Hello. I have been reading some interesting articles, in which navigation least-cost paths are calculated for some study areas. For instance:
Leidwanger, J., 2013. Modeling distance with time in ancient Mediterranean seafaring: a GIS application for the interpretation of maritime connectivity, Journal of Archaeological Science 40(8), 3302–8.
Cost surfaces (and least-cost paths) are based on winds speed and wind directions. The method whereby that is achieved is not so clear to me. If, on the one hand, it is clear that the basic cost is devised on the basis of wind speed, on the other hand it is not clear to me how the wind direction is factored in.
Taking into account wind direction adds complexity to the cost-surface modelling since, other things being equal, moving (for instance) against the wind direction is a limiting factor (i.e., a ship cannot move against wind), while the opposite facilitate the navigation.
As far as ArcGIS is concerned, I have the impression that wind direction can be incorporated as a HORIZONTAL FACTOR (http://desktop.arcgis.com/en/arcmap/10.3/tools/spatial-analyst-toolbox/how-the-horizonal-and-vertical-factors-affect-path-distance.htm), still I cannot wrap my head around how to operationally implement that in GIS.
Has anyone been faced with something similar, or has anyone any comment or suggestion?
Best
Gm
It actually seems very similar tho their version of slope drainage using a d8 algorithm. How I would approach this is to simply classify each neighboring cell as upwind or downwind of the object cell on a iterative basis. So start in a Cell R1). Code each adjacent cell with the direction you desire to travel using the 1N,2NE,4E,8,16,32,64,128 and so on format (r2). Code each cell with the wind direction (R3). The wind direction would be generalized to the nearest direction from S to N = 1, SW to NE = 2 and so (r4). Running a simple subtraction then 0 is the least cost path = wind behind you and based on this math you could assign friction weights, the bigger the difference in values the bigger the friction. You need to check the math to ensure it holds true. You may alway need the direction you want to travel = 1 and corresponding wind direction = 1, then the opposite wind direction -1 or similar.
I'm not sure but Network Analyst or Spatial Analyst models of ArcGIS may help. I used Network Analyst to design least-cost transport network in my research.
Hello, thanks for your reply and suggestion. @Stuart: thank for your guidance. I am wondering if using Path Distance tool would do something like that in a simpler way. I am still looking into that ArcGIS function, but I did not arrive to anything definite yet. What seems intriguing is that the Horizontal Factor that can be specified within that tool should account for "any horizontal friction encountered" while calculating the total cost of moving into a cell.
- Badges
- Science topic
- Similar topics
- Geoscience
- Geoinformatics
- Geographic Information System