Documentation

The GIS resources and tools described in this document can be used without muck knowledge of GIS. If you are interested in how these resources and procedures were put together, you should take a look at some of the GIS user manuals found in the GSD's GIS manual, GIS User Manuals Page

• Using ArcMap for background on GIS data layers and attributes, and how to navigate in the ArcMap interface.
• Geoprocessing in ArcGIS Especially for background on setting the geoprocessing environment for Output Extent.
• Using ArcMap Spatial Analyst for information on some of the Spatial Analysis functions used in these models.
• Using ArcMap 3d Analyst for information about the viewshed tools and working with ArcGlobe.
• Using ArcCatalog for information about creating new shape files, such as our observer points, etc.
• This conference paper by yours truly introduces soem useful terms and techniques

The Tutorial Dataset

Right-click here to download the tutorial project data and models and extract them to your c:\temp directory. You may explore it using windows explorer and then open the ArcMap and ArcGlobe documents to see the data and models.

The Data Sources

The Models

Examining Specific Views froma Human Perspective

In using GIS data to examine views, the most straight-forward approach is to use the GIS to simulate a 3d scene and to simply visit different places in the model to see ewhat they look like. This can be done using the ArcGlobe project provided with the tutorial dataset.

Roll over blue text to see image Here is a photograph from the Science Center Bridge (thanks to Todd). Here is an approximation of the same view from the model. Some views may be tougher to get to, such as this view from the upper story of the new Federal Courthouse on the Fan Peer.. Note that this is a point in our Observer Points layer. Our model will even let us experiment with removing obstructions. the view of Zakim Bridge from Government center. would be pretty good if only those pesky west-end buildings were removed!

Investigating Wholesale Views from Cartographic Perspective

The human perspective is very useful, but like walking around and just looking for specific views, it is difficult and costly to discover view opportunities this way. The Viewshed Tool in your 3d analyst toolbox, under Raster Surface, is useful for discovering what areas on the landscape are visible form specific points. In this, we take advantage of the reflexive nature of views -- If you can see me, I can see you (in the general case) so, if we place a couple of viewpoints about half-way up the spires of the bridge then we can have a map of all of the places that have a view of a decent portion of the bridge.

The first view study we will do will use the plain elevetion model that represents the bare terrain. This will show us the places with potential views if there were no buildings present. The second view study will incorporate the elevation model provided by the MassGIS LIDAR First Return raster. This will show us a closer estimation of the actual views, considering the heights of the buildings and trees. Comaring these two viewshed maps will help us to find places where good views exist, but also may help us find places where the viewe of the bridge might be easily improved by altering or removing or down-zoning a building or two.

Roll over blue text to see image Calculation of a viewshed begins by defining some "observer" points and an elevation model. Our Potential Viewshed Model begins by selecting the bridge points (40m above the elevation given by given elevation surface. Then calculates the Terrain restricted views. People standing in the red areas would see both towers, people in the yellow areas would only see one tower. The bright green areas are publicly accessible openspace.

Calculating Viewsheds on the LIDAR First Return Raster

It is increasingly possible that you may have a raster with the heights of the buildings (and trees and cars) included. This is vrey useful for estimating the actual viewshed. One problem with this is that the bridge itself is part of this model and since our observer points are not above the bridge, but actually half-way down each tower, if we simply repeat the same viewshed operation using this raster, our observation points will be buried and there will be no viewshed at all. If you don't believe me go ahead and try it.

So what we have to do is make layer named Flatten Polys that covers the bridge, and use this area with a Spatial Analyst Conditional function to create a new raster that takes values from the LIDAR raster, except in areas covered by our bridge polygon, in which case the elevation value will be taken from our bare earth raster. These are the functions carried out by the Prepare Lidar model.

Roll over blue text to see image The LIDAR first return raster includes built things -- including the bridge. The bridge is covered by a Flatten Polygon. Then the Prepare LIDAR model is run Producing a Prepared LIDAR model Upon which we can run another viewshed procedure

Comparison of Potential with Actual Views

With maps of the actual and potential views we can find those areas That have good views That could have good views if particular buildings are lowered Or that have good views that may be threatened by likely new construction So we can define zoning overlays that having height restrictions or transfer potential in order to keep important views open.

Back to the Human Perspective

Of course, we can derive all sorts of new information from existing data, but we should never put our confidence in this new info without thinking critically about our inputs and our processes. We may even want to check our result in the field (of course, we should get permission before actually removing any existing buildings!) Before going into the field, we can check our results graphically using the 3d GIS model.

It is nice to go out in the field. When I was exploring the data in GIS, I looked for a public openspace that might have an opportunity for a better view of the bridge. I found this little patch of green in Charlestown that seemed to have its view of the full spires blocked by a single building. It not being far from my house, and also a nice day, I decided to load up mu camera with batteries and ride over on my bicycle. I found the place with an iron fence around it. The sign at the gate announced that it was the first burying ground in Charelstown, established in 1630. The stone carving here is excellent!. At the top of the hill is an obelisk, mirroring the Bunker Hill Monument, and also the spires of the Zakim Bridge. Inscribed HARVARD, it was erected to commemorate the life of John Harvard, which, accoring to the plaque, ended in Charlestown. In keeping with the Harvard enigma that was continued with the statue of Harvard in the Yard of the university that he didn't actually found -- which is not actually John Harvard, because there was no record of what this man looked like -- this is not actually John Harvard's grave, because nobody knows where his body actually ended up.

With maps of the actual and potential views we can find those areas Here is a photo of the Harvard Obleisk Here is the same view from the GIS database removed the skating rink