The Seismo-electric method utilizes dipole antenna to measure potential differences caused by the seismically induced Seismo-electric conversion of compressional P-Wave energy traveling through a saturated porous media, into an electric field. This electric field is then analyzed, to extract geophysical data. This data collection process is subject to a number of noise, variance and uncertainty sources which will affect the outcome of the results and as such, place limitations, variation, risk and uncertainties on the resulting data. We discuss these potential error sources here.

Seismic Source

• Impulse strike amplitude variation with manual seismic sources can cause correlation and SNR issues.
• Strike rebound may cause false SE data responses at depth.
• High ground compressibility may cause signal amplitude variation.
• Inconsistent strike dynamics, causing strike correlation variations.

Dipole Antenna

• Low resistive coupling between dipoles will reduce signal amplitude.
• High resistive connectivity between the ground and electrode, reducing signal amplitude and lowering the SNR value.
• Dipole electrode vibration may cause oscillations in the data.

Site Noise

• The data Signal to noise ratio (SNR) is the driving variable that determines the data quality and overall resultant risk. The lower the SNR, the lower the quality of the data and the higher the risk associated with the data. Site SNR can rarely be controlled and as such the survey location and noise content must be considered when evaluating the data.
• Power line fundamental and harmonic noise, reducing SNR.
• Electric fencing impulse noise, causing false data signatures.
• Seismic noise produced by external sources, such as traffic.
• External mechanical device manipulation during recording of data.
• Noise induced by telephone lines, or electric railways.
• Noise from switching circuits, transformers, and inverters.

Sensors

• Inaccuracies, drift, variations, limits, and errors in the GPS and ADC sensors, will cause uncertainties in the resultant processed data.
• GPS accuracy is subject to site conditions, where a blocked view of the sky, or lack of available satellites will reduce location accuracy.

Filtering

• In the case of noisy data sets, filtering is required to remove the noise. Filters, by definition, remove information from the data set, reducing depth resolution of the data.
• When the system detects high levels of noise, notch filtering is employed to remove the noise. This can cause ringing on the filtered data sets, due to filter instability, which has an effect on late time data interpretations causing inaccurate results at greater depths. This can be seen in the data as oscillations.

Resolution

• Lateral data resolution is determined by interpoint grid spacing distance.
• Vertical resolution is affected by SNR and filtering. By default, the vertical sampling resolution is 8 to 12 cm, however, if the data SNR values are low, filtering is required to extract SE data. This in turn reduces the data's ability to show fine transitions, thus effectively lowering its vertical resolution. Typically, the true resolution of the system is approximately one meter when high data SNR is present and no filtering is required, however, this can reduce to 5 to 10m in very noisy data sets where aggressive filtering is required.

Processing

Processing is dependent on the variable data provided by the sensors, as such there will be uncertainty and risk associated with processing the data and the resulting outcomes.

Interpretation

The system employs artificial intelligence to interpret data, this is highly dependent on the quality of the data provided to it, to produce accurate interpretations. The higher the uncertainties, noise, and variations in the provided data sets, the less accurate the interpretations will be.

Quality Control

• Quality control enforces the minimum acceptable level of data quality, in terms of SNR, strike data correlation, minimum used stack sets, minimum number of strikes detected, recording length, and data clipping. This allows for a minimum level of confidence in the processed data, but it is up to the user whether to accept the data or not.

Interpolation

• This geophysical method is a point sounding based system, in that data is only collected directly under the seismic source, within a 2m radius of the seismic source location. As such, very little to no lateral information beyond this radius is collected. To collect lateral data, soundings must be done on a line or a grid.
• Interpolation is then used to produce 2D and 3D plots of the line or grid data.
• The interpolation used in this report is inverse distance weighted, with a horizontally biased interpolation plane and an interpolation distance determined by the maximum distance between any two points in a project.
• It is important to note that only data collected directly under a sounding point is true data. Interpolated data is a visual interpretation of the similarity between the data collected at discrete locations. The data between points is not actual data, as such should not be considered real collected data.

Fracturing

• Only fractures that are horizontal or thirty degrees slanted can be detected by this geophysical method. Vertical fracturing cannot be detected.
• Fracture width, height, and roughness are estimated as they cannot be delineated by this system.
• Fracture lateral extent is calculated by geometric proximity to fractures detected in surrounding sounding locations. These fracture indicators may, or may not be, truly connected. This can affect yield estimates.

Gravel Beds

• Gravel beds are represented by low probability fractured formations, due to their weathered nature. As fracture dimensions are not delineable by this system, the yield rates are calculated by the worst-case scenario of a small grained, poorly sorted gravel matrix. Thus, the minimum yield is produced for any detected gravel formation.

Clay Formations

• Saturated clay formations can be interpreted as an aquifer system under certain conditions.

Temperature

• Thermal variables must be calibrated against known site thermal gradients.

Liability

All the data sets, interpretations, recommendations, logs, and risk assessments, presented in this document are calculated estimates of the values and parameters they represent, as they are derived from the signal attributes of multiple sets of sensor collected data. These sensors are subject to variance, drift, noise, error and uncertainty. As such, the data collected from these sensors, and the data sets derived from the sensor collected data, are subject to the same variance, drift, noise, error and uncertainty. As such, the data sets discussed in this document cannot, and should not, be viewed or interpreted as being absolute in nature. Any actions or decisions made by the user with reference to the data provided in this document, should consider the uncertainties and risks involved in its use.

Data Quality and Risk Matrix

The system calculates variables to broadly estimate the uncertainties, variations, drifts, and errors stated here, into quantifiable parameters, such as Risk, Confidence, Uncertainty, and Resolution, that the user may assess when reviewing the data. These must be considered by the user when assessing the data provided in this report.