Conducting a Topographic Survey with a Dual-Frequency GPS, and with Total Survey Station

Introduction

Topography, in a very broad sense is the study of surface shapes and features.  Understanding of topography, can be useful when mapping both natural and artificial features in a landscape.  Topographic maps and data can be useful to a variety of different fields ranging from ecology to construction.

We have conducted a topographic survey on the study of elevation on the Eau Claire Campus Mall.  In the last two weeks of class we have learned two different ways of collecting topographic data.  The first way we learned was with use of a survey grade dual frequency GPS.  The second way was by using a topcon total survey station.

The surveys this week tie in many of the concepts that we have been learning this semester.  The survey using the dual-frequency GPS relies on the same concepts we put to use during our excercise in microclimate mapping.  The survey using the total station relies on the same concepts used during the distance azimuth survey.  The interpretation of the data that we collect from each of the survey will rely on knowledge of interpolation from the terrain creation and survey excercise.

One objective of this activity is to learn  practical skills, such as how to connect set up the equipment used for each of the surveys.  Another is to think critically about the advantages and disadvantages of each of the surveys, and to determine what considerations you would take when choosing between them.

Study Area

The study area is campus mall located between the Library, Schofield Hall, and Davies Student Center on the UW Eau Claire Campus located at 105 Garfield Avenue, Eau Claire, WI 54702.  The campus mall is estimated to be about 1 hectare in area.  The campus mall was recently Redeveloped in 2013 as part of the Centenial Hall project.



Figure 1.  Location of Campus Mall shown above.  Map from Esri

Figure 2.  Photo of Campus mall as seen from balcony of Davies Student Center, photo originally from fieldmethodscumminbk.blogspot.com

 

An aspect of this study area that I found interesting was it's adjacency to the steam Little Niagra.  Thearea of land next to the stream on either side is reffered to as the riparian zone.  Critics of past redevelopment projects say that that the stream has been left more exposed to runoff because of lack of trees and other vegetation in the riparian zone.  Although it is not the focus of our project today, it could be interesting research to examine the effect that the topography of the campus mall has on runoff and stream pollution.

Figure 3.  Students shown collecting aquatic invertebrete samples from little Niagra, with the Daivies center in the background.  Picture originally from uwec.edu

Methods

Conducting a topographic survey with a dual frequency GPS.
During classtime we learned how to set up a project using the topcon tesla unit.  In class our group created a job on the tesla unit and named it geog336_topotest_group2.  When we went out to collect the data points we created a new job and named it geog336_group2.  Important concerns when using the tesla were to never try to do a "hard shut down" or powering off by continuously holding down on the power button, because this is very bad for the unit, and causes numerous problems.


Figure4.  Topcon Tesla unit shown above

The tesla connects to the topcon hiper, which is the dual frequency survey grade GPS.  The hiper attaches to the top of a tripod of a known height.  The height is calculated in the Tesla so that the distance from the ground can be accounted for. 

Figure 5. Topcon hiper shown above

The hiper is particularly accurate because the data points are post processed by CORS (Continuosly Operating Reference Station).  This post processing can enhance the precision of the points to within a few centimeters of the true location. 

Figure 6. A screenshot of a map that shows the location of CORS in the United States.  Map can be viewed at http://www.ngs.noaa.gov/CORS_Map/

At each survey point, the tripod is leveled and then a button on the screen is pressed to start collecting points.  In this excersize we had it set up so that at each location it would collect 5 points and than take an average of these points to store. 

 

Our group had three members, so we were instructed to collect 150 data points, or 50 points per person.  Due to scheduling conflicts I collected my 50 data points at a seperate time then my group members.  In the tesla I opened up the job that they had created, and when it was connected to the blutooth from the Mifi unit, I collected 50 data points.  I tried to get the data points to be representative of the area, making effort to collect more data points in places where there was more variability in elevation.  At each of the points it was important to get the tripod level, or else the data would be very inaccurate.  There was a bullseye bubble level attached to the tripod which could be used to determine if the tripod was level enough to collect a point. 

Figure 7. Example of a bullseye bubble level

Conducting a topographic Survey with the Topcon Total Survey Station
In a similar way to the last survey, we learned the techniques for how to use the Topcon Total Survey Station in class and then in our groups conducting the survey outside of class.  

