Denitrification in mountain lakes = Desnitrificación en lagos de montaña

Abstract

The reservoir size and pathway rates of the nitrogen (N) cycle have been deeply modified by the human enhancement of N fixation, atmospheric emissions, and climate warming, doubling the reactive nitrogen (Nr) available in the biosphere. Denitrification transforms nitrate into nitrogenous gas and thus removes Nr back to the atmospheric reservoir. Across ecosystems, there is still rather limited knowledge of the denitrification rates and their relationships with environmental factors and the N-transforming guilds, particularly, for the abundant cold and N-poor freshwater systems.

The main goal of this thesis was to improve the current knowledge of denitrification in mountain lakes. In particular, we studied eleven pristine oligotrophic mountain lakes that have been affected by a high N deposition. The selected lakes showed a gradient of dissolved inorganic nitrogen in the water due to different lake productivity. We focused on spatial rather than in temporal variation, sampling during the ice-free period. Within lakes, we focused on the sediments because of their known higher denitrification rates than the water column. Specifically, we studied sediments near the deepest point of the lake, lithic biofilms from littoral cobbles, and littoral sediments from beds of isoetid and elodeid macrophytes, helophyte (Carex rostrata) belts, and rocky areas.

We aimed to measure the denitrification activity as similar as possible to in situ conditions, with this aim we used little disturbed sediment cores and in situ temperature for the denitrification rate measurements. We characterized the environment by including proximal (benthic) and more distal (lake) descriptors to capture potential drivers acting at different spatial scales. Denitrification is part of the N cycle, and other processes facilitate (e.g. nitrification) or compete (e.g. DNRA) with this pathway. Using molecular tools, we quantified the guilds involved in the main N-transformation pathways in benthic habitats (article I) and related them to the denitrification rates (article IV). We have also developed a method appropriate for estimating denitrification rates in any aquatic system with retrievable sediment cores (article II). Finally, we showed the interest of quantifying the temperature dependence of the process at different degrees of substrate limitation but within the range — or a beat above — the in situ substrate levels, instead of the typical quantification at substrate saturation (article III).

Following, there is a summary of the main findings. There is a complex N-transforming guild composition in benthic habitats of mountain lakes, which is deeply embedded in the overall prokaryotic community. These N-transforming guilds differ in the dominant pathway depending on the habitat and productivity of the lake, with the DNRA-denitrification dichotomy as the greatest differentia- tion (article I). The denitrification temperature dependence increases with nitrate limitation (article III). There is an average current denitrification rate of 1.5 μmol N O m-2 h-1 in the sediments of the Pyrenean lakes with higher activity in littoral than in the deep zones. The factors controlling current and potential denitrification rates differ. Current denitrification is controlled by the NO -/ NO - availability and secondarily by temperature; whereas potential denitrification is controlled by landscape productivity, the sulphate content, and the DNRA-denitrification (nrfA-nirS) competition (article IV).

The best candidate drivers, i.e. temperature, organic matter quantity and quality, and spatio- temporal redox conditions affecting the overall prokaryotic community and the N-transforming guilds are discussed. The estimated denitrification rates are compared to other mountain lake sediments, discussing the spatial variations, the controls, and the methods used. Finally, some unsolved questions of the N cycle in mountain lakes are discussed.

Overall this thesis contributes to increasing the knowledge of denitrification and other processes of the N cycle. The findings are probably not restricted only to mountain lakes encompassing other oligotrophic and remote ecosystems.

El ciclo del nitrógeno (N) ha sido profundamente modificado por la fijación industrial de N, el aumento de emisiones, y el calentamiento global, doblando el nitrógeno reactivo (Nr) disponible en la biosfera. La desnitrificación transforma el nitrato en gases nitrogenados eliminando Nr del sistema. Hay un conocimiento muy limitado de las tasas de desnitrificación, así como de los gremios microbianos implicados en la transformación de N, especialmente en sistemas oligotróficos de aguadulce.

El principal objetivo de esta tesis es ampliar el conocimiento actual de la desnitrificación en lagos de montaña. Para ello, se estudiaron once lagos oligotróficos afectados por una alta deposición de N, que muestran un gradiente de nitrógeno inorgánico disuelto debido a una diferente productividad dentro de la oligotrofia. Concretamente se estudiaron los sedimentos del punto más profundo del lago, las biopelículas de las piedras litorales, y los sedimentos litorales de las áreas rocosas, de los cinturones de Carex rostrata, y de los lechos de macrófitas isoétidas y elodéidas.

Se ha encontrado una compleja composición de gremios transformadores de N profundamente arraigada en la comunidad procariota general de los hábitats bentónicos. La ruta dominante de transformación de N cambia dependiendo del hábitat y la productividad del lago con la dicotomía DNRA-desnitrificación como mayor diferencia, con una dominancia de los desnitrificantes reductores de nitritos (nirS) en las capas superficiales de los sedimentos de los lagos someros, más cálidos y productivos. También una creciente dependencia de la temperatura en la desnitrificación al aumen- tar la limitación de nitratos. Se ha estimado una tasa promedio de desnitrificación de 1.5 μmol N2O m-2 h-1 en los sedimentos de los lagos pirenaicos, con mayor actividad en la zona litoral que en la profunda. Diferentes variables controlan las tasas de desnitrificación actuales y potenciales; las pri- meras están controladas por la disponibilidad de nitratos y secundariamente por la temperatura, las potenciales están controladas por la productividad del sistema, el contenido de sulfatos y la compe- tencia DNRA-desnitrificación (nrfA-nirS).

Esta tesis contribuye a aumentar el conocimiento del ciclo del N, y probablemente, los resultados obtenidos son extrapolables a otros ecosistemas oligotróficos y de áreas remotas.