Seasonal patterns in streamwater nutrient and dissolved organic carbon concentrations: Separating catchment flow path and in-stream effects

Distinct seasonal patterns in streamwater nutrient and dissolved organic carbon (DOC) concentrations are evident in the data record from 7 years of weekly sampling in the West Fork of Walker Branch (WB) and 4 years of weekly sampling in a nearby stream, upper White Oak Creek (WOC), both first-order streams in the Ridge and Valley Province of eastern Tennessee. Concentrations of NO3 and soluble reactive phosphorus (SRP) in both streams showed a repeated pattern of annual maxima in summer and biannual minima in autumn and spring. Concentrations of DOC in WB exhibited distinct autumn maxima. To determine whether temporal variations in catchment hydrological processes could explain the seasonal nutrient and DOC variations in WB, we used an end-member mixing analysis involving Ca and SO4 concentrations to separate stream discharge into three catchment how paths of differing nutrient concentrations. Stream NO3, SRP, and DOC concentrations were predicted solely on the basis of temporal variation in the importance of these how paths and measurements of nutrient concentrations for the different flow paths. Ratios of observed/predicted concentrations in stream water near 1.0 suggested that catchment effects alone explained streamwater concentrations, whereas ratios substantially different from 1.0 suggested that in-stream processes were important determinants of streamwater concentrations. Observed/predicted NO3 and SRP concentration ratios showed repeated annual patterns with values closer to 1.0 during winter and summer (generally 0.8-1.2) and minima (<0.6) in spring and autumn, suggesting substantial in-stream net uptake at these times. Observed/predicted DOC concentration ratios were more variable and generally greater than or equal to 1 but did show consistent autumn maxima (>2.5), indicating substantial in-stream DOC generation at this time. Observed/predicted ratios for all nutrients were generally less variable and were closer to 1.0 at high flow compared to low flow, suggesting that in-stream controls on streamwater chemistry are less important at high discharge than at low discharge. Our results indicate two general modes of control of stream nutrient concentrations: (1) catchment control via seasonal variation in the dominant hydrologic pathway (greater proportion of deep groundwater in summer), which produces lower winter and higher summer concentrations, and (2) in-stream control via high rates of net nutrient uptake during the spring (primarily by autotrophs) and autumn (primarily by heterotrophs).