Take the 2-minute tour ×
Geographic Information Systems Stack Exchange is a question and answer site for cartographers, geographers and GIS professionals. It's 100% free, no registration required.

I am using ArcGIS 10.1 to generate a DEM (bare earth) from LAS dataset. To accomplish this, I use 'Las Dataset to Raster' tool as described in Esri website.

I have read here and there in the web that the the cell size of the output raster should be set at a value corresponding to the average LAS points spacing; somewhere else, they suggest to use a value that is three times the average point spacing.

Now, since I wish to produce a bare earth DEM (i.e., with only ground returns), I am wondering if the cell size has to be set at the average point spacing of the whole LAS dataset (i.e., comprising non-ground, ground, etc), or exclusively of the ground points?

On the basis of the ArcGIS's 'point file information', I managed to get statistics for each lidar class. The average spacing (whole dataset) is 1.5 m, while for ground points only is 0.67.

Any insight into the issue?

share|improve this question
    
The average point spacing should be smaller for the whole dataset (more points, smaller distances among them) rather than the LiDAR cloud with only ground returns. Maybe you are referring to point density? See here, for disambiguation of terminology. –  Andre Silva Jun 13 at 15:49
    
Thank you for the comment. I used the same ArcGIS tool. Now I am rather confused: is the point spacing reported by ArcGis the point density? It is my understanding that what reported shound actually be the point spacing.... –  NewAtGis Jun 13 at 17:42
    
The ArcGis tool 'Point File Information' reports correctly the average point spacing. Maybe you just switched the informations, i.e., 1.5m to ground returns and 0.67m to the whole dataset? –  Andre Silva Jun 13 at 18:11
    
Hello! I took the attribute table associated to the point file information shp file, and i pasted into a stat tool. Then, I calculated the statistics of the point spacing by class code. The average point distance of the ground class (i.e., 2) was 0.667. The same statistic for the whole dataset was 1.54 instead. So, I would go on and use 0.667 as baseline to compute the output raster cell size (according to the hints provided in this same thread). Am I missing something? Thanks –  NewAtGis Jun 13 at 20:03

3 Answers 3

It really depends on what you're going for here. If your goal is to make accurate measurements, you will want the grid spacing to be smaller, around the average point spacing of the dataset. If this is part of a process of creating orthophotos or analyzing watersheds, or any other task which requires a more generalized terrain model, you will want the spacing to be larger.

Also, if you've filtered the data to the ground points, USE THOSE! You can ignore the other classifications.

Three times the size of the point spacing is generally considered a good idea for generalization because it allows the grid cell three points to average out to a theoretically consistent value. In my profession, sometimes when I produce orthophotos I allow a huge grid size, often 8-12 times the size of the point spacing.

My point (heh) is: there is no right or wrong way to produce a bare earth model. Sure, there are specifications sometimes (and if this is one of those cases, I suggest you disregard whatever I'm saying and follow the specifications), but producing what you need to suit your problem can be totally heuristic.

Do you want a more rigid, presumably more accurate model with harder edges on which you can model terrain more precisely? Use a smaller spacing. Do you want a more general terrain model that is smoother, less precise, but shows the overall terrain characteristics of a large area? Use a large spacing.

A lot of this is trial and error. That's the beauty of this kind of thing: it's about 30% science and 70% art.

share|improve this answer
    
Hello! I much appreciate your prompt and informative reply. Now I got an idea of the issue and of the approach that should be context-specific. What do you think about the first part of my question, that is as to use the average spacing of all point vs average spacing of only ground point? –  NewAtGis Jun 13 at 14:24
1  
Again, it really just depends. Your average spacing for all points is going to be skewed because it will include points that are not spatially correlated. Overall average is going to be an OKAY way of analyzing your set but not a GREAT one. If you've taken the time to filter the data, I'd ignore the average overall spacing. If you want to stick really tight to those ground points, use the around average ground point spacing (though I'd round up to the next tenth or so in this case). Otherwise, smooth it out a bit by using a larger size. –  Wes Jun 13 at 14:31
    
Thanks. NOw it is clear. Very appreciated. Best. NaG –  NewAtGis Jun 13 at 14:35
    
Like I said in my rant there, if you're trying to hit some kind of spec, forget what I'm saying and follow those. @Barbarossa gave a good insight into USGS specs in his or her answer. –  Wes Jun 13 at 14:42

I work as part of a production team that acquires LiDAR and generates DEMs for several government agencies, including the USGS. It is common to see specifications that limit cell size to be no smaller than the nominal pulse spacing (NPS) of the entire project (all points). For instance, if the NPS was 2m for a dataset, our LiDAR derivatives were to be no smaller than 2m. The reasoning behind this is that when the cell size is smaller than the NPS, there is considerable interpolation of values between know points. This in turn creates a less than accurate dataset.

EDIT: NPS defined

A good definition in this document:

Nominal pulse spacing (NPS) refers to the average point spacing of a LiDAR dataset typically acquired in a zig-zag pattern with variable point spacing along-track and cross-track. NPS is an estimate and not an exact calculation; standard procedures are under development by ASPRS for NPS calculations

share|improve this answer
    
Thanks. Sorry if I repeat again my question, but I am not a specialist and I had not previous familiarity with Lidar....So, how should I go on? Provided that cell size has to be not smaller that the average point spacing, since I am using ground points, do I have to set cell size 'around' the average spacing of that class or of the whole dataset? Thanks –  NewAtGis Jun 13 at 14:32
    
+1. Could you clarify the terminology NPS? I always had thought that nominal would be the average point spacing planned before the surveying, and then, there would be the actual (observed) point spacing after the survey. As a conclusion, it would make more sense to me to set the specifications regarding the actual point spacing. Maybe I have a wrong idea of the NPS concept. Tks for your help. –  Andre Silva Jun 13 at 16:00
1  
@NaG - Wes & Barbarossa both provide excellent direction for this task, but from strictly an "ease of use/distribution standpoint" I typically want to produce/receive products that have a resolution in whole numbers - in your case, with an average ground density of 0.67, a 1-meter output seems like an ideal compromise between high resolution, accuracy, and ease of use (for example, I have never received a dataset with 0.67m resolution, but I receive 1, 2, 5 etc all the time). This of course depends on your needs and level of OCD :) –  JWallace Jun 13 at 16:02
1  
I agree with @JWallace. DEM resolution, or DEM post spacing, is typically 1, 2, 3m or larger. I would recommend rounding up from the nearest point spacing (actual or nominal). Look at this page for better explanation. –  Barbarossa Jun 13 at 16:15

Just an update on the issue discussed above, hoping that this would stimulate discussion and could prove useful to others.

For comparative purposes, I produced two bare earth DEMs using as cell size value in one instance the average point value of the whole lidar dataset, in the other case the average spacing of only ground points (rounded up: it was 0.668, I entered the value 1, so obtaining a DEM with a resolution of 1 m)

The procedure was made via arcGIS's Las dataset to Raster.

I am noting a neat difference in favor of the second Dem. Focusing on the same portion of the landscape, I do note an improvement in details and quality of the representation of the terrain. So, I am very happy with the second Dem.

Hope it helps. Needless to say, I am open to suggestions and hints (and also to criticism and further caveats).

Cheers NaG

share|improve this answer

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

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