Jacobs Journal of Civil Engineering

Factor analysis to assess pollutant source apportionment and to investigate the relationship between catchment attributes and instream water quality

*Azzellino A
Department Of Civil Engineering, Italy

*Corresponding Author:
Azzellino A
Department Of Civil Engineering, Italy
Email:arianna.azzellino@polimi.it

Published on: 2017-08-26

Abstract

The EU Water Framework Directive (WFD EC60/2000) requires that quality-flow compliance at a particular surface-water reach entail consideration of all upstream inputs, including contaminated land and groundwater contributions. Multivariate statistical techniques may improve our understanding of the pollutant sources affecting river quality. Aim of this study is to analyze the source apportionment and the groundwater contribution to the total pollutant load of Mella river. Factor Analysis (FA) was applied to a series of water quality measurements at seven monitoring sites, located upstream, in the middle and downstream the groundwater recharge area of the Mella river watershed. FA results in the upstream sites were completely different from the lowland stations that were strongly influenced by the groundwater contribution. In the upstream sites, in fact, the major pollutant source resulted to be the contribution of the Gobbia tributary which collects the industrial loads of the Val Trompia metallurgic consortium. On the other hand the groundwater was found to be the most significant pollutant source in the lowland sites. FA proved also useful to distinguish between sources of metals and chlorinated solvents.

Keywords

Factor Analysis; Groundwater Interactions With Surface Waters; Source Apportionment; Macro and Micropollutants

Introduction

The European Water Framework Directive (WFD: 2000/60/ CE) defines a new logic in surface- and ground- water quality management for the European Union, promoting a river basin-approach rather than a local scale approach, stimulating a more integrated approach to mitigate and manage pollution at watershed scale. This is the reason why the analysis of water quality requires today more complex investigations and the identification of all the emission sources affecting water quality at catchment levels. The monitoring and quantification of point and nonpoint sources contributions to the global pollutant load is therefore a key issue for the implementation of management strategies. Diffuse loads may be significant either in wet-weather conditions (i.e. pollutants carried by surface runoff) or in dry-weather conditions (i.e. pollutants carried by subsurface runoff or due to the groundwater exchanges with surface waters). In this respect, the understanding of the interactions between surface waters and ground waters may be the basis for effective water resource management [1]. To fully understand the interactions between surface waters and ground waters, a sound and robust monitoring of surface- and ground water quality data is required. Instream measurements, that are very often instantaneous, can provide information about the total loads in a specific watershed, but do not provide insights about the source apportionment of pollutants, if they are not integrated with other investigative tools, such as mathematical models or statistical techniques. Although experiences are reported concerning the source apportionment of micropollutants (e.g. [2-11], most of the available literature concerns the monitoring either at the emission source [8] or in water bodies [12] of these substances. On the other side, many conceptual models have been developed for the surface-groundwater system [13-14]. Nevertheless, the effect of the groundwater interactions on surface water quality is hard to quantify. Multivariate statistical techniques (e.g. Factor Analysis) may support the understanding of these interactions at watershed scale [18, 19]. Aim of this study has been to apply Factor Analysis (hereinafter FA) to series of water quality measurements collected at seven monitoring sites, located within the Mella river watershed (Figure 1), in order to assess the source apportionment of different macro- and micropollutants.