Lauren Adams

Ubiquitous non-point source (NPS) pollution is a dominant cause of biogeophysical degradation in urban catchment systems, the residual effects of which damage community health, safety and property values.  Remediation of water resources contaminated by NPS requires both political participation as well as scientific information, particularly for drinking water supply sources, where the human impacts of NPS pollution are more acutely realized.  To better understand the relationship between the demand for clean water and the supply of scientific education and information, my research compared actual and perceived pollution risks within the urban Mill River watershed in New Haven County, Connecticut to determine the magnitude and characteristics of the watershed’s manufactured risk.

Measuring the concentrations of different parameters within the watershed offers insight and analysis into a small portion of the actual risks associated with the watershed.  Drilling into a larger dataset collected by the Mill River Regional Water Authority, who funnels portions of the Mill River through a water purification plant and distributes it to a portion of residents living in close proximity to Lake Whitney, helped to determine that minimal risk is associated with drinking the water from this system, particularly after its been purified.  Although every chemical parameter has not been tested within the basin, it is extremely likely that the drinking water sourced by the Mill River is safe, an assumption which is further carried out by the lack of people sick from the local drinking water supply. 

With respect to the risks associated with the Mill River’s local ecosystem, I found high levels of several different chemical parameters indicative of an impaired watershed, likely from urban development that occurred without consideration of its impact on the local watershed.  For example, there were very high levels of coliform, which is indicative of pathogenic bacteria, possibly from leaking septic tanks and high levels of total phosphorous and total nitrogen, meso- to eutrophic-like water conditions could be fertilizer run-off from nearby lawns.  Concentrations of each of these parameters increased as the water flowed downstream through increasing impervious surface cover, further inferring a relationship between impervious surface cover and poor water quality.  Therefore, in this particular watershed, I found that the actual risk of drinking the water the Mill River watershed sources to local treatment facilities is low even though the ecosystem itself is impaired and presents a risk to human health and the environment. 

Separate from the actual risks associated with the Mill River watershed, this research demonstrated that Mill River watershed users have a difficult time describing their local water supplies both at the source and from the tap.  It seems that there is a general lack of interest in and understanding of the mechanistic links between watershed, human and ecosystem health prevails, despite people’s intense preference for the trusted delivery of clean water supplies within their urban homes.

To further quantify the spatial relationship between user perception and actual risks, I will be using remotely sensed imagery to classify each land-use type within the Mill River watershed and then I will match these results with water quality data over a three month time period.  With the assistance of GoogleEarth several different classification methods will be tried, including unsupervised and supervised, to classify the images.