> }|7 bjbjUU 7|7|~8l84e,ZZ4+++++++$!. A0
,
,7,*++H0*+NmF*+5,0e,*1MX1+[Slide Title..]
Introduction to Geographic Information Systems Distance Learning Version
Lecture 2 Script
Introduction: Hello and welcome to Lecture two of the distance learning version of Introduction to Geographic Information Systems. As I mentioned in lecture one, this is an internet version, otherwise known as the eLearning Version, of a traditional course I taught at the University of Hartford. By now, you should have visited the course web site several times and have become familiar with the course policies and procedures. If you have any questions, you can ask me during the web chat session, or e-mail them to me at any time. The session today will be fairly intense (fancy words for long) so try to stay alert (more fancy words for try not to snore too loudly). (Next Slide Please)
[Slide Intro...]
Today, we are going to explore the world of spatial data and how exactly do we measure that spatial data and then how do we describe it to the computer. As well see, we are going to have to accept some inaccuracies in the description method due to certain approximations we are going to be forced to make.
In Lecture one, we were introduced to the two spatial data structures-rasters and vectors. Today, well delve more deeply into the actual construction of these two ways of describing spatial data.
Then, what goes into actually building the spatial data model will be discussed. As well see, the process involves much of the same thought processes as one would experience if you were to draw a paper map from scratch. In Lecture one, we also looked at the three spatial data entities of a vector display namely, point, lines and polygons. Now, two new entities will be introduced, namely the surface, and the network both have direct applications to Civil Engineering.
And finally, the non spatial data relating to the features that is the attributes, will be looked at with respect to what kinds of numbers are used to quantify an attribute. This has ramifications as to what kind of mathematical processes can be performed on the data. OK well see if the definition of spatial data has been retained in your little gray cells, here it is again. (Next Slide Please)
[Slide Definition]
Our definition of Spatial Data is related to the information about the geographic position of features. We discussed last time what a feature is. The geographic position mentioned is an x,y coordinate pair in two dimensional space. In fact, the spatial data are the x,y coordinates. This definition leaves a lot open for discussion. We know that the earth is a three dimensional body a sphere, for the most part, and that maps are two dimensional planes. Well talk about the process of forcing a three dimensional shape in to 2D. Also what x,y pair should we choose? From analytic geometry, there are an infinite set of coordinates to describe the position of a point. Well talk about these seemingly insurmountable obstacles in a few minutes. (Next Slide Please)
[Slide GIS Model...]
Our job, eventually, will be to build a GIS model. A GIS model is a simplified model of the real world the emphasis is on the word simplified since building a model of the real world in its entirety, today, is impossible. For example, if we were building a GIS model of the campus whose theme was the mapping of all underground utilities, details of planting beds and the types of plants located in them would seem irrelevant and would be excluded from the GIS model of the simplified view. Once the necessary features are determined, they have to be placed in a structure so that the computer can recognize them. To tie this all together, we need the spatial data that describes our desired features. This lecture will focus on the spatial data and their structure that allows them to input into the computer. So, first, lets look at the characteristics of spatial data. (Next Slide Please)
[Slide Nature of Spatial Data...]
We can define data as observations we make from the real world. These observations are raw numbers. Without context they are of little value.. Context in this situation means information about how the data was collected, scale of the measurements, units, special conditions, etc. When context is added to data, information is formed information that is useable to others who may not have been present when the data was collected. In GIS terms, this context has a special name Metadata. Since you will sometime in the future be using or incorporating data that you collected or have direct knowledge of (called primary data) and data from many other sources (called secondary data) in a GIS model, metadata becomes vitally important. This will become apparent when work on the term project begins. In that case you will be using secondary data exclusively.
