Bobwhite - Manatee Transmission Line Project Analysis

Intro to GIS - Final Project

Our final project was very long, involved and quite challenging.  We were tasked with analyzing key factors that were originally considered in gaining right-of-way access for a real transmission line project for Florida Power and Light Company (FPL).  We started with the background phase which was providing a cartographic model, basically a plan, for how we would go about the project, what data we would need and where we would get it.  The next phase was an analysis of the data.  There were four objectives we had to analyze and in order to do that we had to create maps for each objective, including datasets that would allow us to examine the data and run analysis tools on it. Most of the data was provided, some of it we had to download, and some we needed to create ourselves.  This process was a real challenge that required skills from pretty much the entire course.

The last phase was the presentation.  We had to create a slide show and transcipt as though we were presenting the data to a high-school level audience.  

Here are links to my slide presentation and the transcripts for it:

http://students.uwf.edu/mr80/Intro2GIS/Finals/Intro2GISFPSldSh.ppsx

http://students.uwf.edu/mr80/Intro2GIS/Finals/Intro2GISFPPrsnt.pdf

I really enjoyed this class and am looking forward to moving on to the next semester.

Cartographic Skills Final Project

Whew!  That's done.  Our final project pulled in many of the skills we learned over the course of this semester.  Our project was to create a bivariate map, one map displaying two thematic datasets over one geographic region, for the Washington Post to use to help explain their article on high school seniors and college entrance scores.

We were provided with a document with the participation rates and the the mean scores for the three portions of the SATs.  We had to add the three scores to come up with the mean composite score for each state and add that, the participation rate, and the state name to a table that could be imported to our map.  We were also directed to a site where we could find an appropriate basemap to start from.  This map had no projection so our first task in ArcMap was to give it a suitable projection.

Because I was dealing with a large land mass that ran generally from east to west I chose to go with Albers Equal Area Conic.  Then it was a simple matter of adding the table to my map and joining it with the attribute table of the U.S. map dataset.  To symbolize my first dataset, the mean composite SAT scores, I chose to go with a Choropleth map classified with the Quantile method and 5 classes.  This gave me a nice, even distribution of the data with enough classes to represent the ranges well, but still allow for a nice distinction between the ranges with my color scheme.  For that I chose to go with a sequential range of blues with the lightest representing the lowest range of scores, and the darkest, the highest.

With the Choropleth map as a base I decided to go with a Graduated Symbol method for my second dataset.  I chose a dark blue color to go along with the color scheme of the basemap, and circles for my symbol.  I used an Equal Interval classification method with 5 classes, the smallest for the lowest participation rate and the largest for the highest.

I think the map does a fair job displaying the two sets of data for a comparison of the SAT scores and the participation rate.  I added two insets to show Alaska and Hawaii separately.  This allowed me to zoom in nice and tight on the contiguous states to make better use of my paper space.  I waited to add my state labels until I moved over to Adobe Illustrator.  Once I was there adding them was quick and easy.  I was very impressed with myself for being able to correctly locate all the states without having to look any of them up.  I guess I'm easily impressed.  I added a medium-light blue background to the map because I didn't want it to contrast or stand out and detract attention from my map, but the color blended just a little too well so I added a drop shadow to the contiguous states.  I felt that added all the distinction I needed to separate the map from the background in such a way that it stood out.  I then added my title, subtext and map information and was done.

This has been a very interesting course.  This project helped me realize I had learned more than I realized and I was quite please most of my reference to previous lessons was just for verification.  I'm still a bit intimidated by AI, but I think my skills there have improved considerable since my first go.  I'm looking forward to moving on to the next set of classes and seeing what else is in store for me.

Google Earth

Cartographic Skills - Mod 12

This week we used our Mod 10 Dot Density Map to create a shareable web map and tour.  We pared down our Mod 10 dot density map of the population of south Florida to just the absolute essentials to achieve the desired affect we wanted in out Google Earth map.  With that done we used the Map To KML Conversion Tool to convert the map to a .kml file that could be viewed in Google Earth.  Next we used Layer to KML to convert the dot density layer.  Converting this layer with this method ignored most of the symbology settings in the layer, but still retained all the attribute data so it would be available in Google Earth.

Once the .kml files were converted we opened them in Google Earth.  After adding the layer.klm it was possible to click within a county boundary and have a pop-up box pop up with the attributes for that county.  

