Lab 2 - Coordinate Data, Projections, & Transformations
Contents
1. Introduction
1.1 Bear Valley Creek Site
1.2 Objectives
2. Unprojected Data
2.1 Methods
2.2 Results
3. Projected Data
3.1 Methods
3.2 Results
4. Transformed Data
4.1 Methods
4.2 Results
5. Conclusions
5. References
1.1 Bear Valley Creek Site
1.2 Objectives
2. Unprojected Data
2.1 Methods
2.2 Results
3. Projected Data
3.1 Methods
3.2 Results
4. Transformed Data
4.1 Methods
4.2 Results
5. Conclusions
5. References
Introduction
Bear Valley Creek Site
The purpose of this lab was to gain a greater understanding of coordinate data, coordinate systems, and projections by working with unprojected total station survey data. The survey data comes from the larger Columbia Habitat Monitoring Program, and the Integrated Status & Effectiveness Monitoring Program. The larger project is funded by Bonneville Power Administration. Bear Valley Creek is one of the areas monitored as part of the larger study and is the focus of this lab.
Bear Valley Creek Site
The purpose of this lab was to gain a greater understanding of coordinate data, coordinate systems, and projections by working with unprojected total station survey data. The survey data comes from the larger Columbia Habitat Monitoring Program, and the Integrated Status & Effectiveness Monitoring Program. The larger project is funded by Bonneville Power Administration. Bear Valley Creek is one of the areas monitored as part of the larger study and is the focus of this lab.
Lab Objectives
1. Understand coordinate systems and transformations between them
2. Plot data from an assumed coordinate system
3. Transform assumed coordinate system data
4. Explain methods
1. Understand coordinate systems and transformations between them
2. Plot data from an assumed coordinate system
3. Transform assumed coordinate system data
4. Explain methods
Unprojected Data
Methods
Datapoints that were collected at the Bear Valley Creek site were collected on an assumed Cartesian coordinate system. With an assumed coordinate system an arbitrary coordinate is chosen and then a direction nearby to that point is set as due north (see points cp1 and BS on the map below). The benefit with using total station survey data and an assumed coordinate system, is that errors associated with collecting data using GPS are avoided. Unfortunately the data lacks a geospatial reference. So, in order to later project the data, control points and bench marks were collected.
For Task 1 I mapped the unprojected data by importing several layers into ArcMap, which included control points, topo points, hard breaklines, and soft breaklines. The data was in an assumed coordinate system, which came to my attention when an error message came up in ArcMap stating the data lacked a spatial reference. The Cartesian coordinate system plots data on an x,y,z axis, but there was no real reference provided, further shown under Data Frame Properties which said the coordinate system is unknown.
To emphasize that this is unprojected data, I symbolized and labeled all control points and benchmarks in such a way that differentiated them. The data is relative to the first control point, and I decided to add a 25 meter increment grid to give some spatial information, even though location is not known. The inset map and compass convey the inability to locate the data spatially in the real world. For more detailed instructions on how to import this unprojected data into ArcGIS, go to Joe Wheaton's website.
Methods
Datapoints that were collected at the Bear Valley Creek site were collected on an assumed Cartesian coordinate system. With an assumed coordinate system an arbitrary coordinate is chosen and then a direction nearby to that point is set as due north (see points cp1 and BS on the map below). The benefit with using total station survey data and an assumed coordinate system, is that errors associated with collecting data using GPS are avoided. Unfortunately the data lacks a geospatial reference. So, in order to later project the data, control points and bench marks were collected.
For Task 1 I mapped the unprojected data by importing several layers into ArcMap, which included control points, topo points, hard breaklines, and soft breaklines. The data was in an assumed coordinate system, which came to my attention when an error message came up in ArcMap stating the data lacked a spatial reference. The Cartesian coordinate system plots data on an x,y,z axis, but there was no real reference provided, further shown under Data Frame Properties which said the coordinate system is unknown.
To emphasize that this is unprojected data, I symbolized and labeled all control points and benchmarks in such a way that differentiated them. The data is relative to the first control point, and I decided to add a 25 meter increment grid to give some spatial information, even though location is not known. The inset map and compass convey the inability to locate the data spatially in the real world. For more detailed instructions on how to import this unprojected data into ArcGIS, go to Joe Wheaton's website.
Results
To download a pdf version of this map please click on the image or click here.
