Cookies

We use cookies to ensure that we give you the best experience on our website. You can change your cookie settings at any time. Otherwise, we'll assume you're OK to continue.

Durham University

Department of Earth Sciences

Postgraduate Students

Publication details for Professor Fred Worrall

Worrall, Fred, Wade, Andrew J., Davies, Richard J. & Hart, Alwyn (2019). Setting the baseline for shale gas – establishing effective sentinels for water quality impacts of unconventional hydrocarbon development. Journal of Hydrology 571: 516-527.

Author(s) from Durham

Abstract

There is a need for the development of effective baselines against which the water quality impacts of industry in general, and shale gas extraction specifically, can be assessed. The salinity, and hence the specific conductance, of fluids associated with shale gas extraction is typically many times higher that of river water. The contrast between these two water types means that testing for salinity (specific conductance) could provide an ideal sentinel for detecting environmental impact of shale gas extraction. Here, Bayesian generalised linear modelling was used to predict specific conductance across English surface waters. The modelling used existing, spot-sampled data from 2005 to 2015 from 123 sites to assess whether this approach could predict variation for subsequent years or for a new site (data from 2002 to 2015). We show that the results were readily projected in to subsequent years for sites included in the initial analysis. The use of covariates (land-use, hydroclimatic and soil descriptors) did not prove useful in predicting specific conductance at further sites not previously included in the analysis. The extension of the approach to 6833 English river monitoring sites with 10 or more observations from more than one year over the period 2005 to 2015 showed that it was possible to reproduce the seasonal variation in river water specific conductance. The approach taken here shows that it is possible to use low-frequency but widespread monitoring data to predict natural variation at monitoring sites to give a probabilistic assessment of whether or not a pollution incident has occurred and the seasonal variation, expressed as uncertainty bounds around the observations, at a specific site has been exceeded.