Previous PageTable Of ContentsNext Page

RTKDGPS and Autodesk Land Development Desktop«; a powerful combination for rapid accurate surveying and land development planning.

I.J. Yule, W.R. Woodgyer and R. Murray

New Zealand Centre for Precision Agriculture,
Massey University,
Private Bag 11-222. Palmerston North, New Zealand.
Phone +64 (6) 3504340, Fax +64 (6) 3505996
i.j.yule@massey.ac.nz

Abstract

This paper provides an example of how modern GPS and GIS technology can be utilised to significantly increase the productivity for surveying in land based applications. The particular example of land subsurface drainage is used.

A Real Time Kinematic Differential Global Positioning System RTKDGPS has been utilised with Autodesk Land Development Desktop« as a means of rapid survey and subsurface drainage system design. The topographic survey is completed with a Trimble AG214« RTKDGPS receiver linked to a Trimble AG170« field computer. This system provided an ideal platform for the capture of accurate topographic data. The system is mounted on a 4WD quad bike which is provided with lightbar driver assistance. Additional sensors can be linked to the system to collect further data, an EM 38« sensor has been used in this case.

Land Development Desktop has many capabilities suitable for professionals involved in land development. It allows analysis to be performed on a wide range of data types relating to land topography and attributes. Initially the system has been used in planning land drainage but is proving to be a versatile tool capable of many other design functions.

The system has dramatically increased the survey output as well as providing a much richer data environment. Previously, a laser was used and measurements taken on a 40 m grid. With the RTKDGPS the survey is completed from a 10m swath width with data being collected every second.

Introduction

Laser technology has previously been the basis for completing detailed topographic surveys required for agricultural field drainage. It is now possible to replace that technology with RTKDGPS in order to achieve significant productivity gains. The new method also offers a richer data environment where information can be used for other purposes and more easily integrated into a GIS system.

A Trimble RTKDGPS system built around the Trimble AG GPS 214 has been used to replace the above laser technology. A full explanation of the system and its performance specifications are given on Trimble’s website (www.trimble.com).

In many instances EM data is also collected in order to give extra meaning to the soil data. Sudduth (2001) stated EM has been used as a surrogate measure for such soil properties as soil salinity, moisture content top soil depth and clay content. In his paper he gave the results of extensive research effort to establish the accuracy of the electromagnetic induction method and factor affecting the accuracy of the instrument. Godwin(2001) also gave a clear explanation of how EM data can be used in assessing field conditions. Both papers contain a large number of useful references in this area of study.

The data acquired from the system is used within the Autodesk Land Development Desktop suite of programs. This software can be used for many design applications dealing with spatial data. In this case the topographic and soils data are used to set design parameters for subsurface field drainage.

Methods

Until recently the initial survey work was completed as the first step towards drainage design was done using a Laser based system. In the last year that system has been replaced with the GPS based system. As a result there have been a number of changes to the design process.

Using laser technology the methodology for completing a drainage survey was as follows:

  • A Spectra Physics EL1 was set up to generate a laser signal. The unit has a range of up to 300m. Where differences in site elevation were greater than 4 – 5 m the unit would have to set up on a second point and a benchmark established to make sure both parts of the survey were consistent.
  • An EL1 receiver was used to take spot elevation points for a 40 x 30m grid. The grid was set up using ranging poles and was manually paced out. This method provided 8.5 elevation points per hectare where the site is reasonably flat and uniform. Where topographic features are visible then these are also logged.
  • Elevation point data was transferred from a field book to spreadsheet. The data was manually drawn onto graph paper and a contour map interpolated by visual means.
  • From this information sections were plotted and a map of the design produced.
  • The productivity of this type of surveying was, 25 – 50 ha surveyed per day, with approximately one week of work to complete the design and prepare costings.

The RTK survey is completed from a 4WD quad bike. GPS and radio antenna are fixed to the rear of the bike while the AG170 computer is placed above the handlebars of the bike in full view of the rider. The touch pad below the screen can be operated from the seat. The lightbar to assist with parallel swathing is placed above the screen. The units are powered by two additional 12 volt batteries which can be easily carried on the bike.

The survey is completed using the following steps:

  • Data is collected every second on a 10 m swath width. Output of up to 8 ha per hour is achieved. This provides 450 elevation points per hectare.
  • Elevation point data is recorded in both “dbf “ and “txt” file formats. Additional sensor data is also recorded. This data can be directly imported into a number of software packages.
  • At present the data is transferred to Trimble Pathfinder Office where it is converted to New Zealand Map Grid. While in this format the data is processed using Vesper(2000) to give a regular grid of points. This is usually completed on a ten meter grid but a 2.5 m has been used on smaller plots where greater detail is required. (other methods could be available to complete this step) A comma delimited file is created.
  • The comma delimited file is then input into Autocad and the surface is generated. A breakline boundary is created (in order to crop contour lines) and the TIN model (illustrated in Figure 1) and contour model (Figure 2) using 0.2 m contours of the data are created. The image can also be provided in 3 D format.

Figure 1 TIN model derived from RTKDGPS survey.

Figure 2. Contour map produced from Krigged date processed through Vesper.

  • EM data can be included as an additional layer on the map. It is generally split into a ten range legend. Vesper is again used to process the data.
  • On the basis of the topographic map and the EM information the drainage design is considered. The EM data helps to define where drainage should take place and where it may not be required. It also provides a great deal of detail in terms of changes in soil type.

