Search

The deterioration of Lake Erie water quality over the last two decades, despite improvements that occurred in the 1980s and early 1990s, has caused an increasing trend in the presence of harmful algal blooms. In this study, changes in the relationship between weather and water and nutrient discharges over time are investigated for the Maumee and Sandusky watersheds, using extensive historical data in statistical models that relate water discharge (flow) and nutrient concentrations to lagged-precipitation and temperature, seasonality, and annual fixed-effects. In Chapter 1, differences in weather effects on flow between three periods that correlate with shifts in water quality trends,1960-1979, 1980-1995, and 1996-2015, are analyzed. Findings indicate a faster flow response to precipitation during 1996-2015 as compared to the previous two periods, and that part of the increase in flow over time is related to factors other than precipitation. The model is also used to assess period-averaged contributions to flow associated with precipitation lag times. It was found that, while short lag time contributions decreased during 1980-1995, they increased significantly during the 1996-2015 period, following same trend as Lake Erie water quality. These results suggest a need to implement a nutrient management strategy that goes beyond the currently adopted best management practices (BMPs). In Chapter 2, differences in the weather effects on nutrient concentration and load between two periods that correlate with shifts in water quality trends, 1980-1995, and 1996-2015, are analyzed. The investigated nutrients are attached phosphorus, soluble reactive phosphorus (SRP), total Kjeldahl nitrogen, and nitrite/nitrate-nitrogen. The model also includes economic variables which are used to analyze the effectiveness of economic policies for nutrient management. Major findings include (1) the large and statistically significant increase in SRP concentration and load response to precipitation during the 1996-2015 period as compared to the 1980-1995 period, and (2) the moderate effectiveness of a fertilizer tax policy for nutrient management. The results indicate that both nutrient input reduction and controlled drainage may be needed to reduce and control short and long-term nutrient loading. In Chapter 3, the flow and nutrient concentration models are used to investigate the effects of anthropogenic factors on flow and nutrient concentration over time, such as subsurface drainage, independently of weather. Similar rain patterns are identified and matched sets of similar sized events for the two watersheds are developed. Major findings are (1) the flow response to precipitation, when controlling for changes in weather over time, is faster for events in the interval 0.75 – 1 in. at both watersheds, and (2) the SRP concentration response to precipitation at the Maumee watershed is faster for events in the interval 0.75 – 1 in. while at the Sandusky watershed it is faster in the interval 0.5 – 0.75. These findings indicate that factors such as subsurface drainage capacity and conservation tillage may have increased over time and suggest the need to focus on management practices that store more water in the landscape, and to continue the implementation of other BMPs such as subsurface placement of phosphorus fertilizer.