Flow Extremes as Spatiotemporal Control Points on River Solute Fluxes and Metabolism

Abstract

Floods are dominant controls on export of solutes from catchments. In contrast, low‐flow periods such as droughts are potentially dominant control points for biogeochemical processing, enhancing spatiotemporal variation in solute concentrations, stream metabolism, and nutrient uptake. Using complementary time series (i.e., an Eulerian reference frame) and longitudinal profiling (i.e., a Lagrangian reference frame), we investigated hydrologic controls on temporal and spatial variation in solute flux and metabolism in the Lower Santa Fe River (FL, USA), where highly colored surface water mixes with exceptionally clear groundwater from springs. Gage measurements suggest groundwater inputs ranged from <1% (during extreme floods) to 86% (during extreme drought) of total discharge (Q). Mass transport of most solutes was dominated by high‐Q periods. Most solute C‐Q relationships exhibited statistically significant slope breakpoints near the transition between surface and groundwater dominance. In particular, parameters controlling water column light attenuation were chemostatic above median Q but markedly reduced at low Q. As a result, river metabolism and assimilatory nitrate (NO3−) uptake were consistently suppressed at high Q and enhanced at low Q, with greater variability in response to drivers other than water column light transmittance. Spatial variation in solute concentrations was also enhanced at low Q, induced by discrete groundwater inflow and biogeochemical processing along the reach. Contrasting reference frames yielded corroborative evidence for transport dominance at high Q, which damps spatiotemporal heterogeneity. In contrast, low‐Q periods enable localized mixing controls on solute concentrations and high rates of metabolism and nutrient processing that increase spatiotemporal variability.

Evaluating spatiotemporal variation in water chemistry of the upper Colorado River using longitudinal profiling

“…The data presented here took eight people 2 weeks to collect, a month to process, and resulted in only a single set of observations under a very specific (and somewhat anomalous) set of hydrologic and climatic conditions. Because there was less water in the river, effects of discontinuities such as lateral inputs, atmospheric heating, and internal biogeochemical processing were amplified, enabling greater spatiotemporal responses to these and other drivers of variation (Hensley et al, ). Fully understanding the spatiotemporal variation of the system requires the ability to easily collect repeated profiles on this scale.…”

“…One approach is to incorporate both fixed‐site and longitudinal data into reactive transport models (Hensley et al, ). Elsewhere, repeated sampling, either day and night (Hensley et al, ; Kunz et al, ) or under different flow condition (Hensley et al, ), have been used to better parse spatial versus temporal variation, but with the obvious constraint of expanding field logistical requirements discussed above. The results suggest profiling is most useful for quantifying spatially distributed inflows, both discrete and diffuse.…”

“…Although earlier grab‐sample‐based spatial profiling undoubtedly advanced our understanding of drivers of spatial heterogeneity (e.g., Dent & Grimm, 1999; Sullivan & Drever, ; Likens & Buso, ), sensors greatly increase the frequency and ease at which measurements can be made, expanding both the resolution and the extents that can be profiled. Multiple recent studies have demonstrated the insights that can be gained from sensor‐based longitudinal profiling of properties including light (Gardner, Ensign, Houser, & Doyle, ), temperature (Ensign et al, ), nutrients (Hensley, Cohen, & Korhnak, ; Kunz, Hensley, Brase, Borchardt, & Rode, ), colour, and dissolved oxygen (DO; Hensley, Kirk, Spangler, Gooseff, & Cohen, ), from reach to regional scales (Gruberts, Paidere, Škute, & Druvietis, ; Kobayashi & Iwata, ; Loken et al, ).…”

River network alteration of C-N-P dynamics in a mesoscale agricultural catchment

“…This issue is reflected in our A-J o u r n a l P r e -p r o o f Q plots, where dates identified as unstable diverge from the regression line for DOC and TP (Figure 7 D and F). To address this issue, future research would need to include highresolution monitoring equipment (Wollheim et al, 2017;Jarvie et al, 2018), Lagrangian sampling (Hensley et al, 2019;Ritz and Fischer, 2019) or time-integrated monitoring techniques (Zabiegala et al, 2010;Knutsson et al, 2013). It is also possible that our assumption of constant delivery from point sources, which we verified on two sampling dates upstream and downstream of each of the seven WWTPs in the catchment, did not capture episodic TP release from some of the WWTPs during storm events.…”

Is the River a Chemostat?: Scale Versus Land Use Controls on Nitrate Concentration‐Discharge Dynamics in the Upper Mississippi River Basin

“…Exponents near 0 indicate chemostatic solute behavior, and strongly negative or positive exponents indicate chemodynamic behavior. Extending this framework, piecewise power law relationships, with different exponents above and below a discharge threshold, often offer substantially better predictive ability and suggest that different solute generation mechanisms are dominant under disparate flow regimes (Hensley et al, 2019; Moatar et al, 2017). A related C‐Q metric is CV C /CV Q , the ratio of the coefficients of variation of concentration and discharge (Thompson et al, 2011).…”