Part two of this assignment was to then use this map to create a tour of several of the locations on the map.  The process itself is simple enough.  All it takes is placing Placemarks in the locations you want to zoom to, then record the process of zooming from one Placemark to another, entering street view to look around at each location.  Where it gets difficult is entering street view in a location that will allow you to pan around and get the best view, and panning smoothly.  It took quite a bit of trial and error to find locations that would work well for the desired affect, but then selecting that location during the course of recording didn't always go so well so the recording had to be redone many times.  Once in a good location it was difficult to pan around smoothly.  Moving the mouse from one side of the screen to another wasn't too bad, but then the mouse had to be moved back to the first side of the screen to pan any further.  This always caused a bit of back and forth motion and several breaks in the pan.  I'm sure there's a better way to do it, but I was not able to figure that out on this assignment.

Another difficulty I had while recording the tour was turning layers off and on during the zoom process and switching to street view quick enough there wasn't too long a pause.  I finally figure out where best to place my Placemarks so I could use them as a better guide to where to enter street view and that cut down a lot on the delay, but it still wasn't as quick and smooth a transition as I would have liked.  I also ran into delays enter the street view because I wasn't quick enough turning off layers that interfered with the view when close up.  By the final recording I had that pretty well under control though.

I think this lesson was pretty easy to complete, but it will take quite a bit of practice to perform the process smoothly.

Georeferencing, Editing and ArcScene

Intro to GIS - Week 13

This lab started with georeferencing two raster images of the UWF campus.  Georeferencing is the method used to tell a raster dataset with no geographically referenced coordinate data built into it where it belongs geographically.  This is done by linking the target layer, the unreferenced layer such as the raster image, to a referenced layer using control points to match up common points between the two.  In this exercise we matched buildings in the raster to buildings in a referenced building layer.  Several points are usually required to create an accurate reference and they should be evenly spread throughout the image.  I found the Image Viewer Tool to be very helpful in placing the control points.

Once the links are places with the control points they can be viewed in the links table.  This table shows the residual value which tells you how much each link agrees with how the layer is currently displayed.  The lower the value, the more accurately the control point is georeferenced. I had two control points that were considerably higher than the rest and when I zoomed in to look at them more closely I saw they weren't that well aligned and there were a couple other buildings around them that were not either.  I deleted those control points and set a couple others.  This made for a much better visual alignment and when I checked the table these residual values were much more in line with the rest.  Once my residuals were more in line and a closer examination of my image was made I felt I had a good match.  I went back to the table to check the RMS (Root Mean Square) Error and saw it was well below 15.  RMSE is frequently used in GIS as an indicator of the accuracy of the spatial analysis and/or remote sensing.  It is a measure of the differences between the calculated values and the actual, or observed/measured, value.  The difference between the calculated and actual value is called the residual.  The RMSE, derived from squaring the difference between the actual and calculated values adding these residuals together, dividing that by the total number of values and taking the square root of the result, aggregates the residuals into a single value.  My RMSE for the north raster of UWF was 4.743 with a 1st Order Polynomial Transformation.  My second raster for the south portion of UWF was much better at 1.66888 with a 2nd Order Polynomial.

The next section of our lab was editing.  In this section we used editing tools to digitize two new features; a building and a road.  To digitize a new feature you must first start an editing session.  This can be done either by clicking on the editor menu of the editor toolbar and selecting start editing or right-click on any layer in the table of contents and select Edit Features then Start Editing.  Selecting Create Features opens a window with the templates in the map in the top panel and tools available to create that type of feature in the bottom panel.  We used the straight segment and endpoint arc tools for both the building and the road, utilizing the snap options on the road to make sure the new road lines met existing road lines and endpoints were placed properly.  Edits are not automatically saved, not even by saving the map.   Before ending the editing sessions with Stop Editing you must first select Save Edits or everything will be lost.

The third section we had to create a multiple ring buffer around an eagle's nest on campus property to show the location of a new conservation easement to protect the nest and the eagle.  First we started with creating a hyperlinked picture of the eagle nest to the attribute data, then through the layer properties tab HTML Popup, choosing to show content as a URL using the picture field.  Next we used Identify to select the point feature in the drawing and clicked on the lightening bolt to verify the link worked.  Next we used the Multiple Ring Buffer tool to create two rings around the protected nest.

