Quantifying how watershed structure influences the exchanges of water among component parts of a watershed, particularly the connection between uplands, valley bottoms, and in‐stream hydrologic exchange, remains a challenge. However, this understanding is critical for ascertaining the source areas and temporal contributions of water and associated biogeochemical constituents in streams. We used dilution gauging, mass recovery, and recording discharge stations to characterize streamflow dynamics across 52 reaches, from peak snowmelt to base flow, in the Tenderfoot Creek Experimental forest, Montana, USA. We found that watershed‐contributing area was only a significant predictor of net changes in streamflow at high moisture states and larger spatial scales. However, at the scale of individual stream reaches, the lateral contributing area in conjunction with underlying lithology and vegetation densities were significant predictors of gross hydrologic gains to the stream. Reach lateral contributing areas underlain by more permeable sandstone yielded less water across flow states relative to those with granite gneiss. Additionally, increases in the frequency of steps across each stream reach contributed to greater hydrologic gross losses. Together, gross gains and losses of water along individual reaches resulted in net changes of discharge that cumulatively scale to the observed outlet discharge dynamics. Our results provide a framework for understanding how hillslope topography, geology, vegetation, and valley bottom structure contribute to the exchange of water and cumulative increases of stream flow across watersheds of increasing size.

Watershed structure influences the timing, magnitude, and spatial location of water and solute entry to stream networks. In turn, stream reach transport velocities and stream network geometry (travel distances) further influence the timing of export from watersheds. Here, we examine how watershed and stream network organization can affect travel times of water from delivery to the stream network to arrival at the watershed outlet. We analysed watershed structure and network geometry and quantified the relationship between stream discharge and solute velocity across six study watersheds (11.4 to 62.8 km 2 ) located in the Sawtooth Mountains of central Idaho, USA. Based on these analyses, we developed stream network travel time functions for each watershed. We found that watershed structure, stream network geometry, and the variable magnitude of inputs across the network can have a pronounced affect on water travel distances and velocities within a stream network. Accordingly, a sample taken at the watershed outlet is composed of water and solutes sourced from across the watershed that experienced a range of travel times in the stream network. We suggest that understanding and quantifying stream network travel time distributions are valuable for deconvolving signals observed at watershed outlets into their spatial and temporal sources, and separating terrestrial and in-channel hydrological, biogeochemical, and ecological influences on in-stream observations. Figure 4. Progression (vertically) of the calculated travel time function for all six study watersheds (horizontally). The basis of the travel time function is the distribution of network distances (the width function) (a). Travel time can be calculated with a constant velocity (b) or a variable velocity (c). The travel times can be weighted by lateral inflows to the stream network (d). When all pieces are incorporated, the result is the inflow weighted variable velocity travel time function (e) presented as proportion of total discharge (Q) in the network at a given time 2677 STREAM NETWORK WATER AND SOLUTE TRAVEL TIMES a Mean and maximum inflow weighted travel time (IWTT) in hours and skewness of each watershed calculated using both and constant and variable velocity (CV and VV, respectively). 2678 A. BERGSTROM ET AL. DISCUSSIONWe present analysis of how watershed structure and stream network geometry can affect solute travel times and discuss implications for interpreting observed watershed outflow signatures. We developed an IW-VV-TTF that represents the probability distribution of conservative solute travel times, as a surrogate for the water molecules themselves, from entry into the stream network to a given watershed outlet. The IW-VV-TTF builds on the commonly used width function and combines direct measures of watershed structure, stream network geometry, and solute velocity to estimate stream network travel time distributions. Network geometry sets the distribution of travel distances for water and solutes from entry to the stream network to the water...

