This repository contains the dataset RADR generated from the work Recent changes to Arctic river discharge (2021), Nature Communications, DOI: 10.1038/s41467-021-27228-1

Lakes are central components of the inland water system distinct from, yet inextricably connected to, river networks. Currently, existing network‐scale biogeochemistry research, although robust, typically treats each of these components separately or reductively. Here, we incorporate lake morphometry into a fully connected stream/lake network for the Connecticut River watershed and model potential evasion of terrestrially sourced headwater CO 2 as transported through the network, ignoring in‐stream production. We found that approximately 25%–30% of total potential soil CO 2 evasion occurs in lakes, and percent evasion is inversely related to streamflow. A lake's ability to evade CO 2 is controlled by residence time and size: most lakes with residence time over 7 days or surface area greater than 0.004 km 2 evade functionally all terrestrial CO 2 entering from upstream, precluding further downstream transport. We conclude that lakes are important for soil CO 2 degassing and that this coupled river/lake approach is promising for CO 2 studies henceforth.

In situ river discharge estimation is a critical component of studying rivers. A dominant method for establishing discharge monitoring in situ is a temporary gauge, which uses a rating curve to relate stage to discharge. However, this approach is constrained by cost and the time to develop the stage‐discharge rating curve, as rating curves rely on numerous flow measurements at high and low stages. Here, we offer a novel alternative approach to traditional temporary gauges: estimating Discharge via Arrays of Pressure Transducers (DAPT). DAPT uses a Bayesian discharge algorithm developed for the upcoming Surface Water Ocean Topography satellite (SWOT) to estimate in situ discharge from automated water surface elevation measurements. We conducted sensitivity tests over 4,954 model runs on five gauged rivers and conclude that the DAPT method can robustly reproduce discharge with an average Nash‐Sutcliffe Efficiency (NSE) of 0.79 and Kling‐Gupta Efficiency of 0.78. Further, we find that the DAPT method estimates discharge similarly to an idealized temporary gauge created from the same input data (NSE differences of less than 0.1), and that results improve significantly with accurate priors. Finally, we test the DAPT method in nine poorly gauged rivers in a realistic and complex field setting in the Peace‐Athabasca Delta, and show that the DAPT method largely outperforms a temporary gauge in this time and budget constrained setting. We therefore recommend DAPT as an effective tool for in situ discharge estimation in cases where there is not enough time or resources to develop a temporary gauge.

This dataset contains georeferenced three-band orthomosaics of green, red, and near-infrared (NIR) digital imagery at 1m resolution collected over selected surface waters across Alaska and Canada between July 9 and August 17, 2017. The orthomosaics were generated from individual images collected by a Cirrus Designs Digital Camera System (DCS) mounted on a Beechcraft Super King Air B200 aircraft from approximately 8-11 km altitude. Flights were over the following areas: Saskatchewan River, Saskatoon, Inuvik, Yukon River including Yukon Flats, Sagavanirktok River, Arctic Coastal Plain, Old Crow Flats, Peace-Athabasca Delta, Slave River, Athabasca River, Yellowknife, Great Slave Lake, Mackenzie River and Delta, Daring Lake, and other selected locations. Most locations were imaged twice during two flight campaigns in Canada and Alaska extending roughly SE-NW then NW-SE up to a month apart. The data were georeferenced using 303 ground control points (GCPs) across the study region.

Recent advances in remote sensing and the upcoming launch of the joint NASA/CNES/CSA/UKSA Surface Water and Ocean Topography (SWOT) satellite point toward improved river discharge estimates in ungauged basins. Existing discharge methods rely on "prior river knowledge" to infer parameters not directly measured from space. Here, we show that discharge estimation is improved by classifying and parameterizing rivers based on their unique geomorphology and hydraulics. Using over 370,000 in situ hydraulic observations as training data, we test unsupervised learning and an "expert" method to assign these hydraulics and geomorphology to rivers via remote sensing. This intervention, along with updates to model physics, constitutes a new method we term "geoBAM," an update of the Bayesian At-many-stations hydraulic geometry-Manning's (BAM) algorithm. We tested geoBAM on Landsat imagery over more than 7,500 rivers (108 are gauged) in Canada's Mackenzie River basin and on simulated hydraulic data for 19 rivers that mimic SWOT observations without measurement error. geoBAM yielded considerable improvement over BAM, improving the median Nash-Sutcliffe efficiency (NSE) for the Mackenzie River from -0.05 to 0.26 and from 0.16 to 0.46 for the SWOT rivers. Further, NSE improved by at least 0.10 in 78/108 gauged Mackenzie rivers and 8/19 SWOT rivers. We attribute geoBAM improvement to parameterizing rivers by type rather than globally, but prediction accuracy worsens if parameters are misassigned. This method is easily mapped to rivers at the global scale and paves the way for improving future discharge estimates, especially when coupled with hydrologic models.

The Surface Water Ocean Topography (SWOT) satellite mission expected to launch in 2021 will offer a unique opportunity to map river discharge at an unprecedented spatial resolution globally from observations of water surface elevation, width, and slope. Because river discharge will not be directly observed from SWOT, a number of algorithms of varying complexity have been developed to estimate discharge from SWOT observables. Outstanding issues include the lack of accurate prior information and parameter equifinality. We developed a new data assimilation discharge algorithm that aimed to overcome these limitations by integrating a data‐driven approach to estimate priors with a model informed by hydraulic geometry relations. A comprehensive simulated dataset of 18 rivers was used to evaluate the algorithm and four different configurations (rectangular channel, generic channel, and geomorphologically classified channel with and without regularization) to assess the impact of progressively adding hydraulic geometry constraints to the estimation problem. The algorithm with the full set of constraints outperformed the other configurations with median Nash‐Sutcliffe coefficients of 0.77, compared with −0.46, 0.31 and 0.66, while other error metrics showed similar improvement. Results from this study show the promise of this hybrid data‐driven approach to estimating river discharge from SWOT observations, although a number of enhancements need to be tested to improve the operational applicability of the algorithm.