DWQ-2024-004887

Summary of discussion of EPC0 calculations
2020-06-10

Goel et al. were asked to compute equilibrium phosphorus (P) concentrations (EPC0) as part of
revisions to the “Utah Lake Sediment-Water Nutrient Interactions” study report. Below is a
summary of discussion about these revisions.

1. Calculation methodology

The calculations referenced in the report and suggested by reviewers use a regression approach
that compares water column P concentrations to the amount of sorbed P in sediment (mg P/kg
dry sediment). This should result in a positive correlation when plotted. The authors correctly
state that any P lost from the water column can be assumed to be sorbed by the sediments, but
the calculations conducted use the change in P in the water column (mg P/L) rather than the
amount of P taken up by the sediment itself. This led to a negative correlation. Equating water
column volume with sediment mass is not appropriate in this case (at least not without a citation
demonstrating this approach is appropriate), as the goal of computing EPC0 is to quantify how
much P a unit of sediment can take up. From personal communications with Dr. Goel, the bulk
density of sediment is known (600 kg/m3), which could be used to convert the volumetric P
uptake to sediment mass-dependent uptake (sediment core depth and volume is known).

On the x-axis, some analyses use the P concentration at the start of the incubation (e.g., Haggard
et al. 2004) and others the P concentration at the end of the incubation (e.g., Pant and Reddy
2001). It is unclear which is appropriate in this case as the authors do not reference an approach
they are following.

2. Incorporation of replicates and variability

The figures presented in the report have only four data points, but the data points represent an
average of several replicates. As reported earlier in the report, the error bars around TDP and
SRP fluxes are substantial and in many cases cross zero, pointing to the large degree of
variability in observed uptake/release. Reporting these values as a single number is misleading,
as it compresses the known variability into single points. Including all replicates in the figure and
calculation would enable a more accurate depiction of the variability inherent in these rates and
better quantification of uncertainty around computed values.

3. Choice of incubation length for calculations

As is shown in the report, the sediments switch between source and sink of P throughout the
incubation, so fluxes could be interpreted differently depending on whether 12, 24, or 72 h was
used as the endpoint of the incubation. If these endpoints are incorporated into the analyses (via
the spreadsheet provided by the authors), the calculated EPC0 concentrations come out to
different values, including some that are negative. The authors have not defended their choice of
using 72 h as the endpoint in the analysis. Further, a discussion of the timescale of equilibration
of sediments was not provided, potentially hindering the usefulness of these calculations as they
relate to the conditions and timescales in Utah Lake.

4. Confidence in reported EPC0 concentrations

For the reasons stated in points 1-3, there is likely a low degree of confidence in the reported
EPC0 values. This confidence could be increased and/or better quantified by incorporating the
suggestions stated above.

5. Interpretation of future scenarios

In theory, the EPC0 value represents a fixed concentration based on the present sediment
conditions. What processes could impact sediment conditions and likely the EPC0 values in
response to reduced P loading? Including aspects of sediment chemistry that will likely affect
their capacity to sorb and/or release P would be a useful addition.