attributevalues without altering the geometric properties of the spatial data. Likewisereclassification functions may involve the removal of boundaries between merged

regions, when data has for instance, been aggregated to a higher spatial unit than thatat which it was collected. There are many choices for this type of analysis and similarlythere are often a number of options with system commands.

7.2.4 Overlay Operations:

Refers to the ability to bring together 2 or more datasets. There are about 3 groups,namely:

Point in polygon - this type of operation overlays point objects on areas andproduces in effect and is contained within relationship.

Line on polygon - this type of operation allows line objects to be overlaid on area,or polygon, objects. This also produces and is contained within relationship.

Polygon on polygon - this allows one polygon theme/coverage/layer to beoverlaid or combined with another polygon coverage. This relates to adjacency.

7.2.5 Buffer Operations:

Generating a region of a specified radius around a feature of interest Created search polygon is temporary - has a query function Both raster and vector data models

The buffer operation involves generating a region of a specified radius around a featureof interest. This type of function has been developed for query purposes. Any search

polygon that is generated is temporary and only exists to allow a specific spatial queryto be performed. Both raster and vector systems can be used to undertake this type ofanalysis.

7.2.6 Network Operation:

Connected lines representing a set of features, for example road or rail network Better with vector - have lines and nodes Each segment has attributes, at joins (nodes) also can have attributes



26/65


Can impose constraints on segments, for example speed limits, water pressure.

A network is generally made up of connected lines that link to represent a set offeatures. Common examples are a road or rail network. A CIS package may or may notinclude network analysis capability, and if it does, it maybe generic or designed forspecific operations. The vector model is the most suitable for network analysis becauseof the explicit representation of lines (roads) and nodes (road junctions). Each segmentor line can have attributes such as the dimension of a pipe, name of road etc. Wherelines connect, the nodes may also have attributes such as valve details or type of

junction recorded. Furthermore, constraints can be placed on the segments, forexample speed limits, directionality and water pressure.

7.2.7 Analysis of Surface:

Analytical Hill-shading - effect of light source Hydrological Modeling - flow of water through a catchments / across asurface Soil Erosion Modeling - function of slope, hydrology, soil, geology,vegetation etc. ideal for modeling scenario and creating "what-if" typemodels, used for hazard assessment and prediction. Visibility Analysis - line of sight of a particular object or feature

All the analysis considered so far involves two-dimensional data. There are also thirddimensional analyses available. Once a surface is represented in digital format, it isstraightforward procedure to derive slope or aspect, whether this is for a triangle in aTIN, or a pixel in a DEM. Once altitude, slope and aspect has been calculated for everypoint across the terrain model, it is then possible to perform a range of further analyses,such as: -

Analytical hill shading - the effect of a light source on the surface is simulated,thus generating shadows were appropriate. Can also alter or move the light sourcethus being able to simulate time period changes (seasons, diurnal) Hydrological modeling represents the flow of water across a surface, can beused to define theoretical water catchments and stream paths. It is possible tocalculate the flow path of water that enters the system (e.g. rainfall) and thus can beemployed for flood hazard modeling and pollution prediction and control Soil erosion modeling - often combined with hydrological modeling. Soil erosion

is partly a function of slope angle, soil type, underlying geology, vegetation andhydrology. All these variables can be easily represented and modeled in a GIS Visibility analysis - it is possible to calculate the visibility (line of sight) of aparticular object or feature on the landscape, such as proposed electrical lines ortelecommunication towers.

7.2.8 Temporal Functions:

All maps out-of-date as soon as produced



27/65


Many GIS variables very dynamic and changing constantly with time Need to try and incorporate the 4-D of time Most software packages not ale to, thus use a series of data items represented

by on-screen animation as a seriesAny map, aerial photography or satellite image is out of date the moment it is produced.Many of the variables that are considered in a CIS are dynamic and continually in astate of flux - for example seasonal change of vegetation patterns and land usechanges through time. It is therefore useful for some applications to be able to representtime. Time represents a fourth dimension to the usual data, so that we may have ourdata structure x, y, z, attribute and time information. Most GIS software does not handletime very well. The simplest way to handle time is to represent using a series of dataitems. This is represented on-screen as a series of animations.

7.2.9 Geo-Coding:

Addresses, street names and house numbers Census and market segmentation data Determines best routes

Distances: Manhattan versus Euclidean & Pythagoras' theorem

7.2.10 Dissolve and Merge:

These are the processes of merging two or more adjacent areas or polygons andremoving the boundary lines between them. Dissolving and merging operations are auseful facility for those who wish to aggregate their data based on particular criteria. It isimportant to note, however, that it is not just the spatial features that are adjusted. The

associated attribute data is also merged or dissolved according to specific criteria.



