High Dissolved Carbon Concentration in Arid Rocky Mountain Streams

Increasing phosphorus loss despite widespread concentration decline in US rivers

The loss of phosphorous (P) from the land to aquatic systems has polluted waters and threatened food production worldwide. Systematic trend analysis of P, a nonrenewable resource, has been challenging, primarily due to sparse and inconsistent historical data. Here, we leveraged intensive hydrometeorological data and the recent renaissance of deep learning approaches to fill data gaps and reconstruct temporal trends. We trained a multitask long short-term memory model for total P (TP) using data from 430 rivers across the contiguous United States (CONUS). Trend analysis of reconstructed daily records (1980-2019) shows widespread decline in concentrations, with declining, increasing, and insignificantly changing trends in 60%, 28%, and 12% of the rivers, respectively. Concentrations in urban rivers have declined the most despite rising urban population in the past decades; concentrations in agricultural rivers however have mostly increased, suggesting not-as-effective controls of nonpoint sources in agriculture lands compared to point sources in cities. TP loss, calculated as fluxes by multiplying concentration and discharge, however exhibited an overall increasing rate of 6.5% per decade at the CONUS scale over the past 40 y, largely due to increasing river discharge. Results highlight the challenge of reducing TP loss that is complicated by changing river discharge in a warming climate.

Emerging investigator series: predicted losses of sulfur and selenium in european soils using machine learning: a call for prudent model interrogation and selection

Reductions in sulfur (S) atmospheric deposition in recent decades have been attributed to S deficiencies in crops. Similarly, global soil selenium (Se) concentrations were predicted to drop, particularly in Europe, due to increases in leaching attributed to increases in aridity. Given its international importance in agriculture, reductions of essential elements, including S and Se, in European soils could have important impacts on nutrition and human health. Our objectives were to model current soil S and Se levels in Europe and predict concentration changes for the 21st century. We interrogated four machine-learning (ML) techniques, but after critical evaluation, only outputs for linear support vector regression (Lin-SVR) models for S and Se and the multilayer perceptron model (MLP) for Se were consistent with known mechanisms reported in literature. Other models exhibited overfitting even when differences in training and testing performance were low or non-existent. Furthermore, our results highlight that similarly performing models based on RMSE or R2 can lead to drastically different predictions and conclusions, thus highlighting the need to interrogate machine learning models and to ensure they are consistent with known mechanisms reported in the literature. Both elements exhibited similar spatial patterns with predicted gains in Scandinavia versus losses in the central and Mediterranean regions of Europe, respectively, by the end of the 21st century for an extreme climate scenario. The median change was -5.5% for S (Lin-SVR) and -3.5% (MLP) and -4.0% (Lin-SVR) for Se. For both elements, modeled losses were driven by decreases in soil organic carbon, S and Se atmospheric deposition, and gains were driven by increases in evapotranspiration.

Escalating Carbon Export from High-Elevation Rivers in a Warming Climate

High-elevation mountains have experienced disproportionately rapid warming, yet the effect of warming on the lateral export of terrestrial carbon to rivers remains poorly explored and understood in these regions. Here, we present a long-term data set of dissolved inorganic carbon (DIC) and a more detailed, short-term data set of DIC, δ13CDIC, and organic carbon from two major rivers of the Qinghai-Tibetan Plateau, the Jinsha River (JSR) and the Yalong River (YLR). In the higher-elevation JSR with ∼51% continuous permafrost coverage, warming (>3 °C) and increasing precipitation coincided with substantially increased DIC concentrations by 35% and fluxes by 110%. In the lower-elevation YLR with ∼14% continuous permafrost, such increases did not occur despite a comparable extent of warming. Riverine concentrations of dissolved and particulate organic carbon increased with discharge (mobilization) in both rivers. In the JSR, DIC concentrations transitioned from dilution (decreasing concentration with discharge) in earlier, colder years to chemostasis (relatively constant concentration) in later, warmer years. This changing pattern, together with lighter δ13CDIC under high discharge, suggests that permafrost thawing boosts DIC production and export via enhancing soil respiration and weathering. These findings reveal the predominant role of warming in altering carbon lateral export by escalating concentrations and fluxes and modifying export patterns.

