Submitted to: Applied Engineering in Agriculture
Publication Type: Peer reviewed journal
Publication Acceptance Date: 4/3/2004
Publication Date: 2/17/2005
Citation: Allred, B.J., Fausey, N.R., Daniels, J.J., Chen, C., Peters, L., Youn, H. 2005. Important considerations for locating buried agricultural drainage pipe using ground penetrating radar. Applied Engineering in Agriculture. 21(1):71-87. Interpretive Summary: Finding buried agricultural drainage pipe is an important problem confronting farmers and land improvement contractors in the Midwest U.S. Ground penetrating radar (GPR) may provide a solution. Results from ongoing research has found GPR to be successful in locating on average 72% of the total amount of pipe present at thirteen test plots in southwest, central, and northwest Ohio. The effective use of GPR for drainage pipe detection requires careful consideration of computer processing procedures, equipment parameters, site conditions, and field operations, all of which were addressed in some detail during this investigation. Some of the more important results obtained include the following. 1) Application of a low frequency noise reduction filter along with increasing the radar signal strength were the computer processing steps found to be most helpful for enhancing the drainage pipe response exhibited within GPR images of the soil profile. 2) GPR maps required additional computer processing in order to show the subsurface drainage system present. 3) A GPR unit having an antenna frequency of 250 MHz seemed to work best for detecting buried agricultural drainage pipe under conditions typical in Ohio. 4) Within limits, increasing the distance between measurements and reducing the amount of data collected at each measurement point still produces good results, while at the same time increasing the speed at which a GPR survey can be conducted. 5) Shallow hydrologic conditions with a saturated soil surrounding a water-filled drainage pipe produce the poorest GPR drainage pipe detection response. 6) Shallow hydrologic conditions with a wet/saturated soil surrounding an air-filled drainage pipe produce the best GPR drainage pipe detection response, especially if the ground surface is frozen. 7) The type of drainage pipe present, either clay tile or corrugated plastic tubing, does not seem to impact the GPR response. 8) The orientation of the drain line, from parallel to perpendicular, with respect to the transect along which measurements are collected, is what governs the GPR profile pipe response that ranges, respectively, from a banded linear feature, to a laterally extended upside-down U-shaped feature, to a laterally compressed upside-down U-shaped feature. 9) Sandy soils often have layers that strongly reflect radar energy, which in turn can potentially interfere with drainage pipe detection. 10) GPR is capable of detecting drainage pipes having a diameter as small as 5 cm (2 in). 11) To avoid missing some of the drainage pipes that are present at an agricultural field site, bidirectional GPR surveys should be conducted that are comprised of two perpendicular sets of parallel measurement lines. 12) The layout of a subsurface drainage system on a GPR map becomes more poorly defined as the spacing distance between GPR measurement lines is increased. As an end product, this research study has accumulated a wealth of information that can be used directly as guidelines to improve the potential for success of using ground penetrating radar to locate buried agricultural drainage pipe.
Technical Abstract: Enhancing the efficiency of soil water removal on land already containing a subsurface drainage system typically involves installing new drain lines between the old ones. However, before this approach can be attempted, the older drainage pipes need to be located. In ongoing research, a near-surface geophysical method, ground penetrating radar (GPR), has been successful in locating on average 72% of the total amount of drainage pipe present at thirteen test plots in southwest, central, and northwest Ohio. The effective use of GPR for drainage pipe detection requires careful consideration of computer processing procedures, equipment parameters, site conditions, and field operations, all of which were thoroughly investigated in this study. Application of a signal saturation correction filter along with a spreading and exponential compensation gain function were the computer processing steps most helpful for enhancing the drainage pipe response exhibited within GPR images of the soil profile. GPR amplitude maps that show the overall subsurface drainage pipe system required additional computer processing, which included 2-D migration, signal trace enveloping, and in some cases, a high frequency noise filter and a spatial background subtraction filter. Equipment parameter test results indicate that a 250 MHz antenna frequency worked best, and that data quality is good over a range of spatial sampling intervals and signal trace stacking. In regard to the site conditions present, shallow hydrology, soil texture, and drainage pipe orientation all substantially influence the GPR response. Additionally, drainage pipe that are as small as 5 cm (2 in) in diameter can be detected. However, the fired clay or plastic material of which the drainage pipe is comprised does not appear to have much of an impact. Finally, with respect to GPR field operations, bidirectional surveys offer the best chance for finding all the buried drainage pipe possible, and for displaying a subsurface drainage system on an amplitude map, the narrower the spacing between GPR measurement lines, the better the result. Although it is important to note that the amplitude maps generated with a wider spacing between GPR measurement lines, still provided plenty of useful data on drainage pipe location. The information supplied by this study can be employed to formulate guidelines that will enhance the potential of success for using ground penetrating radar in locating buried agricultural drainage pipe.