Our group began by setting out flags to mark our occupy point and our backsight points.  These points were taken using the hiper and a tripod in order to get an exact measure for these points.  Similar to the distance azimuth lab, when we know the location of one of the points, we can use this to also get highly accurate measurements using a laser and a reflecting surface. 

The Topcon Total Survey station was set up to be directly over the occupy point.  It took a while to get it level, and to where the laser pointer was close to the placement of the flag.  After we got this level, we connected the tesla to the total survey station, making sure that we first turned the hiper off.  Once this was completed we collect data points.

Data points were collected by aiming the laser of the topcon total station at the total station reflector.  Since the exact location and elevation is known at the topcon total station, it can use the distance of the angle, the consistent height of the reflector (2 meters), and the angle at which the laser is aimed to calculate these data points and store them on the tesla unit.  One group member would move around the campus mall with the reflector, and one would stay at the topcon total station to aim the laser and collect the data points. 

During this survey we were not instructed to take a specific number of points, but instead were asked to collect enough information to represent the topography of the area.  We ended up collecting 53 points, deciding that since the campus mall was gently sloping at a relatively consistent pace, we would get a representative topography as long as the data points were spread out far enough.

Transferring the data to a text file, and then mapping x,y,z coordinates in ArcMap
We would like to thank Martin Goettl for making a video tutorial that explained how to export the data on the tesla as a text file, as well as for all of his other help in our various activities.

In the tesla unit, we opened our job, and then went to exchange.  We exported the data as a text file, going through the steps to make sure that we had labeled it in a way that we could find it on the computer, and making sure that the data was in the corect format in terms of coordinate system, and making sure that east was listed first in the text file, as well as making sure that there were commas between all of the data points.

 

Figure 8. On the left is shown the textfile of the topographic data from the topcon total survey station, and on the right is shown data collected using the dual frequency survey grade GPS.

When the data was all exported as text files and saved in the appropriate folders, I exported the text files into ArcMap.  I right clicked on the tables and went to Display XY data for each of them.  When the Display XY Data window appeared I made sure that the X Field was set as Lon(East), the Y Field was set as Lat(North) and that the z field was set for our elevation (Htt(G)).

Figure 9.Display XY data window shown on left and resulting plotted points shown on the right

 

 

When producing rasters via interpolation, I chose the natural neighbors method.  I chose this over others because I knew from previous excersizes that this interpolation method worked well when mapping gently sloping surfaces. 

 

Results

 

After producing raster interpolations of the elevation data, I opened the rasters in ArcScene so that I could look at them 3-dimensionally, and compare their characteristics.  In the interpolation made from the topography points from the GPS, there is a relatively smooth slope with minor dimples.  In the interpolation made from the topography points from the total survey station there is a similar slope, but also an irregular spike (see figure 12) that might indicate some mistake was made during data collection. 

Figure 10.  Raster Interpolation by Natural Neighbor method of GPS topogrphy points shown in ArcScene

Figure 11. Raster Interpolation by Natural Neighbor method of total survey station topography points shown in ArcScene

 

Figure 12.  Close up of the irregular spike seen in Raster interpolation of total survey station topogrphy points

One way to make a comparison between two different rasters is to look at the histograms.  In viewing the histograms, the raster made with the GPS points seems to be more stretched in comparison with the histogram of the total survey station points.  The differences could be a result of many things, such as accidently surveying more in different spots during each survey, making more errors in one survey in comparison to another, or simply by having less data points in one survey compared to the other.  

Figure 13. Comparison of histograms of the produced Rasters.

When we look at a map produced showing a DEM (Digital Elevation Model) of the campus mall, we can see that there is about a 4 foot change in elevation across the UWEC campus mall.  The highest elevation can be seen in the northeast corner of the map towards Schneider and Centennial Hall.  The lowest elevation can be seen in the southern part of the map as the mall gets closer to the Little Niagra.

Figure 14. Map showing elevation variation on UWEC campus mall

 

The reason that the data from the GPS survey was chosen in producing the map, is because the data from the total survey station did not appear in the right place, indicating that we must have had some significant errors in our methods when using the total survey station.

Figure 15. Map showing inaccuracy of the data points collected with the total survey station.

