GERMANIUM AND SILICON AS INDICATORS OF PREFERENTIAL HYDROLOGIC FLOWPATH ACTIVATION IN SMALL TROPICAL PANAMANIAN WATERSHEDS

In the humid tropics, the hydrologic flow paths taken by water during rain events and how those flow paths interact with groundwater to form stream water are poorly understood. This is due in part to the presence of shallow preferential flow paths, which are more developed in tropical soils than in temperate regions due to continuous and intense biological activity in the shallow soil by animals, insects, tree falls, and root decomposition. Additionally, the effects of land cover on the partitioning of water in tropical catchments is a subject of debate, as these hydrologic processes have been the subject of far fewer investigations than in temperate regions. This work applies germanium (Ge) and silicon (Si) as natural flow path tracers to examine the hydrologic partitioning of event waters in three experimental catchments in central Panama. These small catchments (40-153 hectares) share a common lithology, slope, and soil type, but vary primarily in land cover. We identified three end-member waters in these catchments – stream baseflow (high [Si], low [Ge], low Ge/Si ratio), dilute event water (low [Si], low [Ge], moderate Ge/Si ratio), and shallow soil water (low [Si], high [Ge], high Ge/Si ratio). Using these end-members, a three component mixing model was employed as a hydrograph separation tool. During small magnitude rain events, storm flow is dominated by only the baseflow and dilute event water components, as [Si] is diluted and [Ge] remains relatively constant throughout the hydrograph. During larger events, the third shallow soil water component is activated and reaches its maximum during the falling limb of the hydrograph. This component was responsible for up to almost 80% of the storm flow during the largest observed event with the highest runoff ratio, and remained the dominant event flow component during hydrograph recession. These results agreed within 10% with a two component hydrograph separation model using δ¹⁸O and δD of water to distinguish between “old” and “new” water. We infer that this high [Ge] component represents the activation of long, preferential flow paths in the shallow soil though which event waters acquire an elevated Ge signal during large precipitation events.