Search for: All records

Abstract While inland freshwater networks cover less than 4% of the Earth's terrestrial surface, these ecosystems play a disproportionately large role in the global cycles of [C]arbon, [N]itrogen, and [P]hosphorus, making streams and rivers critical regulators of nutrient balance at regional and continental scales. Foundational studies have established the relative importance of the hydrologic regime, land cover, and instream removal processes for controlling the transport and processing of C, N, and P in river networks. However, particulate C, N, and P can make up a large proportion of the total material in large rivers and during high flows. To constrain the patterns of the biogeochemistry of riverine particulates, we characterized and modeled dissolved and particulate concentration variability at the continental scale using open‐access data from 27 National Ecological Observatory Network (NEON) sites across the United States. We analyzed these data using Boosted Regression Trees (BRTs) to statistically identify if land cover characteristics could predict nutrient quantity and quality of stream particulates. The BRT models revealed that land cover does not strongly predict particulate dynamics across NEON sites but indicate that instream processes might be more important than catchment characteristics alone. In addition, our study demonstrates the consistent importance of particulates relative to dissolved forms, highlighting their likely significance for biogeochemical processes along the freshwater continuum. 

Abstract Studies of stream macroinvertebrates traditionally use sampling methods that target benthic habitats. These methods could underestimate biodiversity if important assemblage components exist outside of the benthic zone. To test the efficacy of different sampling methods, we collected paired reach‐wide benthic and edge samples from up to 10 study reaches in nine basins spanning an aridity gradient across the United States. Edge sampling targeted riparian‐adjacent microhabitats not typically sampled, including submerged vegetation, roots, and overhanging banks. We compared observed richness, asymptotic richness, and assemblage dissimilarity between benthic samples alone and different combinations of benthic and edge samples to determine the magnitude of increased diversity and assemblage dissimilarity values with the addition of edge sampling. We also examined how differences in richness and assemblage composition varied across an aridity gradient. The addition of edge sampling significantly increased observed richness (median increase = 29%) and asymptotic richness (median increase = 173%). Similarly, median Bray–Curtis dissimilarity values increased by as much as 0.178 when benthic and edge samples were combined. Differences in richness metrics were generally higher in arid basins, but assemblage dissimilarity either increased or decreased across the aridity gradient depending on how benthic and edge samples were combined. Our results suggest that studies that do not sample stream edges may significantly underestimate reach diversity and misrepresent assemblage compositions, with effects that can vary across climates. We urge researchers to carefully consider sampling methods in field studies spanning climatic zones and the comparability of existing data sets when conducting data synthesis studies. 

Abstract The interaction of climate change and increasing anthropogenic water withdrawals is anticipated to alter surface water availability and the transport of carbon (C), nitrogen (N), and phosphorus (P) in river networks. But how changes to river flow will alter the balance, or stoichiometry, of these fluxes is unknown. The Lower Flint River Basin (LFRB) is part of an interstate watershed relied upon by several million people for diverse ecosystem services, including seasonal crop irrigation, municipal drinking water access, and public recreation. Recently, increased water demand compounded with intensified droughts have caused historically perennial streams in the LFRB to cease flowing, increasing ecosystem vulnerability. Our objectives were to quantify how riverine dissolved C:N:P varies spatially and seasonally and determine how monthly stoichiometric fluxes varied with overall water availability in a major tributary of LFRB. We used a long‐term record (21–29 years) of solute water chemistry (dissolved organic carbon, nitrate/nitrite, ammonia, and soluble reactive phosphorus) paired with long‐term stream discharge data across six sites within a single LFRB watershed. We found spatial and seasonal differences in soluble nutrient concentrations and stoichiometry attributable to groundwater connections, the presence of a major floodplain wetland, and flow conditions. Further, we showed that water availability, as indicated by the Palmer Drought Severity Index (PDSI), strongly predicted stoichiometry with generally lower C:N and C:P and higher N:P fluxes during periods of low water availability (PDSI < −4). These patterns suggest there may be long‐term and significant changes to stream ecosystem function as water availability is being dramatically altered by human demand with consequential impacts on solute transport, in‐stream processing, and stoichiometric ratios.