Effects of agricultural land use on fluvial carbon dioxide, methane and nitrous oxide concentrations in a large European river, the Meuse (Belgium).

We report a data-set of CO2, CH4, and N2O concentrations in the surface waters of the Meuse river network in Belgium, obtained during four surveys covering 50 stations (summer 2013 and late winter 2013, 2014 and 2015), from yearly cycles in four rivers of variable size and catchment land cover, and from 111 groundwater samples. Surface waters of the Meuse river network were over-saturated in CO2, CH4, N2O with respect to atmospheric equilibrium, acting as sources of these greenhouse gases to the atmosphere, although the dissolved gases also showed marked seasonal and spatial variations. Seasonal variations were related to changes in freshwater discharge following the hydrological cycle, with highest concentrations of CO2, CH4, N2O during low water owing to a longer water residence time and lower currents (i.e. lower gas transfer velocities), both contributing to the accumulation of gases in the water column, combined with higher temperatures favourable to microbial processes. Inter-annual differences of discharge also led to differences in CH4 and N2O that were higher in years with prolonged low water periods. Spatial variations were mostly due to differences in land cover over the catchments, with systems dominated by agriculture (croplands and pastures) having higher CO2, CH4, N2O levels than forested systems. This seemed to be related to higher levels of dissolved and particulate organic matter, as well as dissolved inorganic nitrogen in agriculture dominated systems compared to forested ones. Groundwater had very low CH4 concentrations in the shallow and unconfined aquifers (mostly fractured limestones) of the Meuse basin, hence, should not contribute significantly to the high CH4 levels in surface riverine waters. Owing to high dissolved concentrations, groundwater could potentially transfer important quantities of CO2 and N2O to surface waters of the Meuse basin, although this hypothesis remains to be tested.

Potamon: a dynamic model for predicting phytoplankton composition and biomass in lowland rivers.

POTAMON is a unidimensional, non-stationary model, designed for simulating potamoplankton from source to mouth. The forcing variables are discharge, river morphology, water temperature, available light and nutrient inputs. Given the description of several algal categories, POTAMON allows to simulate algal "successions" at a particular site, as well as longitudinal changes of potamoplankton composition and biomass. The algal categories differ by their physiology, their loss rates, and their sensitivity to grazing by zooplankton. Two zooplankton categories were considered, Brachionus-like and Keratella-like, which differ by their clearance rate, their incipient limiting level, their selectivity towards phytoplankton, and their growth yield. The model simulates satisfactorily the onset and the magnitude of the phytoplankton spring bloom in the Belgian part of R. Meuse, the biomass decrease in early summer, and the autumn bloom. It also renders the major variations of algal assemblages along the river. The model allows to confirm that the main driving variables of potamoplankton dynamics in a eutrophic river are physical factors: discharge and related variables (e.g. retention time), light and temperature. In addition, the simulations confirm that the zooplankton-phytoplankton interaction may result in phytoplankton biomass fluctuations and compositional changes. POTAMON can be useful to explore plankton dynamics in a large river, and it may become a tool to test various management measures.

Primary production in a tropical large lake: the role of phytoplankton composition.

Phytoplankton biomass and primary production in tropical large lakes vary at different time scales, from seasons to centuries. We provide a dataset made of 7 consecutive years of phytoplankton biomass and production in Lake Kivu (Eastern Africa). From 2002 to 2008, bi-weekly samplings were performed in a pelagic site in order to quantify phytoplankton composition and biomass, using marker pigments determined by HPLC. Primary production rates were estimated by 96 in situ (14)C incubations. A principal component analysis showed that the main environmental gradient was linked to a seasonal variation of the phytoplankton assemblage, with a clear separation between diatoms during the dry season and cyanobacteria during the rainy season. A rather wide range of the maximum specific photosynthetic rate (PBm) was found, ranging between 1.15 and 7.21 g carbong(-1)chlorophyll ah(-1), and was best predicted by a regression model using phytoplankton composition as an explanatory variable. The irradiance at the onset of light saturation (Ik) ranged between 91 and 752 µE m(-2)s(-1) and was linearly correlated with the mean irradiance in the mixed layer. The inter-annual variability of phytoplankton biomass and production was high, ranging from 53 to 100 mg chlorophyll am(-2) (annual mean) and from 143 to 278 g carbon m(-2)y(-1), respectively. The degree of seasonal mixing determined annual production, demonstrating the sensitivity of tropical lakes to climate variability. A review of primary production of other African great lakes allows situating Lake Kivu productivity in the same range as that of lakes Tanganyika and Malawi, even if mean phytoplankton biomass was higher in Lake Kivu.