SCANNING LASER PAGE


N.B. the scanning laser data shown have been subset out from the original >100MB textfiles

1. Below : Data from file NORTH2.TXT. Projection UTM-N13, spheroid Clark 1866, datum NAD27 (CONUS). This is an image of a 0.5m raster surface interpolated from the XYZ point data for a 200m x 200m study area centred on 106 deg.52.16'W, 32 deg. 39.04'N. The interpolation used only three data points, resulting in a rather smooth DEM image. Note that the DEM is based on elevations which include LIDAR returns from the mesquite bushes, as well as from soil!.

NORTH2 PLOT

2. Below : Raster data from NORTH1.TXT (left) and NORTH2.TXT (right), both interpolated from point XYZ data. Vector data are contours generated from NORTH2.TXT (left) and points classified as "low vegetation" (from file v1nom.txt). Notes :

(i) area shown in leftmost image corresponds to the area shown in the previous image (above).

(ii) the area shown in the rightmost image corresponds to the area defined by the white AOI box in the leftmost image.

(iii) there is some misregistration between data points from NORTH1.TXT and NORTH2.TXT (original datafiles).

(iv) the anomalous division of the study area by a (non-existent) drop in elevation from E to W; this is thought to be owing to the difference in the elevations calculated from 2 different flight lines. The points cover (sequentially) first the left half of the rightmost image in a downward or southern direction and then the right half in an upward or northern direction. There is a third batch of points which covers the entire area with an even distribution of points, from the S to the N (bottom to top of image). The dark and light features protruding from the central division are artifacts of the interpolation where there is little or no data (in between 2 flight lines / scans?).

(v) the clumped distribution of points on the W-most scan appears to more closely resemble the actual distribution of the mesquite (on top of the dunes). The "fan" distribution on the E side (scanned from S to N) appears to bear no relation to the vegetation distribution.

SAMPLES OVER RASTER DATA

3. Below : an expanded view of the large dune in the north-centre of the area, in which the 0.5m raster cells are clearly visible.
The points are (a) unfiltered (?) from file NORTH2.TXT and (b) filtered or "classified" as "low vegetation" (in this case mesquite).

ZOOMED

4. Below : broad comparison between spatial distribution of vegetation from airborne video (mesquite, in green) and topography from scanning laser (right). Note that the DEM is based on elevations which include LIDAR returns from the mesquite bushes.

1995 Video and 1998 LIDAR

5. Below : as above but with the DEM detrended (i.e. the broad slope has been removed, leaving only the protrusion of the coppice dunes above a virtual slanted surface). The dunes and the depression of the road running roughly N-S are clearer in this DEM image.

ditto, DEM detrended

6. Below : superposition of mesquite boundaries over detrended DEM for the entire study area DEM. Note : range is actually 4347 - 4383 feet rather than the numbers burned in below!!

mesq./DEM

7. Below : pseudo-3D view of the study area, showing the mesquite distributed mostly on the sides and tops of the coppice dunes. The video stills do not cover the green boundary to the S and E. A better version to follow (using Erdas Imagine).

3DView


Back | Disclaimer