Projected Data
Methods
In step one the data could not be spatially analyzed because it did not have a spatial reference to project it on to a map. This section will describe how to transform unprojected data to projected data. In order to transform the data I used the benchmarks, which were collected with total station and GPS, to project the data into NAD 83 UTM Zone 12 N. Data can be transformed in four possible ways, including changing the scale, skew, rotation, or translating the data. The GPS coordinates were used to perform an affine transformation, which just means simply shifting, rotating and performing a datum adjustment. There are two other types of transformations in addition to affine that can be performed if you have control points; similarity and projective. The reason a similarity transformation was not used was because it would not allow for automatic scaling or skewing, and a projective transformation requires a fourth control point, which we did not have. The figure below better explains how to perform an affine transformation and can found on Joe's Wheaton's website.
Methods
In step one the data could not be spatially analyzed because it did not have a spatial reference to project it on to a map. This section will describe how to transform unprojected data to projected data. In order to transform the data I used the benchmarks, which were collected with total station and GPS, to project the data into NAD 83 UTM Zone 12 N. Data can be transformed in four possible ways, including changing the scale, skew, rotation, or translating the data. The GPS coordinates were used to perform an affine transformation, which just means simply shifting, rotating and performing a datum adjustment. There are two other types of transformations in addition to affine that can be performed if you have control points; similarity and projective. The reason a similarity transformation was not used was because it would not allow for automatic scaling or skewing, and a projective transformation requires a fourth control point, which we did not have. The figure below better explains how to perform an affine transformation and can found on Joe's Wheaton's website.
I used bm3 as the point to perform the transformation around. For this step I had to download the Champ Transformation Tool,
and enter in benchmark data. I had to select a point to rotate the data
on and ultimately chose benchmark 3 (bm3) and
GPS 2 as the rotation for two reasons. The residual error calculations
seemed to be lowest (as seen below) and the data seemed to visually line
up with the creek the best. For these reasons, the affine
transformation seemed to work successfully. For more detailed
instructions on how to perform this transformation please refer to the WATS 6920 website.
Residual Error Calculations:
Hinge Bearing Bearing_dH Bearing_dZ Other_dH Other_dZ
1 GPS2 -1.19 1.46 -0.97 1.26
1 GPS3 -1.14 0.22 -1.96 0.87
2 GPS1 1.19 -1.46 0.22 -0.2
2 GPS3 -0.03 -0.04 0.9 -1.71
3 GPS1 1.14 -0.22 -0.82 0.65
3 GPS2 0.03 0.04 0.92 -1.67
Residual Error Calculations:
Hinge Bearing Bearing_dH Bearing_dZ Other_dH Other_dZ
1 GPS2 -1.19 1.46 -0.97 1.26
1 GPS3 -1.14 0.22 -1.96 0.87
2 GPS1 1.19 -1.46 0.22 -0.2
2 GPS3 -0.03 -0.04 0.9 -1.71
3 GPS1 1.14 -0.22 -0.82 0.65
3 GPS2 0.03 0.04 0.92 -1.67
Results
I attempted to keep symbology and labeling consistent so the difference between projected and unprojected data could be observed.
I attempted to keep symbology and labeling consistent so the difference between projected and unprojected data could be observed.
To download a PDF version of this map please click on the image or click here.
Below are both maps laid side by side to better illustrate the difference between an unprojected and projected map. The projected map is shown on the left, and the unprojected map is on the right.
Transformed Data
Methods
I decided to also provide an interactive map of Bear Valley Creek in order to visualize the data at the scale of your choosing. I exported my study site to a KMZ and then imported the data into Bing maps. Detailed instructions can be found on Joe Wheaton's website.
Methods
I decided to also provide an interactive map of Bear Valley Creek in order to visualize the data at the scale of your choosing. I exported my study site to a KMZ and then imported the data into Bing maps. Detailed instructions can be found on Joe Wheaton's website.
Results
To view larger map at source website, click here.
Conclusion
ArcGIS is a useful tool because it automatically projects data for you. Unfortunately, this useful transformation can be misused and lead to the creation of incorrect maps. ArcMap automatically projects data based on the first data layer imported, so your projected data may be transformed to a geographic coordinate system without you realizing because the changes may not be perceptible right away. This lab showed that it is necessary to transform data when you don’t have a real world coordinate reference, and want to have more control over your data.
ArcGIS is a useful tool because it automatically projects data for you. Unfortunately, this useful transformation can be misused and lead to the creation of incorrect maps. ArcMap automatically projects data based on the first data layer imported, so your projected data may be transformed to a geographic coordinate system without you realizing because the changes may not be perceptible right away. This lab showed that it is necessary to transform data when you don’t have a real world coordinate reference, and want to have more control over your data.