28/65


Session 6: Knowing Examples of Applications of GIS inConservation

8.1 Applications of GIS in Conservation

8.1.1 Management Practices in Sabi Sands Complex

Raster-based system. Identification of regions of over-exploitation and grazing surrounding artificially

created water-points, Created to promote tourism, major impact on ability ofvegetation to be sustainable, regenerate and resultant bush-encroachment, Thiscreates a classical spiral of degradation as then needed to burn/cut to removeundesirable species.

GIS implemented as a management tool.

8.1.2 Buffalo River Catchment Study, Eastern Cape.

Vector-based. Physical and anthropological variables considered. Use of catchment principles and processes. Identification of point and non-point pollution sources. Included proposal for setting up a conservancy area -linked to genetic study. Three phases; field work, data input, correction (editing and cleaning) of data,

analysis/decisions. Catchment management use of GIS.

8.1.3 Bell River, Rhodes.

Geomorphological study requiring time series analysis. Use of aerial photography for different time periods. Supplemented with 'other' photographic evidence. Determine past river channel morphology and 'migration' of bank. Bank stability - recent introduction of alien species to stabilize banks, effect on loss

on natural riverine/riparain vegetation, agric. land and natural rates oferosion/deposition

8.1.4 Motao Forest Zimbabwe.

Incorporation of GPS, differential GPS and correction requirements Inventory and management tool, used as a good decision-support system Use in disease control, inclusion of precision agriculture

8.1.5 Biogeography Hogsberg Mountain.

Inventory, prediction, modeling of alien plant species migration Acacia longifolia, A.mearnsii, Opuntia ficus-indica & Solanum sisymbrifolium



29/65


30/65


31/65


Figure 8: Three wave (CCRS/CRT, 2007)

The wavelength is the length of one wave cycle, which can be measured as thedistance between successive wave crests. Wavelength is usually represented by theGreek letter lambda (l). Wavelength is measured in meters (m) or some factor of meterssuch as nanometers (nm, 10-9 meters), micrometers (mm, 10-6 metres) (mm, 10-6

metres) or centimeters (cm, 10-2 metres). Frequency refers to the number of cycles of awave passing a fixed point per unit of time. Frequency is normally measured in hertz(Hz), equivalent to one cycle per second, and various multiples of hertz. Wavelengthand frequency are related by the following formula:

Figure 9: The Electromagnetic Spectrum (CCRS/CRT, 2007)

Therefore, the two are inversely related to each other. The dynamics here are that,shorter the wavelength, the higher the frequency and the longer the wavelength, thelower the frequency. Understanding the characteristics of electromagnetic radiation interms of their wavelength and frequency is crucial to understanding the information to



32/65


be extracted from remote sensing data. The following subsection will be examining theway in which we categorize electromagnetic radiation for just that purpose.

9.3 The Electromagnetic Spectrum

The electromagnetic spectrum ranges from the shorter wavelengths (includinggamma and x-rays) to the longer wavelengths (including microwaves and broadcastradio waves). There are several regions of the electromagnetic spectrum, which areuseful for remote sensing.

For most purposes, the ultraviolet or UV portion of the spectrum has the shortestwavelengths, which are practical for remote sensing. This radiation is just beyond theviolet portion of the visible wavelengths, hence its name. Some Earth surface materials,primarily rocks and minerals, fluoresce or emit visible light when illuminated by UVradiation.

The light, which our eyes - our remote sensors - can detect, is part of the visiblespectrum. It is important to recognize how small the visible portion is relative to the restof the spectrum. There is a lot of radiation around us, which is "invisible" to our eyes,but can be detected by other remote sensing instruments and used to our advantage.The visible wavelengths cover a range from approximately 0.4 to 0.7 m . The longestvisible wavelength is red and the shortest is violet. Common wavelengths of what weperceive as particular colours from the visible portion of the spectrum are listed below. Itis important to note that this is the only portion of the spectrum we can associate withthe concept ofcolours.

