According to Tobler's Law, the influence of one entity on another one can be estimated by the distance, the strait line. But in wind field, this is always not the case, so I need an alternative to estimate the shortest path in wind field.
Maybe my discription is unclear, I mean, how to estimate the shortest path from one point to another point in a wind field, taken this dynamic field as the cost raster.
This morning, I found a tool in ArcGIS named path distance, it says that some dispersion and hydro process can be successfully modeled....But I am not sure if this works...
You can use this tool arcgis: Cost Distance. its disadvantage is that it only serves two points
http://help.arcgis.com/en/arcgisdesktop/10.0/help/index.html#//009z000000z5000000.htm
This can be solved by using Arcgis Engine. However, the model in ArcGIS can only simulate the process with wind speed. However, a surface wind field always consists of wind speed and direction. This tool may be a simplistic. But, if I average the wind speed , the mean may be quite the same...Maybe at this time, I can use this tool in ArcGIS Engine...
Yeah, the description wasn't really clear at the beginning. I was wondering how a wind field can be represented by network (since the term 'shortest path' you generally associate with network distance). However, I have experienced people doing spatial modeling with wind data (I haven't done it myself yet). You are right that the euclidean distance cannot represent the distance in a wind field, instead you could go for great circle distances (function 'spDists' in the 'maptools' package of R). But if you would like to calculate cost distances from raster, R package 'gdistance' is what you are looking for. It also provides options for other types of distances computations on raster.
Hope that helps.
- Badges
- Science topic
- Similar topics
- Geoscience
- Geoinformatics
- Geographic Information System