With this in mind, spatial data can be separated into three types:
Temporal data tells when the data was collected, otherwise known as Metadata
Thematic data Describes the real world feature in other words the attributes, and
Spatial data again the short version the x,y coordinates that locate a feature. Well look in more detail now about the forms in which spatial data occur. To do that, we start first with the analogy of drawing a paper map, for the same processes carry over to GIS operations and model building. (Next Slide Please)
[Slide Traditional Maps...]
Traditional map making procedures influence how we think about spatial data. We will look at each one of these map components in detail. First, the purpose of the map needs to be established. A hand drawn map giving driving directions to a friend will look totally different than a map produced for a presentation in front of town officials. For the purpose of this course, we will be looking at maps that fall into the second category, i.e. professional maps. Next we need to select the scale of the map. Scale is defined as the ratio of the distance on the map to the distance on the ground. As we discussed, a map or GIS model is a simplified view of the real world so we have to select which features we need to include in our map. Then, we need to decide how we want to project the real world spherical model on to a two dimensional map then select a way to measure the x,y coordinates of points on our map. And finally, we need to figure out what annotation we want to put on the map, namely the scale, North arrow, any legend, notes, etc. Lets look at each one of these characteristics in more detail first scale. (Next Slide Please)
[Slide Scale...]
A more literal definition of scale would be that scale indicates how much smaller than reality a map is. A more mathematical definition is that scale is the ratio of the distance on a map to the distance on the ground. For instance a scale of 1 to 5000 can be interpreted as 1 cm on the map represents 5000 cm on the ground or 50 m on the ground. Scale can be shown in graphical terms as shown on the slide. This is the form most utilized on a map. We also use words to indicate the scale of a map. (Next Slide Please)
[Slide Scale, cont...]
If our map covers a large area, then the ratio will be large. Note that in scale ratios, the distance on the map is always given as unity. A map drawn to a ratio such as one to two million is considered small scale map. This seems counter intuitive since two million is a large number but remember, its the ratio of 1 over two million that gives the scale and this is a small number. Conversely, a one to twenty five thousand ratio is considered a large scale map. And will cover smaller areas. Scale is important because it will determine how our features will be drawn. (Next Slide Please)
[Slide Scale Related...]
Every map has some inherent generalization or simplification. This is in keeping with our idea of a map as a simplification of the real world. Scale plays a role in this simplification, as the level of detail shown on a map is directly related to the scale. Here we have successive maps of the Mississippi River. At the top, the scale is very small at 1 to 20 million. At this scale the river is a line. All Details including curves in the river have been smoothed out because they cant be seen at this scale. As we move down with the maps increasing in scale in the third map at 1: 250,000, we can see the meanderings and the banks of the river. In the bottom map at 1:50,000, details along the bank are seen. This is a very vivid example of the power of scale. OK lets move on to our next problem which is how to convert features on the surface of a sphere to a flat plane of two dimensions.(Next Slide Please)
[Slide Map Projections...]
Projection is the process of transferring spatial data from the spherical earth to a flat surface. As it turns out, there are many ways to do this, so that there are literally hundreds of projections but the important point here is all projections have some errors and distortions. Well, how do you actually form a projection? The best way to understand the process is to think about a wire model of the spherical earth with a light bulb source suspended at its center. (Next Slide Please)
[Slide Projection Techniques 1...]
Here we see such a set up. Around the outside of the sphere, we place a two dimensional cylindrical surface. If we unwrap the cylinder, we have a cylindrical projection. You will notice that the lines of latitude and longitude meet at right angles just as they do on the sphere but also note that the distances near the poles is greatly distorted. (Next Slide Please)
[Slide Projection Techniques 2 ...]
In this projection technique, a flat surface is placed next to the sphere. Note in this projection, we can see only a portion of the surface. Actually, this slide is in error, since the view generated has a pole in its center i.e. the flat surface was placed on the top or bottom of the earth, not facing the equator as shown. There will be distortion at the top and bottom edges of this Azimuthal Projection. Distortion will be much less at locations near the point of tangency of the plane and the sphere. (Next Slide Please)
[Slide Projection Techniques 3 ]
The last of these geometric projections we are going to look at is the conic projection. In this projection, a cone is placed over the sphere, and the grid is projected onto the cone. Distortion will be great at the top and bottom edges, and the shape of an area will be distorted. Distortion will be much less in the vicinity of the line of tangency between the cone and the sphere. I did mention that there were literally hundreds of projections. The next slide shows an interesting one the Sinusoidal Projection.
