Probing Subsurface and Stream Particle Composition Through Optical Analysis at the Eel River Critical Zone Observatory

Abstract:

Particles constitute a significant portion of the flux weathered material from continents to ocean basins but little is understood about their seasonal dynamics particularly in subsurface and headwater stream environments. At the Eel River Critical Zone Observatory, located near the headwaters of the South Fork Eel River in the Angelo Coast Range Reserve (Northern California), groundwater from weathered bedrock and stream waters are sampled at a frequency of one to three days from three wells (Well 1 down-slope, Well 3 mid-slope, Well 10 upper-slope) and Elder Creek. Approximately one thousand samples collected by automated ISCO Gravity Filtration System (GFS; Kim et al. 2012, EST) since 2011 have been filtered through 0.45 μm 25 mm diameter Supor filters. Filters imaged under controlled lighting are analyzed for red, green, and blue optical density (OD) to enable rapid assessment of sample loading and color as a prelude to and selection aid for more labor-intensive ICP-MS and Scanning Electron Microscopic analysis. For example, samples with lower red OD relative to green and blue may correspond to samples high in Mn/Fe oxides. Optical imaging of the loaded filters provides a time-series over three years and color anomalies in these filters along with chemical analysis of dissolved and particulate filtrate is used to establish a method for calibrating optical data to interpret chemical composition of water and particles. Results are interpreted within a framework of environmental data such as rainfall, stream discharge and turbidity, and water table depth measured at the heavily monitored forested hillslope. Data from the four locations range up to 0.6 OD units with a typical detection limit of better than 0.01 OD units. At Well 10, wet season filter samples exhibit highest particle loading (OD ~ 0.3) with values rapidly decreasing during the dry season (OD < D.L.) water table recession. At Well 1, particle loads instantaneously reflect intense rain events. Applied at a larger scale, this method - if proven - may be used to estimate basin level particulate flux with an estimation of chemical composition in a highly efficient manner.