Introduction:
This lab is an introduction to the applications of Digital Elevation Models (DEMs) in GIS and how they are relevant to archaeologists. We will work with DEM data, derive slope and aspect data, and also work with the concept of viewsheds and the role of these in archaeological analysis and GIS.
Elevation Data: Digital Elevation Models are important for archaeological analysis, as well as cell tower placement, environmental analysis, military uses, and other purposes. A DEM is a way to represent topography using a raster GIS dataset. Each cell has a single value that represents the elevation above sea level of that cell.
Learn more about how USGS creates DEM products.
Lab Part 1
Using ArcGIS 9 we will process a DEM of Orange county, NC and the Chapel Hill area. ArcGIS allows us to create a number of derived raster layers including:
- Contours
- Slope
- Aspect
- Hillshading
Open ArcGIS and go to the Chapel Hill or Hillsboro(ugh) DEM data located in:
the CD-ROM provided to you.
Look at the data available. You will see two 30 meter DEMs from the National Elevation Dataset (NED) of the Chapel Hill and Hillsborough quads, and several point and vector files as well. There is also the new 20 foot (6 meter) LIDAR derived DEM of all of Orange county.
Go here to learn more about the NC LIDAR DEM project. You can download more here.
You need to open Arc Toolbox and look in Spatial Analyst, so go to 'Tools', then 'Extensions' and turn on Spatial Analyst.
Click on 'Spatial Analyst' to see the available tools, and pull down to 'Surface'. This is where the various processing functions are found.
Contour and Contour list
We can generate a vector contour file with any interval required. Try it out. Zoom in to see some detail, but zoom back out before the next analysis.
Slope
We can generate a new raster slope layer here. Slope is often a good predictor of archaeological sites, as many archaeological sites tend to be located on level ground. Try it out and then look at the resulting raster layer.
Aspect
Aspect is the compass direction that the slope faces. This is commonly used in archaeology, as certain types of sites tend to occur on SE facing slopes and tend to avoid N facing slopes. Right click on the name of the file in the TOC at left and choose 'Properties'. Choose a gray-scale color ramp.
Hillshade
This creates a more realistic elevation display. You will want to have the Sun direction from the NW (315 azimuth and 45 degrees elevation) to make it look right. Put the hillshade image 'under' the DEM in the TOC at left, and then make the DEM 40% transparent. Right click on DEM in the TOC at left, click on 'properties' and then 'display'. Set the transparency to 40%. You will get something like this:
This is a more pleasing and realistic view of the landscape. Try some different Sun angles and azimuth values to see the difference.
Lab Part 2Viewsheds (or line of sight analysis): A viewshed is defined as the area that is in view from a given point on the landscape. We usually think of it as what a person can see, but it can also be, for example, all points visible from the ramparts of a hillfort. This has been a major area of interest in archaeological geomatics (see references below).
Now that we have created the derived layers from the DEM, we will conduct several different viewshed anayses of the Chapel Hill/Hillsborough area.
Look at the DEM file and overlay the schools data (put the schools above the DEM in the TOC at left). See how the school locations are located on the hillshade landscape. Now you will run the viewshed analysis tool and look at the results. This will produce all areas visible from the schools.
Now you are ready to run the Viewshed analysis for those points. Go to the 'Spatial Analyst' pulldown, 'surface', and then 'Viewshed'. Pick the input surface (the DEM) and the observer points (your new point file). You can take into account the Earth's curviture if you are dealing with a large area. The 'Z Factor'(a good name for a band...) is if your DEM is in different units (meters) than your X,Y database. Don't worry about it here. Leave the output cell size the same as the DEM, and give your new file a name. Click 'OK' and you are done!
Check out the results. What are you seeing here? You have several parameters that you can alter. Change some and re-run the analysis.
ArcGIS Editor
Next, you will pick an interesting area, perhaps one on top of a hill or on the slope of a hill, and run the Viewshed Analysis function again. In order to do this you will have to create a new point file of that location. This is not very easy, but here is what you do: Open ArcCatalog and Right click on the right side of the screen and choose 'New' and 'Shapefile' to create a new file. In the dialog box, give your new file a name, and choose as the feature type 'point' (you can choose polygon or multiple points another time to do on-screen digitizing-like from the historic maps). Then click on 'Edit' below. Select 'Projected', and 'state plane', NAD1983', and 'NCfips3200', then click on 'add' and click 'OK'. This will make your new shapefile have the same geographic coordinate system and datum as the other GIS data we are working with in this dataset. It is important that these be the same. Before you add a point, double click on your new shapefile icon in ArcCatalog and you can add attributes that you want.
Now go back into ArcGIS and find that file (it is an empty file). Bring up the DEM and decide where you want to do your viewshed. Pick on the top of a hill or some location where you will get some good results.
Now you have to go to start up the 'Editor' tools, so go to 'view', then 'toolbar' and turn on 'Editor'.This will bring up the Editor menu bar and tools. Now go to the Editor pulldown, and choose 'Start Editing'. 'Create new feature' is the default, and that is what you want. The 'Target' should be what you named the file. Then click on the 'pencil icon' to start using the on-screen editior. Place the mouse icon where you want to add your point and left click there. Now click on the 'Attribute Table' icon at the far right (looks like a text table) and you can see the tabular data. These can be edited as well.