Violet: 0.4 - 0.446 m

Blue: 0.446 - 0.500 m Green: 0.500 - 0.578 m Yellow: 0.578 - 0.592 m Orange: 0.592 - 0.620 m Red: 0.620 - 0.7 m

Blue, green, and red are the primary colours or wavelengths of the visible spectrum.They are defined as such because no single primary colour can be created from theother two, but all other colours can be formed by combining blue, green, and red invarious proportions. Although we see sunlight as a uniform or homogeneous colour, it is

actually composed of various wavelengths of radiation in primarily the ultraviolet, visibleand infrared portions of the spectrum. The visible portion of this radiation can be shownin its component colours when sunlight is passed through a prism, which bends thelight in differing amounts according to wavelength.

The next portion of the spectrum of interest is the infrared (IR) region which covers thewavelength range from approximately 0.7 m to 100 m - more than 100 times as wideas the visible portion! The infrared region can be divided into two categories based on



33/65


their radiation properties - the reflected IR, and the emitted orthermal IR. Radiation inthe reflected IR region is used for remote sensing purposes in ways very similar toradiation in the visible portion. The reflected IR covers wavelengths from approximately0.7 m to 3.0 m. The thermal IR region is quite different from the visible and reflectedIR portions, as this energy is essentially the radiation that is emitted from the Earth'ssurface in the form of heat. The thermal IR covers wavelengths from approximately 3.0m to 100 m.

The portion of the spectrum of more recent interest to remote sensing is the microwaveregion from about 1 mm to 1 m. This covers the longest wavelengths used for remotesensing. The shorter wavelengths have properties similar to the thermal infrared regionwhile the longer wavelengths approach the wavelengths used for radio broadcasts.

9.4 Radiation and the Atmosphere

Figure 10: Scattering (CCRS/CRT, 2007)

Scattering occurs when particles or large gas molecules present in the atmosphere

interact with and cause the electromagnetic radiation to be redirected from its originalpath. How much scattering takes place depends on several factors including thewavelength of the radiation, the abundance of particles or gases, and the distance theradiation travels through the atmosphere. There are three (3) types of scattering whichtake place.

Figure 11: Rayleigh scattering (CCRS/CRT, 2007)

Rayleigh Scattering: occurs when particles are very small compared to the wavelengthof the radiation. These could be particles such as small specks of dust or nitrogen andoxygen molecules. Rayleigh scattering causes shorter wavelengths of energy to bescattered much more than longer wavelengths. Rayleigh scattering is the dominantscattering mechanism in the upper atmosphere. The fact that the sky appears "blue"during the day is because of this phenomenon. As sunlight passes through the



34/65


atmosphere, the shorter wavelengths (i.e. blue) of the visible spectrum are scatteredmore than the other (longer) visible wavelengths. At sunrise and sunset the light has to

travel farther through the atmosphere than at midday and thescattering of the shorter wavelengths is more complete; this leaves agreater proportion of the longer wavelengths to penetrate theatmosphere.

Figure 12: Mie scattering (CCRS/CRT, 2007)

Mie Scattering: occurs when the particles are just about the same size as thewavelength of the radiation. Dust, pollen, smoke and water vapour are common causesof Mie scattering, which tends to affect longer wavelengths than those affected byRayleigh scattering. Mie scattering occurs mostly in the lower portions of theatmosphere where larger particles are more abundant, and dominates when cloud

conditions are overcast.

Figure 13: Nonselective scattering (CCRS/CRT, 2007)

The final scattering mechanism of importance is called nonselective scattering. Thisoccurs when the particles are much larger than the wavelength of the radiation. Waterdroplets and large dust particles can cause this type of scattering. Nonselectivescattering gets its name from the fact that all wavelengths are scattered about equally.This type of scattering causes fog and clouds to appear white to our eyes because blue,green, and red light are all scattered in approximately equal quantities (blue+green+redlight - white light).

Figure 14: Absorption (CCRS/CRT, 2007)



35/65


36/65


Absorption (A) occurs when radiation (energy) is absorbed into the target whiletransmission (T) occurs when radiation passes through a target. Reflection (R) occurswhen radiation "bounces" off the target and is redirected. In remote sensing, we aremost interested in measuring the radiation reflected from targets. We refer to two typesof reflection, which represent the two extreme ends of the way in which energy isreflected from a target: specular reflection and diffuse reflection.

Figure 15: Target Interactions (ccrs, 2007)

When a surface is smooth we get specularor mirror-like reflection where all (or almostall) of the energy is directed away from the surface in a single direction. Diffusereflection occurs when the surface is rough and the energy is reflected almost uniformlyin all directions. Most earth surface features lie somewhere between perfectly specularor perfectly diffuse reflectors. Whether a particular target reflects specularly or diffusely,or somewhere in between, depends on the surface roughness of the feature incomparison to the wavelength of the incoming radiation. If the wavelengths are muchsmaller than the surface variations or the particle sizes that make up the surface, diffusereflection will dominate. For example, fine-grained sand would appear fairly smooth tolong wavelength microwaves but would appear quite rough to the visible wavelengths.