(Next Slide Please)
[Slide Sinusoidal Projection...]
The sinusoidal projection is formed by plotting the lines of longitude as functions of the cosine of latitude, rather than having the lines of longitude vertical as in the cylindrical projection. This technique at least keeps the poles from being greatly distorted; however, notice that the lines of latitude and longitude do not meet at right angles, so directions on this map are incorrect except near the equator and the central meridian. Well, thats all the time we can spend on projections. In a beginning cartography course, projections might take up of the semester. If you are interested, there are several sites on the web that do an excellent job of explaining and displaying different projections. After a projection is decided upon, the next step is to decide on what kind of coordinate system to use to measure the x,y position on the projection. (Next Slide Please)
[Slide Spatial Referencing]
Weve been talking about latitude and longitude, which are actually a three dimensional, spherical coordinate system. We give this system a special name - the geographic coordinate system, which well look at in a minute. If we use a projection technique to produce a planar map, then we have to define an x,y, pair to describe a points position again there are many two dimensional coordinate systems that can be used, and Ill talk about them shortly.
The third category involves non coordinate systems, such as zip codes. This topic we will not cover.
(Next Slide Please)
[Slide Geographic Coordinate System ]
On this sphere, we show lines of latitude and longitude. Lines of Longitude, or meridians, are set as an interior angle measured eastward and westward from the prime meridian which is zero longitude, called the Prime Meridian, and which runs through Greenwich, England. Lines of Latitude or Parallels are set by another interior angle and run North and South from the equator.
(Next Slide Please)
[Slide Geographic Coordinate System Angles]
On this slide, we see a little more clearly the definitions of the angles of latitude and longitude, and we show lines of latitude and longitude. Latitude is set as an interior angle measured eastward and westward from the prime meridian which is zero longitude and runs through Greenwich, England. Lines of Latitude are set by another interior angle and run North and South from the equator.
(Next Slide Please)
[Slide Latitude Longitude Calculation ]
Here we see the calculation for the Latitude and Longitude of Moscow. Note they are given in the units of angles, in this case degree and minutes. We could add seconds for more accuracy, and the angles could also be given and decimal degrees for example 55 degrees 45 minutes could be given as 55.75 degrees. Note the N and E given along with the actual number they must always appear. Before we look at two dimensional coordinate systems, lets take a little Gray Cells Quiz.
(Next Slide Please)
[Slide Little Gray Cells Quiz]
OK Heres our Little Gray Cells Quiz. I may ask some of you for your answers in the chat session, so you might want to jot down some notes. Ill give you a few moments now to conjure up some answers. A hint for question three the answer is not mathematical or technical. Pause here.
Do Not Read
Large scale maps cover large areas. T or F
Why are projections needed?
Why do you think the prime meridian goes through Greenwich, England?
Well, how did you do? Well review the answers later.
(Next Slide Please).
[Slide Rectangular Coordinates]
Here we see the first of our two dimensional systems the rectangular system. This can only be produced by a cylindrical projection. Here, the unwrapped cylinder shows a natural x,y system with Latitude as the y coordinate and Longitude as the x axis. It is obvious that the cylinder in this case was tangent to the sphere at the equator parallel. This kind of projection is called a Mercator Projection. If the cylinder is tangent to the sphere at a meridian, it is called a Transverse Mercator Projection. Note the distortion in area at the poles, and that distortion is much less in the area of the equator where the sphere and the cylinder are tangent. (Next Slide Please)
[Slide Universal Transverse Mercator UTM..]
