Clutter is a term used to describe unwanted echoes in radar systems. In Ground Penetrating Radar, clutter can be caused by poor design or interference to the system, or it can be used to describe all of the targets which the GPR detects, but it is not looking for.
All of these extra targets ‘clutter’ the display and detract away from the object you are trying to find, making it harder to distinguish within the data.
Most modern GPR are not significantly affected by internal noise or interference, therefore it is external sources of clutter which are the most significant. These come from reflections from unwanted targets or interfaces in the ground such as rocks, bricks, tree roots, and additional reflections from water interfaces.
It is necessary to remove clutter in order to correctly interpret GPR data.
In general, clutter means features in the GPR data which you are not looking for. The image above shows a GPR b-scan across a grass area which contains tree roots and stones. The response from a utility would look very similar to the stones and it is necessary to find a method to differentiate between the features in the data you are looking for, and those which you are not.
We can’t prevent unwanted reflections from appearing in our Ground Penetrating Radar data, so we need to find a reliable method for discounting them. This can be done during processing (by post processing the data in such a way that it highlights the features we are looking for) or through the methodology used on site – or a combination of both.
To remove or discount clutter from our data we have to think about what we are trying to detect with our GPR. For example, in a utility survey we will typically be looking for objects which are linear.
A linear object should look the same or similar with a similar size and shape of hyperbola and a relatively consistent depth, in consecutive scan locations. Therefore, a reliable way to highlight this feature and discount the clutter would be to perform consecutive parallel GPR scans and look for targets with similar features which appear consistently in all of the scans – whilst features which appear isolated will be regarded as false alarms and can be discounted.
In general, homogeneous soils such as sand generate less clutter, and in-homogeneous soils such as in brown fill sites generate more clutter.
With GPR, you can detect a wide range of objects below ground level, including both metallic and non-metallic objects such as plastic pipework. GPR will also identify and map any voids below the surface, such as air pockets or mine shafts, as well as any other irregularities including concrete and previously excavated or back-filled areas.
GPR equipment emits an electromagnetic pulse into the ground and records the reflected signals from subsurface structures and voids. It is entirely non-destructive and will not break the ground’s surface or affect any objects below. What’s more, it doesn’t emit any harmful levels of radiation, nor are there any other by-products created throughout the process. This means it’s entirely safe to use by its operators, and on sites of any type, including those open to the public.
While GPR is one of the most effective methods of non-destructive testing available, it can never be 100% accurate. One factor that can adversely affect the accuracy levels include the type of soil being surveyed. Clay soils and soils that contain high levels of salt or minerals can obstruct the GPR reading. Another factor is the experience of the equipment’s operator: interpreting the data collected can be complex, which is why it’s beneficial to commission surveys from an expert firm.
The equipment itself is not difficult to use, but the interpretation of the data recorded tends to be complicated. The results of a GPR survey aren’t automatically translated into an easy-to-understand picture of what lies below the surface; instead, it’s a series of lines and waves and it can take both training and years of practice to master the art of correctly reading the output. Often, it is the experience of the equipment’s operator that plays the most significant role in the accuracy of the results GPR can achieve.