Geographic Information Systems Stack Exchange is a question and answer site for cartographers, geographers and GIS professionals. It only takes a minute to sign up.
Sign up to join this community
I am using a SRTM 30 meters for slope and elevation. When i'm using this 30x30 rasters to generate contours i set a 30m equidistance for contour, and the results are ok.
Now i'm wondering if it's wrong to set a 10m contour from a 30x30 DEM like ASTER or SRTM.
If you want to create 10 m contour interval, so you have to find a primary data ( Your DEM's data) which has spatial resolution 10 meter or smaller like InSAR.
Your can't create an interval contour smaller then it's DEM resolution because when you generate the contour using ArcGIS, the algorithm will try to find the nearest neighborhood between spot height ( the centeroid of cell with its value as elevation) to be interpolated.
If we use the 30m resolution of DEM and try to generate 10 meter of contour intervals, the spot height haven't enough neighborhood to be interpolated and it's also impact on contour's accuracy.
Spatial resolution (cell size) of DEM means a smaller pixel unit that representing earth surface and this depend on a sensor used to record, spot height density, and it's standard deviation. It also impact on map scale and its contour interval..
As a rule of thumb you generally do not want to set your contours to be any smaller than your cell sizes. There is no point in doing so because even if there were 3 different values across the 30m cell, it would all be averaged to the same. If you are using a 30m DEM the lowest contours you should use is 30m, as that preserves the accuracy of the data and creates the smallest contours within that preservation.
I don't entirely agree with the previously given answers. It depends on the landscape structures being present in the DEM.
Equidistance in that context means vertical distance and not horizontal x or y distance. As stated before, coarse resolution can result in inaccurate contour lines, simply because the height measurements are not as dense as in a DEM with higher resolution. But in plain areas, meaning little variation concerning height values, it is valid to choose equidistance values smaller than the given DEM resolution. Imagine a landscape ranging from 0 m to a maximum height of 20 m. Maybe a beach with reoccuring peaks of dunes (mind the max height of 20 m) is a nice picture. The generation of contour lines based on a DEM with 30 m resolution would result in just one contour line - if you would choose an equidistance value of 30 m or higher. In that case you miss out morphological information in your derived contour line dataset. That doesn't mean you can create a highly detailed contour line dataset by setting the equidistance parameter to 1/10th of your DEMs resolution, although the result might look like it. But if the dune structure is large enough to be captured by the spatial resolution of the DEM, you surely can set the equidistance parameter lower than the DEM resolution in order to maintain the dune structure in your contour line dataset.
As the lower limit for the equidistance parameter I would therefore rather suggest the height accuracy of the procedure how the DEM has been created, which is about 6 m for the 30 m SRTM DEM (according to a quick check at Wikipedia). The same applies to a laserscanning derived DEM with a resolution of 1 m, where you could theoretically create contour lines with an equidistance of 0.5 m. That being said, in many cases you would probably want to choose your equidistance parameter higher than DEM resolution anyway because of height accuracy, scale or memory concerns. In the end its always good to compare the contour line output with the original DEM.
Create a 10 m contour lines dataset from a 30 m DEM, but don't expect it to be more detailed than your DEM.