One currently popular coordinate system is the Universal Transverse Mercator or UTM system. In the UTM system, the earth is divided into 60 zones, each zone is 6 degrees in longitude wide, and runs from 800 S latitude to 840 N latitude the latter limits due to distortions at the poles. In effect, each zone is a transverse mercator projection, with the cylinder tangent to the sphere at the central meridian of the zone. However, instead of using the whole unwrapped cylinder, only the portion +/- 3 degrees from the central meridian of the zone is used for that zone, for the next zone, a new cylinder and tangent meridian is chosen. The origin of the x,y system, or Eastings and Northings, in meters, is set at a point where all coordinates in the northern hemisphere are positive. The zones are actually a little wider than 6 degrees so that the zones can be overlapped to make a map of a bigger area.
(Next Slide Please)
[Slide State Plane ]
The last coordinate system we want to look at is the State Plane Coordinate System or SPCS. Most of the state plane systems are built on a conic projection. Since they cover relatively small areas, they are fairly accurate. Again, an origin is chosen to give positive x and y coordinates which are usually given in feet. For instance, the origin of the CT SPCS is in New Jersey. This is a very good local system, but if the study area contains two or more states, then the SPCS system cannot be used since they are independent of each other. Thats why reviewing the metadata of a map is very important because thats where projection information is usually spelled out. OK at this point, lets take a seventh inning break you may stand up and stretch if you wish. Ill begin again in 30 seconds. (Next Slide Please).
[Slide Break .]
30 second break... (Next Slide Please).
[Slide Sources of Spatial Data .]
OK Were back. Weve been concentrating on maps as a source for spatial data. There are other sources which we will look at briefly. One popular source is the US Governments census data. Along with that census data, the US Bureau of Census also produces maps called TIGER files, where the acronym TIGER stands for Topologically Integrated Geographic Encoding and Referencing. This database contains legal boundaries that can be linked to the census data. From the slide we can see that this is a vector display that contains street names, block numbers, node numbers and so forth. There is, however, one word we havent seen yet topology. Well discuss this in a few minutes when we talk about vector data structure. (Next Slide Please).
[Slide Aerial Photographs]
Another source of spatial data is aerial photographs. A couple of things about aerial photos the higher the altitude of the flight, the smaller the scale of the map, and all photos have some distortion near the edges. The next slide shows an aerial photograph of the University of Hartford. (Next Slide Please).
[Slide Aerial Photograph ]
Its clear that this photo was taken a while ago before the two way road system and the magnet school were constructed. Some landmarks are noted. Though GIS software and some ground survey control points are recognizable features in the photo, we could transform this photo into a photo-map. This process is called geo-referencing and would make this photo suitable for use as a layer in a GIS. Youll see more of this technique in the lab exercises.
(Next Slide Please).
[Slide Satellite Images]
We see here a raster LANDSAT satellite image of Morrow Bay in CA. The resolution of this image is 30m x30m which means as we begin to zoom in on this image, eventually, well see blocks of color, each 30m x30 m, or 100 ft by 100 ft. At this resolution, we could not make out the details of a house or small building. What do you think the darker shapes are that are indicated by the red arrows? (PAUSE) If you identified them as shadows, you are correct! (Next Slide Please).
[Slide Surveying .]
Certainly, survey data is a natural source of spatial data and can now be directly input into a GIS. As an extension of surveying, data from the Global Positioning System can also now be directly input into a GIS after conditioning with the GPS software. We could spend the rest of the semester just on these two topics alone. For more information on the GPS system of hardware, software and satellites, go to Trimble.com. Well, now that we know about the model of the real world? (Next Slide Please).
