I'm wondering if anyone can help me out. I'm trying to carry out a viewshed analysis of wind farms and single turbines on a region of the UK. I have the landform data as well as the coordinates of around 1,600 turbines, I'm looking to create a 3D map, plot these turbines on the map and carry out a viewshed analysis to determine what area is suffering from the highest level of visual impact.

I am using ArcGIS Desktop 10.2.1.

  • 1
    I'm assuming you have the Spatial Analyst Extension... Have you gone through the tutorials on that topic? What have you tried? What, specifically, do you need help with? – Fezter Mar 10 '14 at 23:53
  • Hi, thanks for the help. I'm a student so quite new to ArcGIS but know the basics, I'm using it in my dissertation. I have all of the extensions and access to map data from Digimaps. I need help with the methodology. I have all of the turbines coordinates in an Excel file and have created a DEM but I'm not sure how to get the points onto the DEM and carry out the viewshed analysis, I'm looking to identify the town which can see the most turbines in the region. – Ross Mar 11 '14 at 0:19
  • 2
    Did you know that you can use the edit button beneath your Question to revise it with additional details such as these? Doing that rather than using lots of Comments keeps the Q&A easy to read. – PolyGeo Mar 11 '14 at 2:14

See my answer to Understanding Viewshed Analysis in ArcGIS Desktop?. To get your points into ArcMap, export them as a CSV from Excel and then import that using Add XY Data. I'm sure you understand that you do not use this data to edit the DEM (I don't mean to patronise you and only mention this because of the way you phrased your comment, which rang an alarm bell - but maybe that's just the way I read it).

From your question, there are a few points I think you need to consider:

  1. Be warned that 1,600 turbines is a LOT. From experience, this number of turbines will cover a very large area and, presuming you are using at least the OS 50 DTM (this is acceptable though the 10m resolution data is sometimes preferred) the analysis will take a very long time (probably more than one cup of coffee and depending on your computer's power it could be a few hours).
  2. To perform visibility analysis you not only need the coordinates for every turbine but also the tip and/or hub height (and use this for OFFSETA and set OFFSETB to 2).
  3. I'd recommend accounting for the curvature of the Earth and refraction of the air over these sorts of distances.
  4. In your analysis you may need to consider a few other factors such as screening by buildings and trees as the DTM provided by OS is 'bald Earth'. Conservative estimates for woodland blocks and buildings are usually acceptable (15m and 5m respectively are usually OK but you could use others if you can justify the values). You may wish to discuss the differences between actual potential visibility after you account for atmospheric haze etc and theoretical visibility (for instance research suggests that apart from the clearest days you'd be (un)lucky to see turbines much beyond 10KM whereas Scottish Natural Heritage requires ZTVs to be calculated up to 35KM for taller turbines). Screening by buildings is particularly pertinent in a town, where due to the visual equivalent of the canyon effect of the streets, most places in a town can't actually see any turbines, even where your theoretical visibility calculation suggests they can see many. So, if this is serious analysis, then the urban layout is a significant factor (a town with a large open market square where people gather may have a higher practical level of visibility than a more densely built-up urban area that apparently shows a higher ZTV score.
  5. You say you are looking to identify the town which can see the most turbines in the region. That is a rather vague term which needs some definition. Do you mean the town which contains a location with the highest number of visible turbines, or do you mean the town that can see the highest number of different turbines summed over all locations within the town, or possibly the town that has the highest average visible turbine count over the whole town area. What you can't do is sum the total visibility of each cell within the town limits because you will then be double-counting turbines and weighting the results towards larger towns.
  • Hi, your answer is fantastic, thanks so much! Life saver! Even though my university teaches GIS the lecturers don't know how to help me. Anyhoo your help is much appreciated, I think I owe you a pint! Sorry I have't phrased the question very well. I have gotten as far as displaying the XY coordinates so I'm sitting with lots of points on a pretty map. – Ross Mar 11 '14 at 11:24
  • Sorry, ran out of time to edit and accidentally posted. What I was hoping for was to use viewshed on the map to show what areas the hub of a turbine can see and do this with all turbines so that an area that more turbine hubs can see will appear darker than an area only one turbine hub can see. I can then identify a darkest areas as being affected by more turbines and use this to find my study area. Thanks again for your help! – Ross Mar 11 '14 at 11:45
  • The simplest approach is to create an attribute in your turbine points Excel sheet called OFFSETA and set that to the hub-height for each turbine. Create another Attribute called OFFSETB and set it to 2 (2m - actually a bit high for eye level but that's the standard used in the industry). Create a third attribute called RADIUS2 and set it to 35000. 35km is the largest distance used to assess visibility for the tallest turbines. That may be larger than usual for some of your turbines but it stops visibility being calculated over the entire DTM for all turbines. Save as CSV and add XY layer. – MappaGnosis Mar 11 '14 at 12:02
  • The output will be a count of points (turbines) visible at each location by a person with eyes 2m off the ground (!). You can then symbolise the output how you like (the higher the number perhaps the darker, or maybe use a colour ramp). Read the documentation to understand all this! BTW what University are you at? – MappaGnosis Mar 11 '14 at 12:04
  • Thanks for the advice, I think I'm getting the hang of it now. I'll do some more reading on it but I think I know what I have to do now. I'm at Glasgow Caledonian University, we have GIS lecturers but they only know the basics and have no idea how to help me with my project. – Ross Mar 11 '14 at 12:28

Not the answer you're looking for? Browse other questions tagged or ask your own question.