Water quality variations in an urbanising catchment, the Monks' Brook, Hampshire

Prowse, Colin Warwick (1982) Water quality variations in an urbanising catchment, the Monks' Brook, Hampshire. University of Southampton, Doctoral Thesis.

Record type: Thesis (Doctoral)

Abstract

This thesis attempts to identify spatial and temporal water quality variations in a medium-sized (33.03 km2 ) urbanising catchment, with an emphasis on differences in land-use and in relation to storm sewage from existing and developing urban areas. The main aim is the assessment of the impact of urbanisation upon water quality, specifically chemical water quality. Secondary objectives include the explanation of the processes which lead to hysteresis and the identification of the importance of spatial variation upon chemograph response. A total of twenty-six primary water quality parameters (including all major cations and anions) were determined on a regular basis at twenty-five stream sampling sites, and for a series of twenty-one storms at two major sample points (one rural, one urban) from November 1978 to December 1980. The resultant data, which comprised 1577 observations and 266 variables, were analysed using the Mann-Whitney U statistic, trend line regression analysis, direct comparisons of chemograph response, linear stepwise regression and factor analysis. The results suggest that urbanisation is unimportant with respect to changes in average baseflow water quality, but that individual polluting events associated with oil spills and toxic metal discharges have had a serious effect upon the surface waters in the catchment. In addition, urbanisation is seen to effect significant changes in chemograph response including intensifying the dilution of many determinands (e.g. Nat , Mg2+1 and resulting in the flushing of some others (e.g. Ca2+, Zn2). Hysteresis induced by lags (exhibited by a number of determinands) was removed from linear regression models by the incorporation of lagged variables. This provided significant improvement in model fit, especially for Fe 2+ for which an increase in variance explained of 19% was recorded. Source areas were identified using factor analysis. The recognition of spatially distinct sources within the catchment (e.g. a Chalk outcrop), demonstrates the need for a spatio-temporal approach to the modelling of complex basins as opposed to the often used lumped approach.

This record has no associated files available for download.