[Slide Spatial Data Modeling]
This slide shows a flow chart of how one would go about building a spatial data model. We have already talked about deciding which features we want to model, then obtaining spatial data for these futures from maps, aerial photos, etc. Next we have to decide what format to use we have two choices, raster or vector models. Sometimes, the choice is dictated by the spatial data itself for instance if we choose to use satellite images, we are locked into a raster display, whereas if we want to use census data, we will be forced to use a vector model. Whichever model we need to use, it has to be described to the computer. This is usually done by the GIS software package automatically, but we will take a brief look at the computer data structure or the way in which the spatial data is input into the computer.
(Next Slide Please).
[Slide Raster Data Structure]
Here we see a raster feature model, with the accompanying cell values. These values are nominal in nature well talk about this at the end of todays lecture so they probably represent a color, or a feature label, like forest, lake, etc. On the far right, is the way this cell structure is stored in the computer. Note how it resembles a matrix, and that for large numbers of cells, this can become a storage problem requiring great amounts of memory. Techniques have been developed to reduce the size of memory required but these fall into the realm of computer science and wont be covered here.
(Next Slide Please).
[Slide Vector Data Structure]
Now, lets look at the data structure of a vector model. We first notice that the terminology is completely different. We can see points, lines and polygons, except that lines here are called arcs. When we think of lines, we think of a straight segment joining two points. In the vector model, this is generalized so that several lines can be connected to form an arc. Of course, an arc can also be made up of only one line. There are three data files that describe this model a points file which contains the x,y coordinates of the point - an arcs file which shows the points that make up an arc - for instance arc 1 is made up of points 1-7, and a polygons file that contains the arcs that form a polygon - for polygon A that would be arcs 1 and 2, the area of the polygon, and the attributes of the polygon or what the polygon represents. Intuitively, we can see that this structure takes up much less computer space than an equivalent raster model. Did you draw the connection between the definition of a collection of connecting lines or an arc, and the name of our software from ESRI ArcView? There is one more file we have to add to the three shown here, and it involves topology. Remember the first letter in the census data maps TIGER that the T stood for topologically. We now have to introduce topology.
(Next Slide Please)
[Slide Topology]
You can read the definition as it appears on the slide. (slight pause) The idea with topology, is that in order to use analysis techniques with the data, we have to know exactly where each polygon is in relation to other polygons. To be more precise, we have to know which polygons touch each other exactly. To do that, we add another file to the vector data a topological arcs file. (Next Slide Please)
[Slide Topological Arcs File]
Here we see such a file. Note here we indicated that along with node numbers and node coordinates, we add which polygon is located to the left of the arc, and which is located to the right these are the PL and PR entries. Now we have described mathematically that polygons A and B touch along arc 1. Our ArcView software makes use of this data in several analytic operations you will be performing in the lab exercises. OK we could go on and on here about the nature of the computer files in a GIS, but of course we are not computer scientists and that is not our mission. So- lets move on. (Next Slide Please)
[Slide Two New Spatial Entities]
Earlier, we introduced the basic three entities that are associated with spatial models points, lines and polygons. Now, well add two more surfaces and networks both of which have direct applications in Civil Engineering. (Next Slide Please)
[Slide Surfaces]
Here we see a surface drawn in an isometric view. This surface could represent rainfall amounts, temperature variation, elevation of terrain, etc. This particular surface is a digital terrain model, or DTM of Snowdonia National Park in Wales. If we overlay this on an aerial photo of the park, in a process called draping, then(Next Slide Please)
[Slide Snowdonia - Draped]
We see a dramatic view which has a three dimensional look. Note the angle of the sun, the shadows, and the clear view of the road at the base of the mountain. This is not just another pretty picture the surface that the photo was draped on, has real elevation data and can be used for analysis. . I havent mentioned this yet, but DTMs can be used to figure out water runoff characteristics, citing cell phone towers (because you can identify the dead zones in the transmission circle), locating solar panels to catch the most of the suns energy as possible, and so forth. Some Civil Engineering software use DTM data to size storm sewers. The point is that DTMs are very useful to designers.