Discussion
Looking back there were probably things that we could have changed about our methods that would have resulted in better data.  For the total survey station, we probably should have taken more time to make sure that we were as precise as possible when we were setting everything up, this might of prevented everything from becoming as skewed as it did.

An additional technique that I think might have had some benefit is walking a predetermined distance between each point that is being collected.  As it was, we collected data rather haphazardly, which might have meant that we accidently missed parts of the mall.  I think we could have made a better model if we had tried to survey at consistent distances, and then just increased the number of points that we took when we saw that there was a specific area of high variability.

Between the two methods, I can see that there are benefits of each of them, while greatly prefering the survey grade GPS method.  The total survey station was difficult to set up, and easy to make errors that could ruin all of your other data points.  In the GPS method, it was still possible for human error to make data points inaccurate, but it also was also more likely that the error would be limited to a single point, and wouldn't affect the entire survey.  This being said, it is largely about human error, and with more practice and familiarity with the total station this might not be as large of an issue.

Other factors that would make a difference is ease of use between the two methods would be the composition of the terrain you are surveying.  If were trees, or other things that provided visual blocks, then the total survey station would have taken much more time, while the GPS method would have been largely unaffected.

The only real benefit that I see to the total survey station would be in a scenario where the GPS is not working or is in some way unavailable to you. Then using the total survey station would be much more practical.  I think that another good question to ask is if or when both of these survey methods might be proven obsolete by a UAS with Lidar.

Conclusion

This exercise definitely kept us learning new things, but was also very useful in bringing in skills that we had learned in previous activities (such as interpolation and using ArcScene).  This lab was  also useful in that we were able to exposure to different methods of doing a similar survey.  Depending on what you are trying to gain from a survey, and what resources are available to you, it is necessary to be flexible to multiple methodologies.  In a field with technology that changes quickly, flexibility and capacity to learn new skills are necessary to remain efficient and effective at your work. 

Conducting a Distance Azimuth Survey

Introduction
This week I missed the actual field activity because of involvement with a research project in another class, so much of my understanding of the activity comes from the posted description of the assignment, as well as talking to classmates and looking at past blog posts.  The purpose of the activity was to learn how to map locations of features without use of a GPS, using the concept of slope distance and azimuth.

The Azimuth compass starts at 0 degrees as true north, 90 degrees is east, 180 degrees is south, and 270 degrees is west.  Slope Distance takes into account the curvature of the Earth in its measurement, making it more accurate then normal distance.  To conduct a distance azimuth survey, you have a central location (the position of the surveyor) and then you collect information on the distanance of each point from it, as well as the azimuth degree.  You can take measurements of the locations of many different features as long as you have a way to collect this information. Knowing the precise location of surveyor is one of the most important peices of information in the survey. If this point is off by even half of a meter, the same error will be transferred to all of the other points. 

This type of survey is useful because in the field you need to be prepared for a variety of situations.  If you are stuck in a situation where you are not able to use GPS, this method is a simple way that a relatively accurate survey can be completed. 

Study Area
The study area is the a section of the UW Eau Claire Campus located at 105 Garfield Avenue, Eau Claire, WI 54702.  Since I was gone for the activity, I emailed the members of the group that I was listed in on D2L.  The groups were adjusted, but I contacted Professor Hupy who said it was fine to use data collected from Aaron Schroeder's group.  This groups used two stationary surveyor points so that a large area of the northern part of the campus mall could be surveyed.


Figure 1.  Map of study area

Methods
A TruPulse 360 Laser Rangefinder was used .  The device allows you to take a reading of a stationary object and immediately get its slope distance and azimuth from the starting location.

Figure 2. Picture of a TruPulse 360 Laser Rangefinder

The two stationary places were chosen because they were determined to be places that could be easily seen easily in an aerial photo, and therefore easy to find the correct location. A tripod was used to set up the TruPulse in the stationary places.  Features were shot, and the distance, azimuth degree, and type of feature were recorded and later placed into an excel document.



Figure 3.  Portion of the distance-azimuth degree survey excel document

One concern that is necessary to take into consideration for distance-azimuth degree surveys is magnetic declination.  A video from the National Geophysical Data Center shown below is helpful for understanding what magnetic declination is.