Figure 16: Interactions with vegetation (CCRS, 2007)

Let's take a look at a couple of examples of targets at the Earth's surface and how

energy at the visible and infrared wavelengths interacts with them.Leaves: A chemical compound in leaves called chlorophyll strongly absorbs radiation inthe red and blue wavelengths but reflects green wavelengths. Leaves appear "greenest"to us in the summer, when chlorophyll content is at its maximum. In autumn, there isless chlorophyll in the leaves, so there is less absorption and proportionately morereflection of the red wavelengths, making the leaves appear red or yellow (yellow is acombination of red and green wavelengths). The internal structure of healthy leaves actas excellent diffuse reflectors of near-infrared wavelengths. If our eyes were sensitive to



37/65


near infrared, trees would appear extremely bright to us at these wavelengths. In fact,measuring and monitoring the near-IR reflectance is one way that scientists candetermine how healthy (or unhealthy) vegetation may be.

Figure 17: Interactions with vegetation (CCRS, 2007)

Water: Longer wavelength visible and near infrared radiation is absorbed more by waterthan shorter visible wavelengths. Thus, water typically looks blue or blue-green due tostronger reflectance at these shorter wavelengths, and darker if viewed at red or nearinfrared wavelengths. If there is suspended sediment present in the upper layers of thewater body, then this will allow better reflectivity and a brighter appearance of the water.The apparent colour of the water will show a slight shift to longer wavelengths.Suspended sediment (S) can be easily confused with shallow (but clear) water, sincethese two phenomena appear very similar. Chlorophyll in algae absorbs more of theblue wavelengths and reflects the green, making the water appear greener in colourwhen algae are present. The topography of the water surface (rough, smooth, floatingmaterials, etc.) can also lead to complications for water-related interpretation due to

potential problems of specular reflection and other influences on colour and brightness.

Figure 18: Spectral Curve sowing Interactions with vegetation and

water (CCRS/CRT, 2007

We can see from these examples that, depending on the complex make-up of the targetthat is being looked at, and the wavelengths of radiation involved, we can observe verydifferent responses to the mechanisms of absorption, transmission, and reflection. Bymeasuring the energy that is reflected (or emitted) by targets on the Earth's surface overa variety of different wavelengths, we can build up a spectral response for that object.



38/65


By comparing the response patterns of different features, we may be able to distinguishbetween them, where we might not be able to, if we only compared them at onewavelength. For example, water and vegetation may reflect somewhat similarly in thevisible wavelengths but are almost always separable in the infrared. Spectral responsecan be quite variable, even for the same target type, and can also vary with time (e.g."green-ness" of leaves) and location. Knowing where to "look" spectrally andunderstanding the factors, which influence the spectral response of the features ofinterest, are critical to correctly interpreting the interaction of electromagnetic radiationwith the surface.

9.6 Passive vs. Active Sensing

Figure 19: Passive Sensors (CCRS/CRT, 2007)

So far, throughout this section, we have made various references to the sun as a sourceof energy or radiation. The sun provides a very convenient source of energy for remotesensing. The sun's energy is either reflected, as it is for visible wavelengths, orabsorbed and then re-emitted, as it is for thermal infrared wavelengths. Remotesensing systems, which measure energy that is naturally available, are called passive

sensors. Passive sensors can only be used to detect energy when the naturallyoccurring energy is available. For all reflected energy, this can only take place duringthe time when the sun is illuminating the Earth. There is no reflected energy availablefrom the sun at night. Energy that is naturally emitted (such as thermal infrared) can bedetected day or night, as long as the amount of energy is large enough to be recorded.

Figure 20: Active Sensors (CCRS/CRT, 2007)

Active sensors, on the other hand, provide their own energy source for illumination.The sensor emits radiation, which is directed toward the target to be investigated. Theradiation reflected from that target is detected and measured by the sensor. Advantages



39/65


for active sensors include the ability to obtain measurements anytime, regardless of thetime of day or season. Active sensors can be used for examining wavelengths that arenot sufficiently provided by the sun, such as microwaves, or to better control the way atarget is illuminated. However, active systems require the generation of a fairly largeamount of energy to adequately illuminate targets. Some examples of active sensorsare a laser fluorosensor and synthetic aperture radar (SAR).