(Next Slide Please)
[Slide DTM - Njolomole]
Heres another DTM this one of Njolomole, Malawi. Malawi is a land locked country in Southern Africa. This is a black and white photo draped over a raster Digital Terrain Model. Again, we can see great details of the terrain note the pass through the mountains in the lower half of the image. These last two DTM were raster in nature, since rasters are well suited for constructing surfaces. Could we construct a DTM using a vector model? As you might have guessed, the answer is yes but we need to do some preparatory work. (Next Slide Please)
[Slide Vector DTMs - TINS]
In vector space, we can think of constructing a surface as putting together a series of triangular planes to form an approximation of a smooth surface of course the more triangles you use, the smoother the surface will be. This system of connecting triangular planes is called a Triangular Irregular Network or TIN. Its unfortunate that the word network was used because in this context, it doesnt have the same meaning as one of our new spatial entities, as well see shortly. Note in this TIN, more triangles are used in the area of greatest elevation change, and the grid opens up in the flatter areas. Since the position of the triangular vertices is directly under the control of the analyst, a method has been developed which will eliminate points that are unnecessary leaving only what are called surface significant points. (Next Slide Please)
[Slide TINS Surface significant points ..]
By eliminating points that are close together or unneeded, we can reduce storage requirements. If points cannot be interpolated from their neighbors, then they are called surface significant points and are used in the TIN, Conversely, if you can interpolate a points position from its neighbors, then its not needed and is discarded. (Next Slide Please)
[Slide TIN]
Here we see a colorized version of a TIN with vegetation shown by different colors. If you dont look closely at the surface, you cant really see the triangles but it doesnt look as good as the raster DTMs however, it takes up far less room in the computer. Again, we could spend a lot of time going over the details of forming and using a DTM, but alas, we have to move on. The last of these new spatial entities is the Network. (Next Slide Please)
[Slide Modeling Networks]
Lets read this definition of a network - A set of interconnected line features through which material, goods and people are transported, or along which communication of information is achieved. We will drop the second definition which is more appropriate for our EE brothers and sisters. The first definition, however, has great applications in Transportation Engineering. (Next Slide Please)
[Slide Surfaces]
Here we see a model of a network in this case a part of a traffic system. The arrows indicate one way or two way traffic the rectangle represents traffic lights, and so forth. A new term is introduced Impedance - which essentially means resistance to flow. Each component of the network has an impedance assigned to it for instance, a one way street has a greater impedance than a two way street, and traffic signals introduce an impedance. The GIS network software can figure out which path through the network that the least impedance which means which path takes the least time to travel. This path may not necessarily be the shortest path. You can see immediate applications for fire/rescue services, delivery companies like FedEX , and so forth. You should be able to see that the applications in Traffic Engineering are immense. OK, lets summarize using our ski resort example. (Next Slide Please)
[Slide Raster and Vector ]
To set up our spatial model, we first have to decide what entities we want to focus on. Here we see that hotels are modeled by points; ski lifts, by lines; and the forest, modeled by polygons. Next we have to choose what kind of digital format to use. In each case we see the two alternatives, raster and vector. We also studied the computer data structure for the raster and vector formats, and noted that each has advantages and shortcomings. Next, we introduced two new entities networks and surfaces. (Next Slide Please)
[Slide Raster and Vector ]
We can model the roads as a network. While we show the network in the raster format, its most likely that any network will be set up in the vector model. And, finally, we see an elevation model, or DTM, as an example of a surface spatial entity. Using the DTM, we could form a contour model of the ski area. So far, weve focused on the spatial data forming a spatial model. We have to remember that there are also non spatial data in our model otherwise called attributes. So, the last thing we have to cover in this session, is the characteristics of attributes. (Next Slide Please)
[Slide Thematic Charateristics .]
