This grid represents the distribution of total dissolved solids (TDS) in the Scollard / Willow Creek hydrostratigraphic unit (HSU). It is a result of a geostatistical model which was constructed using publicly available water chemistry data from the following sources:
1. Alberta Environment and Parks' (AEP) Alberta Water Well Information Database (AWWID)
2 .Alberta Health Services (AHS)
3. Alberta Energy Research Institute (AERI)
4. Baseline Water Well Test Database (BWWT)
The TDS surface was mapped at a regional scale using many years of data, thus providing a good indication of regional water chemistry within the unit rather than site-specific values.
The following statistics describes the accuracy of the geostatistical layer used to create the final grid:
Samples: 3293
Mean error: 3.2 mg/L
Root mean square error: 458 mg/L
Mean standardized error: 0.006103098
Root mean square standardized: 0.95094
Average standard Error: 459 mg/L
The horizontal accuracy of the input TDS data (AGS DIG 2022-0026 data points) is affected by the locational accuracy of the wells:
A majority of the AWWID wells are located to the middle of the Alberta Township System (ATS) quarter section (occasionally to the middle of the section). Therefore, there can be up to approximately 566 m (or up to approximately 1131 m if the well is located in the middle of a section) of error in the well location and subsequent maps derived from the well records. These sources of error are not unique to this study and are encountered in any hydrogeological study that uses the AWWID, unless further refinement of well locations is undertaken.
The locations of the oil and gas wells may be accurate to within millimetres depending on surveying standards at the time of drilling.
The vertical accuracy of the input TDS data (AGS DIG 2022-0026), is affected by the locational accuracy of the wells:
The potential for up to approximately 566 m (sometimes approximately 1131 m) of horizontal accuracy error for water wells may also introduce errors in the surface elevation, which is derived from the provincial 25 m DEM, based on the location of the well. This error is likely on the order of plus/minus 5 m, but may be 10's of metres in areas where there are large changes in elevation. Any error in elevation may influence the allocation of the well as well as sampling location. These sources of error are not unique to this study and are encountered in any hydrogeological study that uses the AWWID, unless further refinement of well locations is undertaken.
The mid-point of the sampled interval is a calculated value and its accuracy is dependent on the accuracy of the measured depths and elevations provided in the well header. These values are somewhat inconsistent for the water wells because many of them are not surveyed. For the oil and gas wells, the well elevations and locations are surveyed in, so accuracy will be to the same significant digits provided by the source, even though there may be considerable uncertainty as to where the sample actually came from.
The location of the sampled interval in oil and gas wells may be accurate to within millimetres in accordance with surveying standards at the time of drilling.
Step 1 (Modelling the surface): An empirical Bayesian kriging algorithm was used in the ArcMap Geostatistical Analyst extension to interpolate the total dissolved solids values and create a grid of the total dissolved solids surface of the Scollard / Willow Creek HSU.
Method Report:
Output surface type: Prediction
Power: 1
Neighbourhood type: Standard
Maximum neighbours: 15
Minimum neighbours: 10
Sector type: 4 sectors with 45° offset
Samples: 3293
Cell size: 500 m
Step 2 (Grid Alterations): The final grid was clipped based on the extent of the Scollard / Willow Creek HSU, and spatial distribution of data.