Distribution, Drivers, and Threats of Aluminum in Groundwater in Nova Scotia, Canada

Reviewer #1

Response

The manuscript is well-written and well-presented. The approach used is straightforward. I only have a few comments about the manuscript.

Thank you for taking the time to review our manuscript and for providing useful comments for its improvement.

1) It would have been useful to evaluate the geological units that are encountered by the water wells because different units may be encountered in the subsurface even though the geologic maps indicate a particular unit.

To address this comment, we added: “Uncertainty associated with the assignment of bedrock groundwater regions to well water samples increases near geologic boundaries due to factors such as map scale, interpolation of geologic boundaries, and gradational contacts between units. Water samples located near these boundaries were therefore carefully reviewed, and stratigraphy information from well logs was consulted where it was available to verify the assignment of the groundwater region.” [L172].

2) what were the characteristics of the drilled wells? How deep are the wells, where are the wells cased, when were the wells constructed? Some of these characteristics may have a bearing on the results. Did the well database have this information? Could the authors address these concerns?

To address this comment, we added: “Most drilled wells in Nova Scotia are less than 100 m deep, intercepting fractured bedrock aquifers, with about 6 to 12 m of casing to keep unconsolidated sediments from collapsing and the bedrock section left as an open hole (Nova Scotia Environment and Climate Change, 2020).” [L143]

3) The data used in Figure 1 are included in Table 1. Consider deleting the data in table 1 that is shown in Figure 1.

Thank you for identifying this redundancy.  The authors have discussed this question and feel that it is useful to have the percentages in the table so that the statistics are readily available and easy to reference.

4) Line 211 - Figure 3 does not show that the highest exceedances are in plutonic regions. Figure 2 does.

Thank you – we have corrected this reference, removing Figure 4 from the reference in L211.

5) Figure 6 is not very useful. Maybe running the analysis such that only the high exceedances are shown would be more insightful.

To address this comment, we have updated the map in Figure six to show only two categories for the points – small black dot for concentrations less than 100 ug/L and larger red dot for concentrations >=100 ug/L.

6) Line 35 - the word 'they' should be 'there'.

Fixed.

7) Line 37 - A period is missing after "...[4,7,8]"

Fixed.

8) LIne 158 and 171 - "ie." should be "i.e." 

Fixed.

9) It would also have been useful to get a sense of the when the samples were collected because there might be temporal variations in Al concentrations in the groundwater. Does the database have this information? What are the implications for binning all the data collected in different seasons together? 

To address this comment, we have added the following statement about the dataset: “The spatial coverage is not regularly distributed across the province and is biased towards more developed areas of Nova Scotia, where water wells and chemistry data tend to be more readily available. The compiled water chemistry data spans a period of 40 years and comes from a variety of sources ranging from samples collected by homeowners who have shared their well water chemistry results with government, to well water quality surveys conducted by groundwater professionals. Hence, sampling and laboratory analytical methods are not consistent across the dataset. Field measurements of relevant geochemical parameters, such as dissolved oxygen, pH and Eh, were not part of the dataset. Although the quality of the well water chemistry data used in the analysis varies, it is still considered useful for detecting regional-scale (e.g., >1:50 000) trends. Analysis of seasonal or temporal trends is not possible with the data set due to having only one sample per well. We recommend future study on seasonal trends of Al in wells, particularly dug wells, because seasonal patterns in Al concentrations have been identified in surface waters in Nova Scotia (Rotteveel and Sterling, 2020). The well samples were collected most often in the summer (47% of samples), followed by 26% in the spring, 21% in the fall and only 6 percent in the winter.

”  [L339]

If the wells had seasonal patterns in Al concentration, binning data from different seasons would make it more difficult to assess whether the highest annual Al concentration was likely captured.

Reviewer #2

In the reviewed manuscript, the authors investigate the hydrogeological conditions that influence the Al concentrations in groundwaters, based on an extensive set of data gathered from various databases. The article in well written and present important information on the factors (pH, TOC, geology, type of groundwater) that may influence the concentration of Al in groundwaters. I recommend the publication of the manuscript after minor revision.

Thank you for taking the time to provide these helpful comments.

1) P2L56-57 please give details on the dissolved Al complexes that may be leached from soil to water

To address this comment, we added the chemical formulae for the Al complexes that may be leached from soil to water,

i.e., AlF₃, Al(OH)₃, AlPO₄, Al₂(SO₄)₃   [L57]

2) P5Table 1 please define what count detects means. Maybe basic statistics on Ca, TOC, turbidity, pH of each groundwater region would be interesting to be added.

To address this comment, we added a footnote beside the ‘Count Detects²’ title, i.e., 1. Count of the Al results that are greater or equal to the laboratory detection limit’.  Also, we have added median values for Ca and pH to Table 1. We did not add statistics for TOC and turbidity data due to the left-censored (mostly non-detect) nature of the distributions.

3) P6L205-211. Please extend discussions on Al in sedimentary geology region as it could give interesting results. High max Al content but low 95 percentile and mean may indicate some extreme values?

To address this comment we added the following to L 281]:

“While the sedimentary groundwater region was generally associated with lower Al concentrations compared to the plutonic and surficial regions, there were also several very high Al concentrations (>2,000 µg/L) detected in sedimentary bedrock aquifers (Table 1). Most of these extreme Al concentrations, however, were also associated with elevated turbidity due to insufficient well development following drilling (e.g., test well), which can result in elevated Al when the water sample is not filtered prior to acidification, as recommended for assessments of drinking water quality.”

4) P7L220-223 please verify the statement with the data in Table 2.

To address this comment, we clarified the statement about the data in Table 2 as: “After pairwise comparison of Al concentrations across the groundwater regions, the plutonic and surficial groundwater regions were interpreted to have significantly higher concentrations (p <0.05) than all other regions whereas the carbonate-evaporite groundwater region had lower concentrations than all other regions (Table 2). There was no significant difference interpreted between Al populations for metamorphic, sedimentary, and volcanic groundwater regions (Table 2).”

5) Please format the references according to the journal requirements.

As far as we are aware, the references are properly formatted according to the journal requirements. We used the MDPI style file for Zotero that was recommended on the journal’s submission instructions page.