calculation of salt front

[galengorski]

I have been trying to recreate the salt front time series supplied by Salme at al. That dataset gives both a daily and a 7-day averaged estimate of the salt front location. The drb has a dashboard where they report the salt front location and they give the methods for calculation:

[galengorski]

They use an interpolation method with the four sites listed there. Here is the specific conductivity at those four sites for 2019, the red line is a mistake and should be ignored

[galengorski]

It looks like you can convert between specific conductivity and salinity in a few ways. There is a standard method and there are also regional experimental regression equations like those used in this paper. For this, I used the regression equation from the paper, they have a piece-wise regression equation developed from Mid-Atlantic sites (none on the Delaware). Using those methods I came up with the following salt front location plot

:

[galengorski]

Clearly something is off, I think maybe the specific conductivity to salinity conversion isn't accurate for higher values

[galengorski]

I went back and calculated the salt front river mile with a few changes and improvements:

1. I used the Practical Salinity Scale, which appears to be standard practice for sea water, the other method is more appropriate for lower salinity values found inland. This paper does a good job of explaining the method.
2. Converted mg/L salinity to mg/L Cl, I had mistakenly not done that before, which I thought might account for the mismatch. Here is the conversion factor salinity to Cl

I am still seeing an offset of roughly 10 river miles! I'm not sure what is going on here.

[galengorski]

If I shift the salt front location downstream by 10 miles the match is great

[galengorski]

Sent an email to Amy Shallcross of DRBC and she sent me an updated version of the salt front location on 02/04/2022. DRBC is still working on the salt front location white paper with the detailed methods. I still haven't been able to reproduce the salt front location time series with the published equations from the drb dashboard.

[galengorski]

I compared the newer record emailed by Amy Shallcross in 2/22 with the older one emailed in 10/21, there doesn't seem to be much difference. Below is a crossplot of the two records from 2000-2020, the older record is labeled as October (x-axis) and the new one as February (y-axis). From 2000-2020 there are 136 days where they do deviate, with the maximum difference of 2.7 miles and median distance of < 0.01 miles. We will use the updated one going forward, but I doubt there will be any difference in our analysis.