6 Shortcomings of Well Record Databases
A common shortcoming of well record databases is that they typically do not capture the more detailed geological data interpreted from boreholes by experts during hydrogeological or geotechnical investigations. Because water well contractors often lack specialized geological training, the geological data captured by well contractors can be rudimentary and lead to inconsistent classification of different geological units by different well contractors, which makes direct comparison of adjacent records challenging. In addition, water wells are usually drilled using rotary techniques that provide only an approximate representation of the subsurface geology because an intact geologic core is not obtained, but rather geology is interpreted from rock chips that rise to the surface during drilling, where the original depth of the chip is imprecisely known. Water well records can reveal more about aquitards than aquifers because it is common to terminate well construction when a suitable aquifer is encountered without drilling to the base of the aquifer.
Another common shortcoming of well record databases is the poor accuracy of water well locations recorded in the database, especially for domestic wells. Prior to the proliferation of handheld GPS technology, the coordinates of older wells were typically recorded relative to a map reference (e.g., approximate position on a topographical or land subdivision map) and may be only accurate to several hundred meters at best. In jurisdictions where there is no system for physically tagging wells with a unique identifier label, it can be challenging for well owners to find a record of their well in the database, especially where location coordinates are not accurate and property ownership has changed a number of times since the well was constructed. Surface or wellhead elevation information can be useful for interpreting water level and groundwater flow patterns; however, this data is seldom recorded by well contractors during well construction due to the challenges associated with accurately measuring elevation using simple, inexpensive devices.
Well record databases can have large gaps where well construction activity pre-dates the adoption of well record reporting. A voluntary well registration program for older wells can help fill in some of these gaps. For example, in as the Province of British Columbia, the practice of voluntary reporting of well records before the implementation of mandatory reporting in 2016, and the subsequent implementation of a program to register older wells, has contributed to building a considerable database of historical well records.. In other cases, well records may be missing, incomplete, or contain significant errors due to a lack of regulatory compliance or enforcement of submission requirements. Basic well screen details are often lacking in well record databases (e.g., presence of screen, depth of screen), which can make it impossible to associate water well information with aquifers, especially in layered glacial aquifers (Holysh and Gerber, 2014), and therefore difficult to meaningfully interpret water level elevations and yields.
Some well types, such as dug wells, may be under-represented in well record databases due to a bias with respect to well log submission rates by contractors. For example, the distribution of well types in the Nova Scotia Well Logs Database (Nova Scotia Environment, 2020) indicates that only about 10 percent of water wells are dug wells, whereas regional field survey data in mostly rural areas of the province suggests that the proportion of dug water wells may be as high as 30 percent (Kennedy and Drage, 2020). Dry wells may also be underreported (Misstear et al., 2017) because failed boreholes or ‘unfinished’ wells are often not reported by well contractors. This information, however, may be especially important for predicting areas where groundwater quantity is limiting or for constructing accurate yield distribution curves as detailed in Box 5 (Misstear et al., 2017).
Well records tend to be static, capturing information only at the time of well construction, which can limit their usefulness in terms of identifying active wells and current well usage patterns. A common problem is that the water level in the well is in the process of recovering to its static level when the well contractor leaves the site, so the recorded static water level is lower than the water level in the aquifer. Misstear and others (2017) emphasized the need for well record data to be more dynamic in order to be used effectively as a management tool, recording operational data such as water quality, water level and yield over time, in addition to any maintenance activities or modifications to the well.
Standard methods for estimating the yield of domestic wells (e.g., in drilled wells the placement of the drill stem toward the bottom of the borehole, injecting air, and measuring the rate of displaced water as return flow over a period of about an hour with a bucket and stopwatch) often result in the overestimation of well yield. This effect is largely due to the short-term duration of the test, as short-term yields are controlled mainly by the transmissivity of the geologic materials/fractures near the borehole and may not be representative of the longer-term sustainable yield of the aquifer. For fractured bedrock aquifers the longer-term yield will be controlled by the connectivity, storage and recharge properties of the fracture network in the wider aquifer (Misstear et al., 2017). On the other hand, if the well has been insufficiently developed, the well yield estimated by the well contractor following well completion may be underestimated.
In jurisdictions that receive hard copies of well records and then manually enter the information into a database format, keeping the databases current is a persistent, resource intensive challenge. As a result, most jurisdictions favor recording information as submitted on the well logs, making only minor corrections rather than conducting a robust expert review of the accuracy of submitted information.
A lack of consistent policy for governing groundwater data management across national and subnational scales can result in dissimilar data availability and access. Even where groundwater records are widely available, a lack of consistent terminology or standards for the reporting of well record information, especially lithology, is a major challenge for researchers looking to aggregate well record information at spatial levels that cross political boundaries. There can be significant differences in how data is collected and managed even at small spatial scales (e.g., county or subregional level). Inconsistent terminology can be mitigated through expert interpretation and correction, but this process is time-consuming and costly.