Abstract. The McMurdo Dry Valleys (MDVs) of Antarctica are a polar desert ecosystem consisting of alpine glaciers, ice-covered lakes, streams, and expanses of vegetation-free rocky soil. Because average summer temperatures are close to 0 ∘C, the MDV ecosystem in general, and glacier melt dynamics in particular, are both closely linked to the energy balance. A slight increase in incoming radiation or change in albedo can have large effects on the timing and volume of meltwater. However, the seasonal evolution or spatial variability of albedo in the valleys has yet to fully characterized. In this study, we aim to understand the drivers of landscape albedo change within and across seasons. To do so, a box with a camera, GPS, and shortwave radiometer was hung from a helicopter that flew transects four to five times a season along Taylor Valley. Measurements were repeated over three seasons. These data were coupled with incoming radiation measured at six meteorological stations distributed along the valley to calculate the distribution of albedo across individual glaciers, lakes, and soil surfaces. We hypothesized that albedo would decrease throughout the austral summer with ablation of snow patches and increasing sediment exposure on the glacier and lake surfaces. However, small snow events (<6 mm water equivalent) coupled with ice whitening caused spatial and temporal variability of albedo across the entire landscape. Glaciers frequently followed a pattern of increasing albedo with increasing elevation, as well as increasing albedo moving from east to west laterally across the ablation zone. We suggest that spatial patterns of albedo are a function of landscape morphology trapping snow and sediment, longitudinal gradients in snowfall magnitude, and wind-driven snow redistribution from east to west along the valley. We also compare our albedo measurements to the MODIS albedo product and found that overall the data have reasonable agreement. The mismatch in spatial scale between these two datasets results in variability, which is reduced after a snow event due to albedo following valley-scale gradients of snowfall magnitude. These findings highlight the importance of understanding the spatial and temporal variability in albedo and the close coupling of climate and landscape response. This new understanding of landscape albedo can constrain landscape energy budgets, better predict meltwater generation on from MDV glaciers, and how these ecosystems will respond to changing climate at the landscape scale.

In low‐nutrient streams in cold and arid ecosystems, the spiraling of autochthonous particulate organic matter (POM) may provide important nutrient subsidies downstream. Because of its lability and the spatial heterogeneity of processing in hyporheic sediments, the downstream transport and fate of autochthonous POM can be difficult to trace. In Antarctic McMurdo Dry Valley (MDV) streams, any POM retained in the hyporheic zone is expected to be derived from surface microbial mats that contain diatoms with long‐lasting silica frustules. We tested whether diatom frustules can be used to trace the retention of autochthonous POM in the hyporheic zone and whether certain geomorphic locations promote this process. The accumulation of diatom frustules in hyporheic sediments, measured as biogenic silica, was correlated with loss‐on‐ignition organic matter and sorbed ammonium, suggesting that diatoms can be used to identify locations where POM has been retained and processed over long timescales, regardless of whether the POM remains intact. In addition, by modeling the upstream sources of hyporheic diatom assemblages, we found that POM was predominantly derived from N‐fixing microbial mats of the genus Nostoc. In terms of spatial variability, we conclude that the hyporheic sediments adjacent to the stream channel that are regularly inundated by daily flood pulses are where the most POM has been retained over long timescales. Autochthonous POM is retained in hyporheic zones of low‐nutrient streams beyond the MDVs, and we suggest that biogenic silica and diatom composition can be used to identify locations where this transfer is most prevalent.

Concentration‐discharge (C‐Q) relationships are often used to quantify source water contributions and biogeochemical processes occurring within catchments, especially during discrete hydrological events. Yet, the interpretation of C‐Q hysteresis is often confounded by complexity of the critical zone, such as numerous source waters and hydrochemical nonstationarity. Consequently, researchers must often ignore important runoff pathways and geochemical sources/sinks, especially the hyporheic zone because it lacks a distinct hydrochemical signature. Such simplifications limit efforts to identify processes responsible for the transience of C‐Q hysteresis over time. To address these limitations, we leverage the hydrologic simplicity and long‐term, high‐frequency Q and electrical conductivity (EC) data from streams in the McMurdo Dry Valleys, Antarctica. In this two end‐member system, EC can serve as a proxy for the concentration of solutes derived from the hyporheic zone. We utilize a novel approach to decompose loops into subhysteretic EC‐Q dynamics to identify individual mechanisms governing hysteresis across a wide range of timescales. We find that hydrologic and hydraulic processes govern EC response to diel and seasonal Q variability and that the effects of hyporheic mixing processes on C‐Q transience differ in short and long streams. We also observe that variable hyporheic turnover rates govern EC‐Q patterns at daily to interannual timescales. Last, subhysteretic analysis reveals a period of interannual freshening of glacial meltwater streams related to the effects of unsteady flow on hyporheic exchange. The subhysteretic analysis framework we introduce may be applied more broadly to constrain the processes controlling C‐Q transience and advance understanding of catchment evolution.

