Detection of Soil pipes and Tunnel using Multi-electrode Resistivity imaging Techniques, Western Ghats, Kerala.
A case study in Kannur and Kasaragod Districts.
1. Introduction
Land subsidences and tunnel formations are now become common features in the Western Ghats, Kerala during high rainfall seasons. This phenomenon is caused by physical and chemical processes of soil. The major reason for land subsidence is subsurface erosion called “soil piping”. The incident was first reported in 2005 in Cherupuzha, Kannur(fig.1) since then many incidences of land- subsidence were noticed in different parts of the State, most of them often go unreported. This work aims to identify the suitable electrical resistivity imaging technique to detect places that are susceptible to soil piping process.

 Figure1. Land subsidence due to soilpiping in Kannur

Soil Piping
“Soil Piping”, also known as tunnel erosion is the subsurface erosion of soil by percolating waters to produce pipe-like conduits below ground especially in non-lithified earth materials.During rains percolating waters carries finer silt and clay particles and forms passage ways.
•   The resulting "pipes" are commonly a few millimetres to a few centimetres in size, but can grow to a meter or more in diameter.
•   They may lie very close to the ground surface or extend several meters below ground.
•   Once initiated they become cumulative with time, the conduits expand due to subsurface erosion leading to roof collapse and subsidence features on surface.
•   Cavities in the form of tunnels usually grow faster during high rains
Figure 2. Data base on Soilpiping(Geology map: Courtesy GSI)

During the last decade many piping incidences were reported from different places of the state. Data base on soil piping (Fig. 2) indicate that this phenomenon occurs in many areas in the Western Ghats irrespective of the lithology. Many of them are located at Idukki and Kannur, then followed by Kozhikode,  Palakkad,  Ernakulum, Pathanamthitta, Kasaragod, Wayanad.

2. Objective
The objective of this study is to determine the suitability of geophysical techniques such as Electrical Resistivity surveys by using Digital multi-electrode resistivity imaging techniques to map the underground pipes/cavities and their extent .

3. Study Area
The study area comes under the two districts, located at Nelliyedukkam (fig. 3) in Kinanoor village, Vellarikund taluk in the Kasaragod district (lies between latitudes 11º40' to 12º 48' North and longitudes 74º 52' to 76º 56' East) and Kottathalachimala (fig. 4) in Thaliparamba taluk in the Kannur district (lies between latitudes 12º15' to 12º 17' North and longitudes 74º 12' to 76º 15' East), The affected locality in Nelliyadukkam has been the side slope region at 55 meters above MSL in elevation. Pipe outlet is located at 50m north of this area. In this area a huge tunnel shaped pipe was formed as a result of sudden collapsing of the surface soil due to the continued erosion taken place beneath the top cover. Surface fissures were also identified in the premises of the tunnel. In Kottathalachi mala the affected locality is situated in the highland region with elevations reaching up to 801m. The upper slope is characterized by s high slopes (+30 0 ) where as the lower slopes are about 24 0 . The Kottathalachi mala is characterized by radial drainage indicating its shape. Many lower order streams originate from this hill. Charnockites rocks dominates Precambrian crystalline basement of the area The overburden material above the basement is clay rich loamy soil. The side slope of the Kottathalachi mala is characterized by thick soil cover especially the area where the incident (tunnel) occurred. The average overburden thickness is about 15m.

Figure 3 Location map at Kannur and
Kasaragod district
  Figure 4. Piping locations of
  Figure 5 Cross-section of piping
in Nelliyadukkam

4. Materials and Methods
Multi-electrode electrical resistivity imaging techniques were used for this survey. The instrument used consisted of a multi-function Digital DC Resistivity/IP Meter (fig. 6), with a WDJD-4main frame, WDZJ-4 switcher box (Multiplex Electrode Converter), 12V rechargeable battery as a transmitting power source ,electrodes, multi-electrode cables etc. In order to conduct multi-electrode2D electrical resistivity survey, the instrument system will automatically select current electrodes and potential electrodes according to a specific electrode array, and gives measured results of all the data points of a cross-section. The data so gathered is processed and interpreted using RES2DINV Software.

Figure 6 Field survey at Kannur by using WDJD-4 Multi-function Digital DC Resistivity/IP Meter
In the multi-electrode resistivity survey the maximum 60 electrode were used, The maximum length of the profile was depending on the electrode separation.WDZJ-4 switcher box used for connecting electrode and resistivity meter. In order to nullify the contact resistance, if any, at the
electrodes, Grounding Resistance (Rg) was initially measured for the set of electrodes by setting the desirable maximum limit of Rg to 5 K. Ohms considering the requirement of improving the signal to noise ratio. To improve the ground contact by tight pegging of the electrode and/or by pouring saline water. After ensuring that all the electrodes are well grounded without contact resistance beyond the desirable limit, the switcher box is connected and measurement mode initiated.

