Phytoplankton growth in relation to network topology: time‐averaged catchment‐scale modelling in a large lowland river

Abstract

Summary Using the unregulated, eutrophic S omeş R iver ( R omania/ H ungary), a tributary of the U pper T isza R iver as a model system, we related phytoplankton growth to the structure of river network modified by human intervention. The T ime‐ A veraged P hytoplankton I ncrease in R ivers ( TAPIR ) model was developed by adding two new modules (nitrogen emission and phytoplankton growth) to the PHOSFATE model. The latter was a raster‐based, long‐term average GIS tool that computed the flow tree, mean hydrology and phosphorus concentration at fine spatial resolution in the entire catchment. Long‐term monitoring data at the outflow and phytoplankton profiles obtained along the S omeş R iver in 2012 were used to calibrate the models. In summer 2012, a few centric diatom species dominated the phytoplankton, with mean apparent net growth rate of 2.5 day −1 along a ˜100‐km‐long upstream section of the S omeş R iver. This high growth rate was explained by enhanced benthic retention of meroplanktonic diatoms, while further downstream mean algal biomass varied between 80 and 100 mg C hl m −3 . The TAPIR model realistically reproduced observed mean biomass profiles of phytoplankton in the S omeş R iver. According to modelling results, P ‐determined biomass capacity prevented further growth of algae along the lower two‐thirds of the river. Channel slope and form were important reach‐scale determinants of phytoplankton recruitment, while the absence of large tributaries along the S omeş R iver resulted in marginal longitudinal changes in dilution and light availability. Therefore, phytoplankton might achieve high biomass in a wide range of instantaneous net growth rates realised under strongly fluctuating environmental conditions, which combined with the abundant nutrients from urban areas underpins advanced eutrophication. Large montane reservoirs on tributaries increased mean water age at the outflow of the S omeş catchment by an order of magnitude compared with pristine conditions, but had a negligible overall impact on the phytoplankton of the S omeş R iver. Growth of river phytoplankton is intrinsically linked with network topology, including spatial distribution of nutrient inputs. Stochastic variability in discharge tends to destabilise, while specific network properties may stabilise phytoplankton growth in large rivers. Shallow rivers support meroplanktonic diatoms that, depending on discharge, may have adequate time to develop summer blooms, making shallow rivers particularly susceptible to eutrophication.