For our final Water Group seminar of this term we have presentations on two interesting research questions!

First up, What caused variability in DOC concentrations in the River Thames between 1884-2014? by Valentina Noacco and second, Fanny Sarrazin will give a talk entitled: How much does subsurface heterogeneity alter the impact of climate and land use changes on groundwater recharge?  To find out the answers to these mysteries, come to Queen's Building 1.68 this Friday 18th March at 1pm.

This event is open to University of Bristol students and staff only.

Abstracts

How much does subsurface heterogeneity alter the impact of climate and land use changes on groundwater recharge?

Fanny Sarrazin, Andreas Hartmann, Francesca Pianosi and Thorsten Wagener

Karst aquifers are an important source of drinking water in many regions of the world, but their resources are likely to be affected by changes in climate and land use. In fact, climate characteristics control the supply of water to karst systems and the evaporative demand, while land use characteristics control the actual evapotranspiration losses. Understanding karst hydrology and estimating karst groundwater resources at a large-scale is critical for preventing threats to water supply in a changing world. Hartmann et al. (2015, Geosci. Model Dev.) introduced a parsimonious karst recharge model, called VarKarst-R, which allows for large-scale simulations of groundwater recharge while explicitly taking into account karst heterogeneities, i.e. preferential flow paths. The first objective of the present study is to introduce vegetation processes into the VarKarst-R model to better estimate evapotranspiration losses depending on the land use characteristics. We test the model at Fluxnet sites located in carbonate rock areas. Secondly, the VarKarst-R model so modified is used to assess the relative influence of changes in climate and land use on aquifer recharge. We establish a sensitivity analysis framework to analyse the interactions between climate descriptors (e.g. mean precipitation, precipitation seasonality), vegetation parameters (e.g. canopy storage capacity, rooting depth) and soil parameters (e.g. soil storage).

What caused variability in DOC concentrations in the River Thames between 1884-2014?

Valentina Noacco

Climate and atmospheric circulation patterns influence the variability of basin hydrochemistry, therefore understanding their influence is essential to put short-term water quality trends into the right context and to predict future hydrochemistry responses in the face of climate change.
 We investigate the drivers of DOC concentration variability in the Thames basin over 130 years. Our previous work has shown that increased urbanization since the 1880s in the Thames basin was the major driver for the increase in riverine DOC, but it does not explain DOC variability. Our current work investigates the links between hydro-climatic variability (temperature, precipitation and runoff) and teleconnections, and the variability in DOC concentration. Moreover we compare the impact of hydro-climatic variability on riverine DOC, to the impact of land-use change and population increase.

We use singular spectrum analysis to identify and then compare the dominant oscillatory components of hydro-climatic and hydro-biogeochemical variables. We use phase-plane trajectories of the noise-free, intra-annual to inter-annual reconstructed components to elucidate the biogeochemical and hydro-climatic dynamics of the system. This allows us to elucidate the links between the variability of hydro-climatic variables and DOC. Moreover they enable the identification of points in time where the dynamics of the system have changed, e.g. due to anthropogenic influences. We also compare the links between hydro-climatic variables and DOC, together with nitrate to investigate common drivers and elucidate the dynamics of the system. We apply Generalized Additive Models (GAM) to understand how the various hydro-climatic variables and land-use changes influence the variability of DOC concentration.

We show that the high seasonal to inter-annual variability in DOC concentration is linked to the variability of precipitation and runoff, rather than temperature. The dominant inter-annual modes of variability in DOC are connected to the ENSO oscillatory components. During strong El Niño and La Niña years there is statistically significant positive correlation between DOC concentration and ENSO. Moreover peaks in DOC concentration (as in the 1940s and early 1980s) are connected to very strong El Niño events. ENSO impacts riverine DOC indirectly, by influencing climate and hydrology, in turn these influence DOC export, transport and production. If El Niño events will strengthen in the future, we might expect higher DOC concentration peaks. Nonetheless not all the variability of DOC concentration can be explained with the variables considered, especially in the later period when the variability of the systems increases.