As a review, we remember that attributes give information about the feature, and allow certain GIS operations like Query. If we are going to perform mathematical operations on the attribute data, then we have to be aware that the type of number that measures the attribute is important. Again, there is an unfortunate use of the term scale this time scale has nothing to do with maps. Mathematically, there are four different scales, or levels, that our attribute can exhibit. We use the nominal scale to indicate classes like 1 represents a well, 2 a catch basis, and so forth. You would never try to say add nominal data together because the result would be meaningless. Ordinal scale data has rank assigned to them, so that we can establish order to the data, however, we cannot determine how much heavier heavy is as compared to moderate. Again, arithmetic operations would be meaningless. Interval data is measured on a relative scale, like elevations for example. Differences between interval data are meaningful, but the data has no absolute zero value. Elevations could have a negative value depending on the datum selected. The same with temperature zero degrees Celsius is not as low as the temperature could go. But the difference between a 10 ft contour and a 20 ft contour is 10 ft. And finally, we have the ratio scale. The ratio scale is like the interval scale except it has an absolute zero, so there can be no negative values. Data measuring snow depth would all be positive and would as a ratio scale. OK, as promised, this was a long, information rich session fancy words for tiring. The homework assignment will reinforce what Ive gone over today, but lets take another big picture look. (Next Slide Please)
[Slide How to Construct a. ]
Going back to one of our very early slides in this lecture, we talked about how, in setting up a spatial data model, we follow the same procedures as one would follow when drawing a traditional paper map. Lets look at these steps again but now focusing on what has been presented today. Ill give you a little time to reflect on each of these steps, and tie them in to what you have absorbed from the lecture. Hopefully, something will click. (Pause) Well, whats in store for next time.(Next Slide Please)
[Slide Whats Next]
In the next lecture well explore how attribute data is set up in a database, and how we actually get the spatial data into the computer.
Thats all for now see you next time. Scale
PAGE
Page PAGE 22 of NUMPAGES 22
pr~+>Qde1$"7"8"9"T"%%% &''(z))))++++-.*.11134-4555~777999::;;;;
<>>>>56B*CJaJph B*ph B*ph56B*ph 56>*>*CJaJCJaJ5B*CJaJphN^_pqr+>?s9 e.}01d`d$a$19"T"% &'())+++-.*.11334-455d55}7779999;;:;V;;;;