Go back to the 'Editor' pull-down and choose 'Stop Editing'. This completes the digitizing session. You have now finisheded creating a new point shapefile. Easy!
Now run the viewshed analysis on that point.
Try some different options. Does this LOOK right to you? You can do viewsheds for single points, multiple points, or vector (road) files. There are some additional optional parameters, including offset parameters (like a hillfort rampart that is above the DEM altitude).
Lab Part 3
Interpolation: The Hageman and Bennett and Weimer articles we discussed today consider the various options available for the interpolation of point data into a raster surface. ArcGIS has a variety of these tools available.
In the ArcToolbox, go to Spatial Analyst tools and click on Interpolation. You will see the various options that are available, including IDW, Kriging, and several others. You can learn more about them by using the help function.
There is an excellent website on Multidimensional GIS Spatial Interpolation.
Some Viewshed Issues to consider for next class (be prepared to discuss these):
- problems such as presence of trees, observer heights etc . How can we deal with this?
- influence of scale and spatial resolution of raster data
- where does a viewshed actually end (can't see/doesn't influence you much into infinity)
- curviture of the Earth
- what can this really do for us, what does it really have to do with what people do on the landscape?
Lab Part 4 (to be done later)
Review several online Viewshed websites for discussion next time:
Scott Madry's (again?!?) Scenic byway viewshed analysis in New Jersey
A comparison of ArcInfo and GRASS viewshed results (they are different)
Viewshed analysis along the Apalachian trail
Mars lander viewshed analysis
ArcGIS Spatial Analyst: Advanced Analysis Using Raster and Vector Data: an ESRI white paper.
Archaeological viewshed references:
Madry S.L.H. Rakos L. 1996 Line-of-sight and cost-surface techniques for regional research in the Arroux river valley in Maschner, H.D.G. New Methods, Old Problems: Geographical Infromation Systems in modern archaeological research Southern Illinois University Center for Archaeological Investigations. Occasional Paper No. 23. Carbondale (line-of-sight, cost surface, hillforts - looking at the relationship between Iron Age hillforts and roads in Burgundy - routes preferentially selected within sight of hillforts)
Ruggles C.L.N Medyckyj-Scott D.R. Gruffydd, A. 1992? Multiple viewshed analysis using GIS and its archaeological application: a case study in northern Mull in CAA92 125-132
Gaffney V. Stancic Z Watson, H eds 1996 Moving from catchments to cognition: Tentatative steps towards a larger archaeological context for GIS in M.S.Aldenderfer and H.D.G.Mascher Anthropology, Space and Geographical Information Systems Oxford University Press Oxford Mascher 1996 (viewsheds - positioning of monuments)
Lock G.R Harris T.M. 1996 Danebury Revisited: An English Iron Age hillfort in a digital landscape in M.S.Aldenderfer and H.D.G.Mascher Anthropology, Space and Geographical Information Systems Oxford University Press Oxford (viewsheds - long barrows not seen from other barrows)
Wheatley D.M. 1995 Cumulative Viewshed Analysis: A GIS-based Method for Investigating Intervisibility and Its Archaeological Application. In Lock, G.R. & Z. Stancic (eds) Archaeology and Geographic Information Systems: A European Perspective. Taylor & Francis London (viewsheds - cumulative viewshed analysis at Neolithic long barrows at Avebury and Stonehenge)
Lake M.W. Woodman P.E. Mithen, S.J. 1998 Tailoring GIS software for Archaeological Applications: An example concerning viewshed analysis JAS 25 27-38 (viewsheds)
Ruggles C.L.N. Medyckyj-Scott D.J. 1996 Site location, Landscape visibility and symbolic astronomy: A Scottish Case Study in Maschner, H.D.G. New Methods, Old Problems: Geographical Infromation Systems in modern archaeological research Southern Illinois University Center for Archaeological Investigations. Occasional Paper No. 23. Carbondale (landscape perception, viewsheds, multiple viewsheds - Bronze Age monuments on Mull relationship between monuments and astronomical phenomena on Mull)
Ruggles C.L.N Medyckyj-Scott D.R. Gruffydd, A. 1992 Mutiple viewshed analysis using GIS and its archaeological application: a case study in northern Mull in CAA92 125-132
Fischer P. Farrelly C. 1997 Spatial analysis of visible areas from the Bronze Age cairns of Mull JAS 24 581-592 (viewsheds)
Maschner H.D.G. 1996 The politics of settlement choice on the northwest coast: Cognition, GIS and coastal landscapes M.S.Aldenderfer and H.D.G.Mascher Anthropology, Space and Geographical Information Systems Oxford University Press Oxford (viewsheds, settlement changes - showed changes in location of settlements from the Middle to Late Phases)
Feel free to contact me if you have any questions. See you after spring break week after next. Have a great break.
~ Main page ~ Schedule ~ Web Resources ~ Bibliography ~
