I cannot say anything on geochemistry of boreal wetlands. However, I am working on analysis of compositional data. I feel that your results may appear confusing if you do not transform the geochemical composition into an appropriated set of log-ratios. Time series analysis is mainly based on correlations. And correlations between raw compositional data are known to be spurious. The first one pointing out this circumstance was K. Pearson (1896). If your results are not understandable, may be this is due to the spurious correlation and they require a compositional treatment.

I would suggest to have a look on compositional data analysis. An up to date reference is

Pawlowsky-Glahn, V. and Buccianti A. (Eds.)

Compositional Data Analysis: Theory and Applications,

ISBN-10: 0-470-71135-3, Wiley, Chichester UK, 2011.

which is made of papers going from basics to applied and theoretical issues.

On the other hand, I have recently publish a paper on differential models for evolutionary compositions that may be related to your topic (from time series --difference models-- to processes based on differential models)

Egozcue, J. J. and Jarauta-Bragulat, E.,

Differential Models for Evolutionary Compositions,

Mathematical Geosciences, 46, 4, 381--410, 2014.

You do not give sufficient information to allow informed comment; nevertheless, ion-exchange onto sorption sites of Ca displacing Na (freshening) or of Na displacing Ca (salinization) may be worth looking at. If the change in meq/L is the same for both (i.e they balance) you have a partial answer. Then your question would be what drives changing water composition in the source regions.

Thanks John. I thought about it, but since this is not a coastal wetland, where Ca is brought seasonally with each snow-melting episode the question concluding your answer is the core of the problem. The recharge is strictly local, within the perimeter of the wetland. The flow is radially outwards (in all directions), so the source of Ca must be within the wetland. I suspect that it is associated with biochemical selection during the process of biomass growth and the selection may be even specific to a particular species.