Essentially, Magnetic declination is the difference between magnetic north and true north.  The NGDC has magnetic field calculators that can be used to determine the amount of magnetic declination in your geographic location at the time of the survey.

Figure 4. Magnetic Declination shown for the Eau Claire area

The next part of the activity would be mapping the location of these features in ArcMap.  Before doing anything else in ArcMap, the first step was to create a geodatabase which I named distance_azimuth.  This was important for being able to organize and keep track of all the data and layers produced.

I tried using the Bearing Distance to Line command to import the data into ArcMap, but I got error messages when I tried doing this.  I looked at other blogs and found that some of the blogs mentioned using a Table to Table Command to before using the Bearing Distsance to Line Command.  The Table to Table Command, found in the ArcToolbox under conversion is a way to take an excel worksheet and turn it into a geodatabase table.


Figure 5. ArcMaps's description of the Table to Table  tool

When I had the data as a table in ArcMap, I was able to use the Bearing Distsance to Line Command found under data management tools under features.  This allowed me to take the data in the table and plot it as lines on a map.


Figure 6.  ArcMaps's description of Bearing Distance to Line command
 

When I first performed the bearing to line command, I set the x field as new x and the y field as new y.  I had thought that these were the most updated values, but when I looked back at the excel document I saw that these rows contained values with less decimal places (refer to Figure 3.)  This meant that these values were not precise enough which is why I got values that were in the wrong areas, as I am reletively sure that the group did not set up a tripod in the river.

Figure 7. Input of the Bearing Distance to line command shown on the left, output of the command shown on the right.

After checking the table I realized that I should use the origin x and origin y values instead, and when I used these the surveyor points moved to more realistic areas.

Figure 8. Input of the Bearing Distance to line command shown on the left, output of the command shown on the right

 The next step was to get points by turning the vertices of the lines into points.  This was done using the Feature Vertices to Point too under data management under features. 

Figure 9. Input of the Feature Vertices to Points command shown on the left, output of the command shown on the righ

 Results

Using the layers generated in the methods section it is possible to make a map that overlays the recorded location of the features with an imagery background.  Below are two maps, one that show the locations of just the feature points, and one that shows the locations of the feature points with the lines. 

Figure 10.  Location of features collected from survey 

Figure 11. Location of features shown with lines.

 

 

It is also easy to generate other maps that contain informations about the feature points collected.  If a join is performed between the points feature that was generated, and the original survey table, than the information can be used to make maps on the locations of features with certain characteristics.  Two examples of this can be seen below.


Figure 12. Map showing the location of trees in the survey area

Figure 13. Map showing the location of lightpoles in the survey area.

Discussion
The biggest issue that I saw when making maps using this data, was that there was a large level of inaccuracy.  For example,the picture underneath show an aerial view of zorn arena.  When I looked back at the table, the points that you can see on top of zorn were listed as being taken on the edges of the building.  One of the furthest points in was close to 20 meters away from the edge of the building, meaning that in some areas the measurements were very noticably off.

Figure 14.  Aerial view of Zorn arena with feature points from the survey.

Since I was not present for the actual data collection, I don't feel like I can speculate how much of it came down to things like human error or miscalculation.  I would say however, that the points were at least relatively close, and if the features being measured were larger, the proximity to the actual locations probably wouldn't be as apparent.  Also I would assume that with more practice, the errors that arise when using this method would be decreased. 

There was another concern that I saw when reviewing other people's blog posts, and that is the concern that this technique, at least in the way it was used in class might not be that useful, because it relies on an expensive laser to work, and therefore is less likely to be useful in an emergency situation where a GPS isn't working. However, the same method could still be used with a different way of measuring distance and angle, so it is still useful to learn the process even if you don't have access to this specific technology.
Conclusion
This week's assignment was different for me because I was not present for the actual field activity.  Despite this, I still feel like I was able to learn about the surveying methods and gain an understanding of how it could be useful.  The assignment was well explained, which helped greatly when I was going through the indivdual parts of the assignment, and reading through blog posts from last year was also very useful.

Understanding how to conduct a distance azimuth survey is a useful skill to have in circumstances where you do not have access to a GPS.  It may be more time consuming, and more difficult to get to a similar level of accuracy, but it is still a good skill to have when out in the field collecting data.