Dissolved silicon concentrations and silicon concentration-discharge relationships have been used widely as a tracer of subsurface flowpaths in catchment hydrology. This study will employ Ge/Si ratios as a novel hydrologic tracer, and will test emerging ideas regarding the role of terrestrial plants in controlling streamwater Si fluxes. Ge/Si data from a research catchment in Puerto Rico (Rio Icacos) will be supplemented by oxygen isotope analyses, hydrometric data, and physical hydrologic modeling to test the following hypotheses: A) High Si concentrations observed in streams at base flow reflect weathering at the saprolite-bedrock interface. This silicon is delivered to streams via deep groundwater discharge; B) Low Si concentrations observed during runoff events reflect rapid release of silicon from secondary soil minerals to water flowing along shallow permeability discontinuities; C) High Si concentrations sometimes present in shallow soil porewater reflect plant cycling and dissolution of phytoliths. This Si does not comprise a significant part of the overall streamwater Si flux at our site. This project is aimed at developing a new tool that will increase our understanding of the underground movement of water, delivery of water to streams, and the processes that control the chemistry of stream water. These issues are important determinants of water quality and are central to basic research in geochemistry and hydrology. Here we apply this tool (concentrations of naturally-occurring silicon and the trace element germanium in stream water) to an established research watershed in the Caribbean National Forest in Puerto Rico. The project promises to integrate information from additional chemical tracers of water sources plus physical measurements of soil moisture and water table levels, and use mathematical models of water flow. Data collected and ideas developed from this study should lay the groundwork for widespread application of this tool by hydrologists and geochemists in other settings.