Surveys at Nelliyadukkam, Kasaragod

The surveys were laid (figure 7), on above the known soil pipe. Electrode spacing at 1m and 2m, so, total survey line went up to 60 and 120m respectively. In survey line S 1 and S 2 , the instrument was placed at the centre of the target (above the cavity) to arrive greater depth of penetration. S 3 is laid north of the survey line S 1.  ,in between the subsidence 2 and outlet of the piping. The best resultant cross-sectional Electrical Resistivity Tomographic sections are plotted below.

Figure 7 ER surveys layout at Nelliyadukkam

5. Results

Survey line 1 (S1)
In profile 1 laid on the top of the tunnel at 1m electrode spacing, the total depth of information obtained from the figure 8 is 11.5m. and high resistivity value obtained was 2366Ωm. From the schlumberger and Wenner images, a sudden increase of resistivity (~500 to 2366Ωm) in a depth of 2.69m indicating the starting point of tunnel section. The high resistivity zone continued up to a depth of 11m. The lateral extends of the tunnel was more visible in this profile.

Survey line 2 (S 2 )
This profile was also laid at the same point of profile S 1 with an electrodeinterval of 2m. Depth of information obtained was 23m.The tunnel roof was observed at a depth of about 3.98m from the surface in all the three configurations, the high resistivity zone could be seen extending at a depth of 3.98 to 10.5m, in the schlumberger configuration. The data generated by schlumberger configuration was more accurate than Dipole-Dipole (Figure 9c) and Wenner array mode.

Survey line 3 (S 3 )
The survey line 3 is laid across the soil pipe (east to west direction) and the mid point of the profile is 12m north of the pipe. The electrode interval was 2m, the total distance of the profile was 120m and the maximum depth of penetration achieved was 23m.(Fig10) Two high resistive (>1551 Ωm) zone was identified at a depth of ~ 5.37m was, indicated piping branching.

Surveys at Kottathalachimala, Kannur
Two ERPs are laid in the study area (Kottathalachimala) as per the layout (Fig 11). The length of the ERP was 90m, and 60m for the current electrode spread of 1.5m and 1m respectively. The profiles 1 and 2 were laid over a known soil pipe, perpendicular to the direction of piping. The resistivity measurement on each profile was taken for three different electrode arrays namely Wenner, Schlumberger and Dipole-Dipole. The resistivity data on each profile using different arrays were labeled and stored.
Figure 11 ER surveys layout at Kottathalachimala
Survey line 1
Figure 12 ERT image at a.1m electrode spacing b. 1.5m electrode spacing c. 2m electrode spacing
From profile 1 (Fig12a), it can be observed that Schlumberger and Wenner are not very efficient in mapping the entire tunnel cross-section but only the top of it. Both the configurations show the tunnel extending indefinitely with depth. The two configurations fail to map the tunnel bottom even though from physical observations it is known that the vertical extent of the tunnel is lesser than the investigation depth of the two arrays. The dipole-dipole array effectively maps the entire tunnel cross-section.
Survey line 2 and 3
Figure13 ERT image at a.1.5m electrode spacing b. 2m electrode spacing c.1.5m electrode spacing
The survey line 2 consist of two profiles (Figure 13a & b), with an electrode spacing of 1.50m and 2.00m respectively, exhibit pipe like anomaly, can be seen in dipole –dipole array, the tunnel size is smaller compared to tunnel entrance (survey line 1). As a continuation from survey line 2, In Survey line 3 (figure13c) laid on the outlet portion of the pipe an anomaly were observed on the surface of RS image, it shows a high resistivity region. The anomaly could be indicative of small underground soil pipe or a boulder. Schlumberger response for soil-pipe as observed for survey lines 1, 2 & 3, the source of anomaly remains inconclusive.

6. Conclusion
The image generated using schlumberger, Wenner and Dipole-Dipole configuration for station interval of 1m, 1.5m, and 2m exhibited resistivity anomalies. The profiles laid on the top of the piping showed high resistive region which indicates tunneling. The resistivity of cavity is much higher than resistivity of hard rock. Surrounded by this high resistive region there were lowering of resistivity indicating saturated soil beds. The Schlumberger configuration has the highest investigation depth, Dp-Dp array configuration clearly brings out the entire tunnel cross-section. The qualitative interpretation of the resistivity section indicates that the technique could delineate the conducive zones where the soil pipes are present. The interpretations are based on the field observation; actual physical measurements are matching with the data generated by electrical resistivity tomography.
PREV:Multi-geophysical approaches to detect karst channels underground
NEXT :The comprehensive geophysical method for detecting surface-groundwater interaction and contamination in India