Riverine exports of silicon (Si) influence global carbon cycling through the growth of marine diatoms, which account for ∼25% of global primary production. Climate change will likely alter river Si exports in biome‐specific ways due to interacting shifts in chemical weathering rates, hydrologic connectivity, and metabolic processes in aquatic and terrestrial systems. Nonetheless, factors driving long‐term changes in Si exports remain unexplored at local, regional, and global scales. We evaluated how concentrations and yields of dissolved Si (DSi) changed over the last several decades of rapid climate warming using long‐term data sets from 60 rivers and streams spanning the globe (e.g., Antarctic, tropical, temperate, boreal, alpine, Arctic systems). We show that widespread changes in river DSi concentration and yield have occurred, with the most substantial shifts occurring in alpine and polar regions. The magnitude and direction of trends varied within and among biomes, were most strongly associated with differences in land cover, and were often independent of changes in river discharge. These findings indicate that there are likely diverse mechanisms driving change in river Si biogeochemistry that span the land‐water interface, which may include glacial melt, changes in terrestrial vegetation, and river productivity. Finally, trends were often stronger in months outside of the growing season, particularly in temperate and boreal systems, demonstrating a potentially important role of shifting seasonality for the flux of Si from rivers. Our results have implications for the timing and magnitude of silica processing in rivers and its delivery to global oceans.

Evaluation of carbon (C), nitrogen (N), and phosphorus (P) ratios in aquatic and terrestrial ecosystems can advance our understanding of biological processes, nutrient cycling, and the fate of organic matter (OM) in these ecosystems. Eutrophication of aquatic ecosystems can change the accumulation and decomposition of OM which can alter biogeochemical cycling and alter the base of the aquatic food web. This study investigated nutrient stoichiometry within and among wetland ecosystem compartments (i.e., water column, flocculent, soil, and aboveground vegetation biomass) of two subtropical treatment wetlands with distinct vegetation communities. Two flow-ways (FWs) within the network of Everglades Stormwater Treatment Areas in south Florida (USA) were selected for this study. We evaluated nutrient stoichiometry of these to understand biogeochemical cycling and controls of nutrient removal in a treatment wetland within an ecological stoichiometry context. This study demonstrates that C, N, and P stoichiometry can be highly variable among ecosystem compartments and between FWs. Power law slopes of C, N, and P within surface water floc, soil, and vegetation were significantly different between and along FWs. Assessment of wetland nutrient stoichiometry between and within ecosystem compartments suggests unconstrained stoichiometry related to P that conforms with the notion of P limitation in the ecosystem. Differences in N:P ratios between floc and soil suggest different pathways of organic nutrient accumulation and retention between FWs. Surface nutrient stoichiometry was highly variable and decoupled (or close to decoupled as indicated by < 25% explained variation between parameters), in particular with respect to P. We hypothesize that decoupling may be the imprint of variability in inflow nutrient stoichiometry. However, despite active biogeochemical cycles that could act to restore nutrient stoichiometry along the FW, there was little evidence that such balancing occurred, as the degree of stochiometric decoupling in the water column did change with distance downstream. This information is only the beginning of a larger journey to understand stoichiometric processes within wetland ecosystems and how they relate to ecosystem function.

Hurricanes cause landscape-scale disturbances that affect biogeochemical cycling and water quality in coastal ecosystems. During Hurricane Irma’s passage through northern Florida, water movements driven by wind velocities up to 105 km h⁻¹ caused a salinity peak in an estuary/blackwater river complex. Water quality was monitored across the 15 km site to detect the magnitude and duration of disturbance. Saline water intruded 15 km inland into a freshwater portion of the river that peaked at a salinity of 2 psu. Due to the volume of precipitation from the hurricane, significant runoff of freshwater and dissolved organic matter (DOM) caused a decrease in salinity, dissolved oxygen (DO), and Chlorophyll-a concentrations while increasing turbidity and fluorescent dissolved organic matter (fDOM). The disturbance caused rapid changes observed by in-situ water quality monitors over a 3-week period, but some effects persisted for longer periods as shown by 3-month weekly water sampling. This disturbance caused shifts in DOM loading, altered salinity dynamics, and reshaped landscapes due to wind and wave surge both in upland marsh and downstream estuary. Hurricane disturbance temporarily and abruptly alters the aquatic continuum, and observations of system response can help us understand the mechanisms associated with ecosystem resilience and recovery.