Revisiting Solute Dynamics in the Luquillo Critical Zone Observatory Using Young Water Fractions

Abstract:

Rivers carry solutes derived from mineral weathering, atmospheric deposition, and biological processes and thus record the net biogeochemical effects of Critical Zone processes. At a single site, riverine solute concentrations can vary considerably and these changes are often correlated with discharge. Two possible drivers of this concentration-discharge (C-Q) behavior are changes in the transit time of water through the Critical Zone and the activation of preferential flowpaths. While changes in transit time affect the progress of chemical reactions, solute concentrations are also affected by the mixing of water that has reacted with distinct substrates (e.g., flow through soil vs. bedrock). Distinguishing the effects and relative importance of these different drivers of C-Q behavior is difficult, however, since few watersheds that exhibit preferential flowpath activation have also been characterized in terms of fluid transit times.

Here, we report new time-series observations of water isotopic ratios from the Luquillo Critical Zone Observatory (LCZO) and use them to constrain transit times and their relationship to C-Q behavior. The LCZO, which is located in a montane tropical rainforest in Puerto Rico, includes multiple intensely-weathered study catchments underlain by either granitoid or volcaniclastic rocks. In previous work, solute variations during storm events have been attributed to the activation of preferential flowpaths, but associated changes in transit times are poorly constrained. Using samples taken weekly over the course of 2 years from 7 watersheds as well as from rainfall and throughfall collectors, we compute the young water fraction (F_yw­), which is equal to the proportion of discharge younger than about 3 months. Preliminary results imply a F_yw­ around 25%, but this estimate will be revised after more analyses. Specifically, we will determine the discharge-dependence of F_yw­ in order to more directly compare transit time constraints with the C-Q behavior.