Tracking Infiltration Front Depth Using Time-lapse Multi-offset Gathers Collected with Array Antenna Ground Penetrating Radar

The overall goal of this experiment is to track the infiltration front in field soil using array antenna ground penetrating radar. This method can help answer key questions in the vadose zone hydrology field such as understanding how water infiltrates into the soil during the rainfall. The main advantage of this technique is that time lapse multi-offset gather can be collected seamlessly with minimal effort during dynamic subsurface processes such as infiltration.

Demonstrating the procedure will be Nobuhito Nagai and Yukio Tobe. The antenna array is at the core of this experiment. Within this housing are the 21 antenna for this experiment.

This schematic provides additional details. There are 10 transmitting and 11 receiving bow-tie mono pull antennas controlled by a step frequency continuous wave form radar unit. The array can switch through all 110 transmitter receiver pairs.

Identify a site for the infiltration test. Make sure the surface is bare, flat, and about three meters by three meters. Obtain 2.5 meter long porous tubes.

The tube porosity allows water to be released onto the ground. Use the tubes to construct an irrigation infiltration system on the plot. This system's tubes are parallel and separated by 15 centimeters.

Connect the tubes at one end with a valve regulated water inlet connected to a water source. And connect the other end to the outlet. Next, get a thin wood panel, slightly larger than the antenna array, and place this over the porous tubes.

The panel should be flat and cover the tubes. Move on to install soil moisture sensors near the panel. This rod type sensor measures moisture at several depths.

First, install the sensor's access tube in the ground next to the wood panel. Then place rod type sensors into the access tube. Start by placing the antenna array on the wood panel, centered on the infiltration system.

Connect the array to the controller with coaxial cables and connect the controller to a computer. Set the antenna sequence to perform a full scan through all transmitter receiver combinations and begin collecting data. At this point, go to the water source and start the flow and infiltration.

Stop the water when a predetermined amount has been injected. Then, stop collecting data with the radar array. After collecting data, analyze it to find a velocity estimate.

This is an example of the time lapse radargram from the antenna array. The data were taken over 60 minutes of the experiment. Each region bounded by black lines corresponds to data collected by one transmitter and the 11 receivers.

The location of the transmitter is indicated by the red triangle. Along the third axis is the total time from transmission to reception of the reflected signal. The different colors indicate signal amplitude.

Using this data, construct the common midpoint data cube. The new axis is the separation of the transmitting and receiving antennas. This is a cross section of the common midpoint data cube at a given time during the experiment.

Identify the reflection from the wetting front for each such frame. Use this equation to fit the hyperbolic curves by adjusting the two parameters, t0 and vr. The white curve represents the travel time of the reflection from the wetting front.

The purple curves are travel times for the air and ground waves. This is a representative radargram for an infiltration experiment. Each section is associated with a transmitter.

Along the vertical axis is the travel time from transmitter to reflector to receiver. The gray scale color map indicates the signal amplitude. In this time lapse from one of the common midpoint data regions observe the high amplitude signal moves steadily downward as the experiment progresses.

The signal is produced by the reflections at the wetting front as the water penetrates the subsurface. For velocity analysis, use common midpoint data taken every minute. In this sample data, five minutes into the experiment, the best fit curve for the reflected wave is the solid white line and that of the airwave is the dashed line.

As time goes on, the travel time increases linearly. These are plots of the estimated wetting front depth as a function of elapsed time. The triangles are for a model of a uniformed medium.

The squares are for a two layer model that takes the wooden panel under the array into account. The intervals in black indicate when a moisture sensor reading at a given depth started to increase and when it stabilized. We first had the idea for this method during a demonstration for array antenna ground penetrating radar.

It occurred to us to correct time lapse multi-offset data. The most great advantage of this procedure is that we did not need to move the antennas to collect multi-offset gather unlike common ground penetrating greater systems. Once you start the measurement, it's just monitor and do nothing.

While attempting this procedure, it's important to remember not to move the antennas so that the reproducibility of the data is insured. After its development, this technique will pave the way for researchers in the field of vadose zone hydrology to explore water movement in field soils. After watching this video you should have a good understanding of how to use array antenna ground penetrating radar to track infiltration fronts.