Post

2 followers Follow
0
Avatar

2005 lidar point data

Hello, I expect you know that the 2005 lidar gridded datasets have a processing error that results in multiple artefacts - see image. Many years ago I was informed that the original, 2005 point data were erased which explains why it cannot be downloaded from the Defra site.

Can you confirm that the point data have definitely been erased, no longer exists anywhere, on any stray tape(s), in some academic data store, etc.? I ask because in many areas the 2005 data is all that is available, i.e. I would grid the point data to get a clear dataset.

Thank you and regards. Steve Kaye

Wumptid //

Please sign in to leave a comment.

4 comments

0
Avatar

Hi Steve, 

Thank you for your question, please can you provide us with the OS grid reference for the image you attached and we will get back to you as soon as possible.

Many thanks

Ella Fotheringham 

Environment Agency 

Ella Fotheringham 0 votes //
Comment actions Permalink
0
Avatar

Hello Ella,

The buildings in the rough centre, alongside the N-S road, are located at ...

OS X (Eastings)     234767
OS Y (Northings)    81360
Nearest Post Code   PL15 9NA
Lat (WGS84)         N50:36:31 (50.608497)
Long (WGS84)        W4:20:12 (-4.336586)
Lat,Long            50.608497,-4.336586
Nat Grid            SX347813 / SX3476781360

The 2005 DTM 1m grid is SX38. You can look in any 2005 grid to see the same aberrations. The image supplied is 'simple local relief' over coloured slope but the same effects are apparent in any filtered imagery, for example, sun shaded view will show them as etched contours.

To be clear, this is not a new problem; it was introduced when the original points were gridded. I suspect it is a scaling issue related to going from, say, an original grid of 0.5 m resolution to one of 1 m (or something like that). The problem seems not to have been noticed and then later the points were deleted/erased from storage media - apparently a cost-saving measure. At least, that is what I remember being told years ago.

I'm asking if the points might still exist somewhere because 2005 data is all that is available for large areas of the SWest and it is old enough to show some features in the land surface that the large agricultural machines have/are removing, i.e. the data themselves are historically important.

Regards, Steve Kaye

Wumptid 0 votes //
Comment actions Permalink
0
Avatar

Hi Steve,

The issue is that our timestamped tiles, which you have downloaded, are provided in Millimetres (rather than metres).

Here’s a solution that should solve your issue:

Divide the raster by 1000 to convert mm to m,  then re – compute the hillshade.

We are aware of this issue and currently in discussions to have the mm acsii archive replaced with geotiffs in metres.

I hope this helps, let me know if the problem persists.

Many thanks

Ella Fotheringham

Environment Agency 

Ella Fotheringham 0 votes //
Comment actions Permalink
0
Avatar

Hello Ella,

The issue is not mm or m. Nor is it due to a conversion between the two. The following is of small area of dtm_V0040624_20041219_20050220_mm_units.asc (a 2005 tile within SX38). It was produced within SAGA, converted to m and then sunshaded. The artefacts are clear.

 

As I wrote earlier, the artefacts are due to a processing error introduced during the creation of the 2005 grids. If you create new geotiffs from the original grids then the artefacts will remain.

You can only produce 'clean' grids by processing the original point data but, as these reportedly were deleted, you need to find a set of points from somewhere either within Env. Agency (Geomatics) or hope that some external agency/Uni./utility/company purchased all the points and still has them.

Some detective work is required.

By the way, it may be true that other point clouds from other years were also deleted but I don't recall.

Hope this helps.

Regards, Steve Kaye

Wumptid 0 votes //
Comment actions Permalink