Section 4 had us create a 3D view of UWF in ArcScene.  We started with a DEM file of UWF which had elevation data, then draped the other layers over it in the Base Heights tab of each layer's properties and selecting "floating on a custom surface:" and selecting the UWF_DEM to use its elevation data.  Next we used the height field of the buildings layer as our z-value to extend those features above ground by checking "Extrude features in Layer" option in the Extrusion tab of the buildings layer properties. We then set the vertical exaggeration in the scene properties to accentuate the building heights.  With that done we exported the scent to a 2D image to finish off in ArcMap since ArcScene does not allow the addition of map elements.  I had to go back and forth between ArcMap and ArcScene a couple times to adjust my angle so the digitized road would show in the image but there would still be enough relief to show the height of the buildings.  I ended up having to increase the line weight of the road symbol to make sure it showed up.

I really enjoyed the editing portion of this assignment and am looking forward to doing more such work in later courses.

3D Mapping

Cartographic Skills - Mod 10

This week we learned about different types of 3D data and visualization techniques.  The first part of this three part assignment was done in Esri's Virtual Training.  The two broad categories of data used in 3D Analyst are surfaces, such as rasters, terrain data sets, or Triangulation Irregular Networks (TINs) and 3D features, features with discrete boundaries such as buildings, wells or roads.  Some 2D data can also be used.  The difference between 2D and 3D is that 3D has a z-value.  The z-value is a value represented on the z-axis in a three-dimensional x,y,z coordinate system.  

The z-value does not only represent elevation, it can also be used to visualize other attributes such as population or cost.  In one of our later exercises we extruded land parcels based on the property value as the z-value.  The more valuable the property, the taller it appears.

2D spatial data can be displayed in 3D perspective as long as you have a surface layer with elevation values and the same extent as your 2D data.  One of the techniques used to accomplish this is setting the base heights.  You start with a surface for your study area, such as an elevation raster, and set its base heights, turn on shading and choose symbology.  You may then drape your 2D feature class on top of the topography using the elevation surface to set the base heights for your feature class. These steps are all done through the layer properties.

Base heights establish the elevations of surface locations and features.  It is what tells 3D Analyst the elevations values of layers and their features.  Shading and symbology increase the perception of depth, enhancing details within the topography.

Other techniques for enhancing 3D views are setting vertical exaggeration, illumination and background color.  These techniques are all scene properties rather than layer properties.  They will affect the whole scene rather than one layer or another.  Vertical exaggeration is used to amplify the surface, emphasizing subtle changes when the horizontal extent is significantly greater than the amount of vertical change. Illumination can add realism such as season or time of day by controlling the position of the light source as though it were the sun.  Relief effects are harder to distinguish in shadow so changing the position and angle of the light source can help to emphasize them.  Background also helps to add realism to the scene.

Extrusion is a simple technique used to create 3D symbology from 2D features.  The process stretches a flat 2D shape vertically to create a 3D object.  Extrusion can also be applied negatively as in the case of wells, basements, pilings or other underground features.  Extrusion is typically used to set heights for features, such as buildings, but it can also be used to visualize other attributes such as populations or cost in 3D perspective basing the z-value on a value other than elevation as discussed earlier.

The second part of our assignment was to actually convert a 2D building layer to a 3D building layer.  We started with a building polygon layer that needed elevation data and a raster layer that contained elevation data. To add the raster elevation data to the building polygons we started using the Create Random Points tool to generate random points within the building shapefile so we could then add the elevations information to those points using the Add Surface Information tool.  Once we had our z-value we then used the Summary Statistics tool to generate a single elevation value for each building then joined the Mean Z-value to the building footprint layer.  With this done we then had all the data we needed to create a 3D image in ArcScene by extruding the features in our building layer to the Mean Z-value we created in previous steps.  With this done we converted the data to a KML file to be shared on Google Earth.  Viewing this assignment in Google Earth was pretty cool.

Part III of our assignment was to compare and contrast the 2D and 3D versions of Minard's map of Napoleon's Russian Campaign of 1812.  I found the 3D map to be pretty cool and liked some of the added information that was available in the display that wasn't available in the 2D version, but all in all, for the general population I thought the 2D map was still the most useful.  All the information it contained was right there to be seen.  In the 3D version there was always some bit of information that could not be viewed without changing the view a bit.  And when you did, some other bit of information disappeared from view.  So while I liked the 3D version more, I thought the 2D version to be more usable.