40/65


41/65


convey the same information (although some detail may be lost when converting backand forth).

In previous sections, we described the visible portion of the spectrum and the conceptof colours. We see colour because our eyes detect the entire visible range ofwavelengths and our brains process the information into separate colours. Can youimagine what the world would look like if we could only see very narrow ranges ofwavelengths or colours? That is how many sensors work. The information from anarrow wavelength range is gathered and stored in a channel, also sometimes referredto as a band. We can combine and display channels of information digitally using thethree primary colours (blue, green, and red). The data from each channel is representedas one of the primary colours and, depending on the relative brightness (i.e. the digitalvalue) of each pixel in each channel, the primary colours combine in differentproportions to represent different colours.

When we use this method to display a single channel or range of wavelengths, we areactually displaying that channel through all three primary colours. Because thebrightness level of each pixel is the same for each primary colour, they combine to forma black and white image, showing various shades of gray from black to white. Whenwe display more than one channel each as a different primary colour, then thebrightness levels may be different for each channel/primary colour combination and theywill combine to form a colour image.

10.2 On the Ground, In the Air, In Space

Figure 22: Platforms (CCRS/CRT, 2007)

In an earlier section, we learned some of the fundamental concepts required tounderstand the process that encompasses remote sensing. We covered in some detail

the first three components of this process: the energy source, interaction of energy withthe atmosphere, and interaction of energy with the surface. We touched briefly on thefourth component (recording of energy by the sensor), when we discussed passivevs. active sensors and characteristics of images. In this section, we will take a closerlook at this component of the remote sensing process by examining in greater detail, thecharacteristics of remote sensing platforms and sensors and the data they collect. Wewill also touch briefly, on how that data is processed once the sensor has recorded it. Inorder for a sensor to collect and record energy reflected or emitted from a target or



42/65


43/65


Session 9: Understanding Satellite Characteristics-Orbits

and Swaths

Figure 23: Satellite orbits (CCRS/CRT, 2007)

11.1 Geostationary Orbits

We learned in the previous section that remote sensing instruments can be placed on avariety of platforms to view and image targets. Although ground-based and aircraftplatforms may be used, satellites provide a great deal of the remote sensing imagerycommonly used today. Satellites have several unique characteristics, which make themparticularly useful for remote sensing of the Earth's surface.

The path followed by a satellite is referred to as its orbit. Satellite orbits are matched tothe capability and objective of the sensor(s) they carry. Orbit selection can vary in termsof altitude (their height above the Earth's surface) and their orientation and rotationrelative to the Earth. Satellites at very high altitudes, which view the same portion of theEarth's surface at all times have geostationary orbits.

These geostationary satellites, at altitudes of approximately 36,000 kilometers, revolveat speeds, which match the rotation of the Earth so they seem stationary, relative to theEarth's surface. This allows the satellites to observe and collect informationcontinuously over specific areas. Weather and communications satellites commonlyhave these types of orbits. Due to their high altitude, some geostationary weathersatellites can monitor weather and cloud patterns covering an entire hemisphere of theEarth.

Many remote sensing platforms are designed to follow an orbit (basically north-south)which, in conjunction with the Earth's rotation (west-east), allows them to cover most ofthe Earth's surface over a certain period of time. These are near-polar orbits, sonamed for the inclination of the orbit relative to a line running between the North andSouth poles. Many of these satellite orbits are also sun-synchronous such that theycover each area of the world at a constant local time of day called local sun time. Atany given latitude, the position of the sun in the sky as the satellite passes overhead will



44/65


be the same within the same season. This ensures consistent illumination conditionswhen acquiring images in a specific season over successive years, or over a particulararea over a series of days. This is an important factor for monitoring changes betweenimages or for mosaicking adjacent images together, as they do not have to be correctedfor different illumination conditions.Most of the remote sensing satellite platforms today are in near polar, an orbit, whichmeans that the satellite travels northwards on one side of the Earth and then toward thesouthern pole on the second half of its orbit. These are called ascending anddescending passes, respectively. If the orbit is also sun-synchronous, the ascendingpass is most likely on the shadowed side of the Earth while the descending pass is onthe sunlit side. Sensors recording reflected solar energy only image the surface on adescending pass, when solar illumination is available. Active sensors, which providetheir own illumination, or passive sensors that record emitted (e.g. thermal) radiationcan also image the surface on ascending passes.

