5

I have about +100 points of various descriptions from an archaeological survey. They are the backlog from years of excavations BEFORE systematically using Total Station at the site. For each artifact I have the upper and lower elevation data, in addition to the distance from either the NE, SE, SW, or NW point of the grid square. The location was recorded using two meter tapes from two points intersecting at the point of the artifact. The Excel file I have records the grid corner point and the distance from that corner to the artifact (ID1 and Radius1, ID2 and Radius2).

First, I need a method that will help me systematically upload the points. How do I arrange my Excel file, and what do I do in QGIS to get it plotted properly?

Second, I want to be able to answer questions of spatial relations based on elevations.Is there something special that I need to do for the input of the data in the .csv file?

I'm new to QGIS.

|improve this question
2

Your problem looks like a bilateration issue.

This way it works for me:

We start of with the corner coodinates in a projected CRS (here: EPSG:31466):

file corners.csv

corner,E,N
A,2530114,5725986
B,2530338,5726216
C,2530525,5726401
D,2530655,5726370
E,2530917,5726264
F,2531381,5725999
G,2531197,5725785
H,2530698,5725334

and the points and measured distances, one line for every distance and two for every point:

file pointdist.csv

point,corner,dist
1,A,146
1,B,211
2,B,134
2,C,158
3,A,369
3,H,516
4,A,425
4,B,303
5,D,192
5,E,120
6,G,203
6,H,563
7,H,125
7,G,586
8,E,216
8,F,462
9,F,163
9,G,166

Load corners.csv and pointdist.csv with Add Vector Layer

Rightclick on the pointdist layer Properties->Joins

Join layer: corners
Join field corner
Target field: corner
Save As... pointcornerdist.csv

Load pointcornerdist.csv as Delimited text and EPSG:31466

remove the other two csv layers

Rightclick -> Set Project CRS from Layer

Settings->Snapping options: pointcornerdist.csv 10 pixels

Create a New Shapefile Layer of type polygon EPSG:31466 name site.shp

Draw a polygon along the corner points

Rightclick on the polygon layer -> Properties->Style: Simple fill : No brush

Vector->Geoprocessing->Buffer(s)

Input vector layer: pointcornerdist
Segments: 20
Buffer distance field: dist
Output shapefile: buffers.shp
x Add result to canvas with EPSG:31466
Close

Vector->Geometry Tools -> Polygons to lines

Input Vector layer: buffers
Outpt shapefile: bufferlines.shp
x Add result to canvas with EPSG:31466
Close

Remove layer buffers

Vector->Analysis Tools-> Line intersections

Input line layer = Intersect line layer: bufferlines
Input unique ID field = Intersect unique ID fiels: point
Output Shapefile: intersect.shp
x Add result to canvas with EPSG:31466
Close

Rightclick on intersect.shp -> Filter

Click on point_1 = and point_2

Vector->Geoprocessing Tools -> Clip

 Input vector layer: intersect
 Clip layer: site
 Output shapefile: points.shp
 x Add result to canvas with EPSG:31466
 Close

Re-Add corners.csv, this time as Delimited text and EPSG:31466

label points layer with point_1, and corners layer with corner

and you are done: enter image description here

|improve this answer
  • Thanks Andre for your quick reply. Yes, someone suggested this, but what you're suggesting seems like I would have to do this for each point. Is there a systematic method I can apply to the whole dataset? If so, can you please include a specific directive? Thank you! – JPDA Sep 16 '14 at 12:16
  • Unfortunately, your points may have been measured to any two of the four edge points. If you group them by common reference points, you could try to do the trilateration mathematically. See gis.stackexchange.com/questions/48937/… – AndreJ Sep 16 '14 at 12:51
  • Hi Andre. Thank you for following up... I don't have to guess which two edge points the measurements were taken from. Does that change the method? – JPDA Sep 16 '14 at 13:14
  • No. The only point of guessing is if both possible intersections are within your research area. – AndreJ Sep 16 '14 at 13:17
  • When you say "group them by common reference points", do you mean that I should create a separate csv file for the artifacts that are measured from the same two points, and then load that csv file? – JPDA Sep 16 '14 at 13:20

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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