In polar regions, where many glaciers are cold based (frozen to their beds), biological communities on the glacier surface can modulate and transform nutrients, controlling downstream delivery. However, it remains unclear whether supraglacial streams are nutrient sinks or sources and the rates of nutrient processing. In order to test this, we conducted tracer injections in three supraglacial streams (62 to 123 m long) on Canada Glacier in the Taylor Valley, of the McMurdo Dry Valleys, Antarctica. We conducted a series of additions including nitrate (N), N + phosphate (P), N + P + glucose (C), and N + C. In two reaches, N-only additions resulted in N uptake. The third reach showed net N release during the N-only addition, but high N uptake in the N + P addition, indicating P-limitation or N + P colimitation. Coinjecting C did not increase N-uptake. Additionally, in these systems at low N concentrations the streams can be a net source of nitrogen. We confirmed these findings using laboratory-based nutrient incubation experiments on sediment collected from stream channels on Canada Glacier and two other glaciers in the Taylor Valley. Together, these results suggest there is active biological processing of nutrients occurring in these supraglacial streams despite low sediment cover, high flow velocities, and cold temperatures, modifying the input signals to proglacial streams. As glaciers worldwide undergo rapid change, these findings further our understanding of how melt generated on glacier surfaces set the initial nutrient signature for subglacial and downstream environments. Plain Language Summary In polar regions most glaciers are frozen to their beds, meaning that meltwater from these glaciers comes from the surface, or "supraglacial" environment. Active biological communities exist in this supraglacial environment where they generate and recycle nutrients. However, it remains unclear whether nutrients are removed by the supraglacial streams, and what limits this removal process. We did experimental nutrient additions in three streams on Canada Glacier in the Taylor Valley of the McMurdo Dry Valleys, Antarctica. We found that nitrogen in the form of nitrate was quickly removed. In one of three locations, the addition of phosphorus stimulated more nitrogen removal than just adding nitrogen alone. We also found that at low nitrogen concentrations, the streams can be a net source of nitrogen. We confirmed these field experiment findings using laboratory nutrient incubation experiments on sediment collected from stream channels on Canada Glacier and two other glaciers in the Taylor Valley. Supraglacial streams act as a filter and can remove nitrogen from meltwater before it leaves the glacier and affect the quantity and type of nutrients transported to downstream ecosystems. This is important worldwide because glaciers are often the headwaters of the stream network.

Glacierized watersheds play an important role in global hydrologic and biogeochemical cycles (e.g., Hood et al., 2015;Milner et al., 2017) and regional ecology (Cauvy-Fraunié & Dangles, 2019). For example, they sustain late-summer streamflow and maintain colder water temperatures relative to adjacent nonglacierized watersheds (

Glaciers of the McMurdo dry valleys (MDVs) Antarctica are the main source of streamflow in this polar desert. Because summer air temperatures hover near 0 C small changes in the energy balance strongly affect meltwater generation. Here we demonstrate that increased surface roughness, which alters the turbulent transfer of energy between the ice surface and atmosphere, yields a detectable increase in meltwater runoff. At low elevations on the glaciers, basin-like topography became significantly rougher over 13 years between repeat lidar surveys, yielding greater melt. In