Deep learning for water quality

Understanding and predicting the quality of inland waters are challenging, particularly in the context of intensifying climate extremes expected in the future. These challenges arise partly due to complex processes that regulate water quality, and arduous and expensive data collection that exacerbate the issue of data scarcity. Traditional process-based and statistical models often fall short in predicting water quality. In this Review, we posit that deep learning represents an underutilized yet promising approach that can unravel intricate structures and relationships in high-dimensional data. We demonstrate that deep learning methods can help address data scarcity by filling temporal and spatial gaps and aid in formulating and testing hypotheses via identifying influential drivers of water quality. This Review highlights the strengths and limitations of deep learning methods relative to traditional approaches, and underscores its potential as an emerging and indispensable approach in overcoming challenges and discovering new knowledge in water-quality sciences.

Enhanced Role of Streamflow Processes in the Evolutionary Trends of Dissolved Organic Carbon

Investigating dissolved organic carbon (DOC) dynamics and drivers in rivers enhances the understanding of carbon-environment linkages and support sustainability. Previous studies did not fully consider the dynamic nature of key drivers that influence the long-term changing trends in DOC concentration over time (the controlling factors and their roles in DOC trend can undergo alterations over time). We analyzed 42 years (1979-2018) of hydrometeorology, sulfate SO4, and DOC data from a 5.42 km2 watershed in central-southern Ontario, Canada. Our findings reveal a significant (p ≤ 0.01) overall increase in DOC concentrations, mainly due to the coevolution of SO4 and streamflow trends, especially the extreme flows. Over the 42-year period, the changing trend of streamflow (especially the extreme high or low flows) have significantly (p < 0.05) intensified their influence on DOC trends, increasing by an average of 30%. Conversely, the impact of SO4 has weakened, experiencing an average decrease of 32.6%. The upward trend in the annual average DOC concentration is attributed to the increasing number of maximum flow days within a year, while the decreasing trend in the number of minimum flow days has a contrasting effect. In other words, changes in maximum and minimum flow days have a counteracting effect on the DOC concentration trends. These results underscore the importance of considering the effects of altered streamflow processes on carbon cycle changes under evolving environmental conditions.

The recovery of river chemistry from acid rain in the Mississippi River basin amid intensifying anthropogenic activities and climate change

Acid rain has degraded the environmental health of many regions worldwide since the Industrial Revolution. Signatures of river chemistry recovery from acid rain since the Clean Air Act and similar legislation have been reported extensively in small streams but are often subdued or masked in large rivers by complex, co-occurring drivers. Here we assess the recovery of river chemistry from acid rain deposition in the Mississippi River Basin (MRB), the largest river basin in North America. We combine analysis of temporal trends of acid rain indicator solutes with Bayesian statistical models to assess the large-scale recovery from acid rain and characterize effects of anthropogenic activities. We found evidence of river chemistry recovery from acid rain; however, the effects of other anthropogenic activities, including fertilizer application and road salting, and changing climate, are likely intensifying. Trends of pH, alkalinity and SO4 export suggest acid rain recovery at large in the MRB, with stronger evidence of recovery in the historically afflicted eastern region of the basin. The concentrations of acid rain indicators generally correlate positively to NO3 and Cl, indicating that N-fertilizer application may have significantly increased weathering, and possibly acidification, and road salt application likely increased cation loss from catchments and contributed to SO4 export. Temperature correlates positively with solute concentrations, possibly through respiration-driven weathering or evaporation. The concentrations of acid rain indicators correlate negatively and most strongly to discharge, indicating discharge as a predominant driver and that lower discharge during droughts can elevate concentrations of riverine solutes in a changing climate. Using long-term data, this study represents a rare, comprehensive assessment of the recovery from acid rain in a large river basin, taking into consideration the entangled effects of multiple human activities and climate change. Our results highlight the ever-present need for adaptive environmental management in a constantly changing world.