<>>qBBBEEF
&Fd
&Fd
&Fdd^`d>>?????????qBBBEEEEFF6FIII:I`JuJJ[LpLLNNTNhNiNNaPuPvPPSSTaVuVvVV[[[[[n]]]]]_`-`a%a&a9abbbbbneeeԺ^J5B*ph B*ph5B*ph6B*ph56B*phB*CJaJphB*CJH*aJph B*ph5B*CJaJphHF6FI:IvJJ[LqLLiNNvPPSST`VvVV[[[]]
`-`&a9abbdbneeeggAkrkllnnWpjpstt.vLvxxǁQdeggggg,k?k@kAkrklllnnnnnVoXoCpVpWpjpsssttv-v.vLvxxx|ǁ~ùùCJaJmHnHujCJUaJCJaJ0J
j0JU B*phB*]ph5B*CJaJph56B*ph B*phBd$h]ha$h]h&`#$$&P1h/ =!"#$%ne8+~nU =PNG
IHDR}+WPLTEUbKGDHcmPPJCmp0712OmIDATXX˒0*t'Upul,mQ4Xb$H O(?L>I
t2d@YTm3 Í~ =O=5rM.dd[~GCܲ_
˂uM=sqJP׃&}J4ZU
A#8pDlJqh'<4ՑԆFfІrJA۠(6-3D,MRk6#(H[ҙ=V/m*jD*I~îrU{'.k"DrזJ 9YV+e3GC#FpX"l˲*5k݉ eiOVR|l
Qݲ}6W,nx8h8fzI3$ent l5(ĬJP]Zs`v_!d&h=k,۪f*B*
ph>V@>FollowedHyperlink>*B*
phff@"@Balloon TextCJOJQJ^JaJ, @2,Footer
!&)@A&Page Number,R,Header
!~ ^_pqr+>?s9e.}019T! "#$%%''')***--//0-0111}333555577:7V7777
8::q>>>AAB6BE:EvFF[HqHHiJJvLLOOP`RvRRWWWYY
\-\&]9]^^naaaccAgrghhjjWljlopp.rLrtt{{}}}Q~~~~~~~00000000000000000000000000000000000000000000000000000000000000000 0 0 0000000000000000000000000000000000000000000000000000000000000000000000@0@0
0777:>eGLO15FbHJKMNPI!&14:!!l,b$e8+~nU =&I@0(
j
FA ?toptxtrKT
9'B'k-n-..88t=w=TT``JaOaaaaabbcciiOoTott}}}~~~io"*+6 y
#6=G!!%%''***l-n-..#0,0<11223g333]5l5778899::;;u=w=>>AA-B5BBBFFFFGGrJJLLQQTTzW~W#X)X@XIX|^^^^fgqgkkllCmKmmmnnOoVoppppBrKrVt^ttt{{o}}~~~3333333333333333333333333333333333333333333333333333333333333333333vv||}}~~~~~~~~
LeoneDonald LeonesaleonfC:\Documents and Settings\saleon\Application Data\Microsoft\Word\AutoRecovery save of Lecture 2F03.asdsaleonGC:\Documents and Settings\saleon\My Documents\Worddocs\Lecture 2F03.docsaleonGC:\Documents and Settings\saleon\My Documents\Worddocs\Lecture 2F03.docVޱ'6t$0_"9 22{}/eM[~}Au\Pfd^`OJQJo(n^`OJQJo(npp^p`OJQJo(n@@^@`OJQJo(n^`OJQJo(n^`OJQJo(n^`OJQJo(n^`OJQJo(nPP^P`OJQJo(n^`OJQJo(n^`OJQJo(npp^p`OJQJo(n@@^@`OJQJo(n^`OJQJo(n^`OJQJo(n^`OJQJo(n^`OJQJo(nPP^P`OJQJo(n^`OJQJo(n^`OJQJo(npp^p`OJQJo(n@@^@`OJQJo(n^`OJQJo(n^`OJQJo(n^`OJQJo(n^`OJQJo(nPP^P`OJQJo(n^`OJQJo(n^`OJQJo(npp^p`OJQJo(n@@^@`OJQJo(n^`OJQJo(n^`OJQJo(n^`OJQJo(n^`OJQJo(nPP^P`OJQJo(n^`OJQJo(n^`OJQJo(npp^p`OJQJo(n@@^@`OJQJo(n^`OJQJo(n^`OJQJo(n^`OJQJo(n^`OJQJo(nPP^P`OJQJo(n^`OJQJo(n^`OJQJo(npp^p`OJQJo(n@@^@`OJQJo(n^`OJQJo(n^`OJQJo(n^`OJQJo(n^`OJQJo(nPP^P`OJQJo(n^`OJQJo(n^`OJQJo(npp^p`OJQJo(n@@^@`OJQJo(n^`OJQJo(n^`OJQJo(n^`OJQJo(n^`OJQJo(nPP^P`OJQJo(neM[{}/ $0'u\PfVV.
1ązt$`pH\<Q^JBPzT
6\ċFyHU(J>By:wLZB%&`ްi,nuNM݄u~x$Hĳ\Vr""Dz=T`SBfH"E4h4~Ȁ"pFL*p~6IJj%.fDJ4ZbFPZx]̎?@ABCDEFGHIJKLMNOPQRTUVWXYZ[\]^_`abcdefghijkmnopqrsuvwxyz{~Root Entry F`ͫm1TableS1WordDocumentSummaryInformation(lDocumentSummaryInformation8tCompObjjObjectPool`ͫm`ͫm
FMicrosoft Word Document
MSWordDocWord.Document.89q