We measured groundwater apparent age (τ) and seepage rate (v) in a sandy streambed using point‐scale sampling and seepage blankets (a novel seepage meter). We found very similar MTT estimates from streambed point sampling in a 58 m reach (29 years) and a 2.5 km reach (31 years). The TTD for groundwater discharging to the stream was best fit by a gamma distribution model and was very similar for streambed point sampling in both reaches. Between adjacent point‐scale and seepage blanket samples, water from the seepage blankets was generally younger, largely because blanket samples contained a fraction of “young” stream water. Correcting blanket data for the stream water fraction brought τ estimates for most blanket samples closer to those for adjacent point samples. The MTT estimates from corrected blanket data were in good agreement with those from sampling streambed points adjacent to the blankets. Collectively, agreement among age‐dating tracers, general accord between tracer data and piston‐flow model curves, and large groundwater age gradients in the streambed, suggested that the piston flow apparent ages were reasonable estimates of the groundwater transit times for most samples. Overall, our results from two field campaigns suggest that groundwater collected in the streambed can provide reasonable estimates of apparent age of groundwater discharge, and that MTT can be determined from different age‐dating tracers and by sampling with different groundwater collection devices. Coupled streambed point measurements of groundwater age and groundwater seepage rate represent a novel, reproducible, and effective approach to estimating aquifer TTD and MTT.

In this study, we explored the use of statistical machine learning and long-term groundwater nitrate monitoring data to estimate vadose zone and saturated zone lag times in an irrigated alluvial agricultural setting. Unlike most previous statistical machine learning studies that sought to predict groundwater nitrate concentrations within aquifers, the focus of this study was to leverage available groundwater nitrate concentrations and other environmental variables to determine mean regional vertical velocities (transport rates) of water and solutes in the vadose zone and saturated zone (3.50 and 3.75 m yr⁻¹, respectively). The statistical machine learning results are consistent with two primary recharge processes in this western Nebraska aquifer, namely (1) diffuse recharge from irrigation and precipitation across the landscape and (2) focused recharge from leaking irrigation conveyance canals. The vadose zone mean velocity yielded a mean recharge rate (0.46 m yr⁻¹) consistent with previous estimates from groundwater age dating in shallow wells (0.38 m yr⁻¹). The saturated zone mean velocity yielded a recharge rate (1.31 m yr⁻¹) that was more consistent with focused recharge from leaky irrigation canals, as indicated by previous results of groundwater age dating in intermediate-depth wells (1.22 m yr⁻¹). Collectively, the statistical machine learning model results are consistent with previous observations of relatively high water fluxes and short transit times for water and nitrate in the primarily oxic aquifer. Partial dependence plots from the model indicate a sharp threshold in which high groundwater nitrate concentrations are mostly associated with total travel times of 7 years or less, possibly reflecting some combination of recent management practices and a tendency for nitrate concentrations to be higher in diffuse infiltration recharge than in canal leakage water. Limitations to the machine learning approach include the non-uniqueness of different transport rate combinations when comparing model performance and highlight the need to corroborate statistical model results with a robust conceptual model and complementary information such as groundwater age.

We compared three stream‐based sampling methods to study the fate of nitrate in groundwater in a coastal plain watershed: point measurements beneath the streambed, seepage blankets (novel seepage‐meter design), and reach mass‐balance. The methods gave similar mean groundwater seepage rates into the stream (0.3–0.6 m/d) during two 3–4 day field campaigns despite an order of magnitude difference in stream discharge between the campaigns. At low flow, estimates of flow‐weighted mean nitrate concentrations in groundwater discharge ([ ]_FWM) and nitrate flux from groundwater to the stream decreased with increasing degree of channel influence and measurement scale, i.e., [ ]_FWM was 654, 561, and 451 µM for point, blanket, and reach mass‐balance sampling, respectively. At high flow the trend was reversed, likely because reach mass‐balance captured inputs from shallow transient high‐nitrate flow paths while point and blanket measurements did not. Point sampling may be better suited to estimating aquifer discharge of nitrate, while reach mass‐balance reflects full nitrate inputs into the channel (which at high flow may be more than aquifer discharge due to transient flow paths, and at low flow may be less than aquifer discharge due to channel‐based nitrate removal). Modeling dissolved N₂ from streambed samples suggested (1) about half of groundwater nitrate was denitrified prior to discharge from the aquifer, and (2) both extent of denitrification and initial nitrate concentration in groundwater (700–1300 µM) were related to land use, suggesting these forms of streambed sampling for groundwater can reveal watershed spatial relations relevant to nitrate contamination and fate in the aquifer.

Image‐based water level measurements offer data quality assurance through visual verification that no other method can provide. GaugeCam Remote Image Manager‐Educational 2 (GRIME2) is a mature, open‐source commercial friendly software application that automatically detects and measures water level in laboratory and field settings. The software relies on a dedicated target background for water line detection and image calibration. The system detects the change in pixel gray scale values associated with the intersection of the water level at the target surface. Fiducials on the target background are used to precisely create a pixel to real world coordinate transfer matrix and to correct for camera movement. The presented software package implements the algorithms and automates the water level measurement process, annotation of images with result overlays, creation of animations, and output of results to files that can be further analyzed in a spreadsheet or with R or Python. These GRIME2 features are illustrated using imagery from a coastal marsh field site. Tradeoffs between workflow and algorithm complexity and ease of use are discussed and future improvements are identified with the intention that this Findable, Accessible, Interoperable, and Reusable‐inspired software can be adopted, modified and improved by the user community. While image resolution, quality and other factors associated with field deployment (e.g., water surface roughness, sun glare, shadows, and bio‐fouling) will have an impact on measurement quality, previous controlled laboratory testing that did not manifest these issues showed potential for accuracy of ±3 mm (Gilmore et al., 2013, https://doi.org/10.1016/j.jhydrol.2013.05.011).

Image-based stage and discharge measuring systems are among the most promising new non-contact technologies available for long-term hydrological monitoring. This article evaluates and reports the long-term performance of the GaugeCam (www.gaugecam.org) image-based stage measuring system in situ. For this we installed and evaluated the system over several months in a tidal marsh to obtain a good stratification of the measured stages. Our evaluation shows that the GaugeCam system was able to measure within about ±5 mm for a 90% confidence interval over a range of about 1 m in a tidal creek in a remote location of North Carolina, USA. Our results show that the GaugeCam system nearly performed to the desired design of ±3 mm accuracy around 70% of the time. The system uses a dedicated target background for calibration and geometrical perspective correction of images, as well as auto-correction to compensate for camera movement. The correction systems performed well overall, although our results show a ‘croissant-shaped’ mean error (-1 to +4 mm,) varying with water stage. We attribute this to the small, yet present, ‘fish-eye’ effect embedded in images, for which our system did not entirely correct in the tested version, and which might affect all image-based water level measurement systems.