The processes that have led to soil formation are very often the result of alluvial deposition rather than erosion. (although this situation is also common) Figure 4 helps to illustrate the intricate link between micro topography and soil texture.

Figure 4a Illustration of Soil EM data showing soil variation.

In the above example the brown areas correspond to lower EM value, Light (yellow and green) medium value and purple high values. The coarser textured soil is in a strong band through the property with finer textures being seen in lower lying areas.

In figures 4A and B alluvial deposition has taken place whereby the courser material has been deposited first in the north part of the block. Towards the southern part of the site more finely textured soils are found. It can be seen that the textural boundaries are also influenced by surface water movement in the form of old creek beds. This is in essence the natural state of the soil. What can also be observed is the influence of human and animal activity on this land. The more finely textured soils are more liable to suffer from compaction, which was observed in this instance.

In the case of Figure 4B the drains would be placed in a northeast – southwest axis in order to achieve the correct drain gradient, the paddock would then be mole drained across the pipe drains in order to achieve a grid pattern. The EM data indicates where moling is likely to succeed. It will also indicate where mole drainage is likely to fail (in the courser textured soils) and should therefore not be attempted.

Figure 4B Southern Section of property.

From considering the whole map a decision has to be taken regarding the general drain layout, once this is done the detailed drainage plan can be created:

  • The cross section and long section of the drainage line can be drawn. This can be done either as a straight line or where appropriate bends and changes in direction of the pipe can be represented. The soil surface is represented on the cross section and two additional lines offset to a depth representing the upper and lower limits of the drain trench are included. Figure 5 illustrates a long section.

Figure 5 Long Section along drainage line illustrating surface profile.(From a further property)

  • The actual sub surface drainage can be put in, this can be started from known invert level of existing drains which are required to connect to the new system. From this starting point drainage lines can be drawn up the section ensuring that they do not go outside the maximum or minimum depth levels. Standard design parameters are used in terms of pipe gradients etc.. In many cases (depending on topography) this will require that more than one main drain is needed. When this is the case care must be taken to ensure that gradients of the drainage pipes are appropriate and that levels also match where pipes intersect. It is to some extent an iterative process. When a drainage line is completed satisfactorily it can be added to the “system’ design.
  • In addition to each drainage line being profiled an overall drainage location plan can be produced for the contractor to operate to. The plan view can include information regarding profile depth at clearly marked points.
  • The navigational capabilities of the RTKDGPS can also be used in setting out (pegging out) the location plan of the system in the paddock. Information and co-ordinates from the plan can be loaded as waypoints to a device such as the Trimble Ag214 RTKDGPS receiver connected to TSC1 data logger. This enhances the accuracy of pegging out and can speed up the process significantly.
  • Additional maps and a complete plan can be given to the property owner. Some of the data, such as EM measurements can be used in terms of precision management of the property if practised.

Conclusions

The system has proved highly successful from a number of stand points. A significant increase in productivity has been achieved and the system has proven to be very reliable and give repeatable results.

Specific achievements have been:

  • Survey time has been cut in half
  • Map processing and design time has been reduced to 25% of original design methods with laser and manually drawing up the plan.
  • The system can be used in all weathers, no manual recording of information is required and receivers are waterproof.
  • The system gives greater degree of process control and quality assurance. For example the drainage contractor has a much clearer, well proportioned plan view to follow.
  • Presentation of the complete plan is much more professional and is sufficiently flexible to be offered in a number of formats.
  • Drainage plans can be easily combined into one larger scheme. So not all surveying has to be completed at the same time.
  • Although the initial capital cost of the equipment and software is high, if sufficient work is generated it is in fact cheaper than previous methods due to the very large labour savings made.

Clearly one of the major drawback to this system is cost. Costs are estimated at (December 2000) :

Initial capital cost of RTKDGPS, with AG170 Field Computer, Trimble MS750 base station, Trimmcom radios and lightbar option.

NZ$145,000

EM 38 sensor form Geonics Canada with polycorder

NZ$25,000

Autodesk Land DevelopmentDesktop

NZ$13,000

A high powered personal computer is also required. The machine used in this study is specified as follows;

1G processor (PC133), with 512Mb of RAM, 32Mb video graphics card. 21’ monitor. It is also best to have access to an A0 printer.

Additional problems are :

  • Trees, there are occasions that trees are in the way and prevent the receiver from getting the signals required to calculate its position. This can be overcome by such devices as laser range finders in order to complete the physical survey, it is less easy to get elevations from such a method. Provided a benchmark can be provided a laser survey would be feasible for a small area.
  • Although the same basic equipment is used to get invert levels the receiver has to be placed in a backpack and a TSC1 datalogger connected. This is mounted on a pole with the antenna on top. This equipment has to be removed from the quad bike in order for this to happen. The weight of the backpack and pole is an issue when using it over long periods.
  • It is a reasonably complex system and some effort is required to learn how to use it effectively.

References

Godwin, R. (2001) Field Condition Mapping Technologies. SMART FARMING II: Workshop on Automation for Agriculture. 13 – 15 March 2001, Putrajaya, Malaysia.

Sudduth, K.A. Drummond, S.T. Kitchen, N.R. 2001. Accuracy issues in electromagnetic induction sensing of soil electrical conductivity for precision agriculture. Computer and Electronic in Agriculture, 31 (2001) 239 – 264.

Vesper (2000) Variogram Estimator and spatial Predictor with Error. Software available from the Australian Centre for Precision Agriculture. University of Sydney. http://www.usyd.edu.au/su/agric/acpa/pag.htm

Previous PageTop Of PageNext Page