As a satellite revolves around the Earth, the sensor "sees" a certain portion of theEarth's surface. The area imaged on the surface, is referred to as the swath. Imagingswaths for space borne sensors generally vary between tens and hundreds ofkilometers wide. As the satellite orbits the Earth from pole to pole, its east-west positionwouldn't change if the Earth didn't rotate. However, as seen from the Earth, it seemsthat the satellite is shifting westward because the Earth is rotating (from west to east)beneath it. This apparent movement allows the satellite swath to cover a new area witheach consecutive pass. The satellite's orbit and the rotation of the Earth work togetherto allow complete coverage of the Earth's surface, after it has completed one completecycle of orbits.

If we start with any randomly selected pass in a satellite's orbit, an orbit cycle will becompleted when the satellite retraces its path, passing over the same point on theEarth's surface directly below the satellite (called the nadirpoint) for a second time. Theexact length of time of the orbital cycle will vary with each satellite. The interval of timerequired for the satellite to complete its orbit cycle is not the same as the "revisitperiod". Using steerable sensors, a satellite-borne instrument can view an area (off-nadir) before and after the orbit passes over a target, thus making the 'revisit' time lessthan the orbit cycle time. The revisit period is an important consideration for a number ofmonitoring applications, especially when frequent imaging is required (for example, tomonitor the spread of an oil spill, or the extent of flooding). In near-polar orbits, areas at

high latitudes will be imaged more frequently than the equatorial zone due to theincreasing overlap in adjacent swaths as the orbit paths come closer together nearthe poles.

11.2 Spatial Resolution, Pixel Size, and Scale

For some remote sensing instruments, the distance between the target being imagedand the platform, plays a large role in determining the detail of information obtained and



45/65


the total area imaged by the sensor. Sensors onboard platforms far away from theirtargets, typically view a larger area, but cannot provide great detail. Compare what anastronaut onboard the space shuttle sees of the Earth to what you can see from anairplane. The astronaut might see your whole province or country in one glance, butcouldn't distinguish individual houses. Flying over a city or town, you would be able tosee individual buildings and cars, but you would be viewing a much smaller area thanthe astronaut would. There is a similar difference between satellite images and airphotos.

Figure 24: Spatial Resolution, Pixel Size, and Scale (CCRS/CRT, 2007)

The detail discernible in an image is dependent on the spatial resolution of the sensorand refers to the size of the smallest possible feature that can be detected. Spatialresolution of passive sensors (we will look at the special case of active microwavesensors later) depends primarily on their Instantaneous Field of View (IFOV). TheIFOV is the angular cone of visibility of the sensor (A) and determines the area on theEarth's surface, which is "seen" from a given altitude at one particular moment in time(B). The size of the area viewed is determined by multiplying the IFOV by the distancefrom the ground to the sensor (C). This area on the ground is called the resolution celland determines a sensor's maximum spatial resolution. For a homogeneous feature to

be detected, its size generally has to be equal to or larger than the resolution cell. If thefeature is smaller than this, it may not be detectable, as the average brightness of allfeatures in that resolution cell will be recorded. However, smaller features maysometimes be detectable if their reflectance dominates within a particular resolution cellallowing sub-pixel or resolution cell detection.

As we mentioned in the previous sections, most remote sensing images are composedof a matrix of picture elements, or pixels, which are the smallest units of an image.Image pixels are normally square and represent a certain area on an image. It isimportant to distinguish between pixel size and spatial resolution - they are notinterchangeable. If a sensor has a spatial resolution of 20 metres and an image from

that sensor is displayed at full resolution, each pixel represents an area of 20m x 20mon the ground. In this case, the pixel size and resolution are the same. However, it ispossible to display an image with a pixel size different from the resolution. Many postersof satellite images of the Earth have their pixels averaged to represent larger areas,although the original spatial resolution of the sensor that collected the imagery remainsthe same.



46/65


Images where only large features are visible are said to have coarse or lowresolution. In fine or high-resolution images, small objects can be detected. Militarysensors for example, are designed to view as much detail as possible, and thereforehave very fine resolution. Commercial satellites provide imagery with resolutions varyingfrom a few metres to several kilometers. Generally speaking, the finer the resolution, theless total ground area can be seen. The ratio of distance on an image or map, to actualground distance is referred to as scale. If you had a map with a scale of 1:100,000, anobject of 1cm length on the map would actually be an object 100,00

Training Manual - GIS and Remote Sensing

Documents

Transcript of Training Manual - GIS and Remote Sensing