Eutrophication is the process of excessive plant growth in water bodies, typically due to human induced nutrient loading and can lead to harmful algal blooms (HABs) (Schindler, 2006). HABs are a national and global problem mainly influenced by anthropogenic inputs of nitrogen (N) and phosphorous (P), specifically non-point sources such as agricultural fields. N and P are key nutritional elements for plants and therefore, an accumulation in waterways promotes growth which can lead to eutrophication and HABs. (Ludsin et al. 2001; Michalak et al. 2012). HABs create extensive hypoxic (low-oxygen) areas resulting in significant consequences to water quality with emphasis on drinking water and fish life (Chislock et al. 2013). Zooplankton are highly effective in controlling algal growth, however, they are also very sensitive to warming water. With the onset of climate change the effect of zooplankton will likely be limited, therefore increasing eutrophication (Moss et al. 2011).
Lake Erie is the most susceptible Great Lake due to its shallow depth allowing for greater biodiversity (Government of Canada, 2018). Lake Erie has been plagued with HABs for decades, with its most devastating bloom in 2014 where toxic bacteria contaminated the drinking water of Toledo, Ohio (Bentanzo et al. 2015). An additional proxy for increased runoff is areas of extensive agriculture and large urban populations, the Lake Erie watershed has over 11 million people residing in it (Government of Canada 2018; Erie County, 2019). For these reasons, it is critical that the factors affecting N and P levels be spatially analyzed and predicted in hopes to implement best management practices and prevent excessive inputs from entering waterways.
There are many different sources (emitters) and sinks (absorbers) of N and P that determine the levels of N and P at a given location within a stream. One main factor affecting N and P levels is agriculture (Chen & Lu, 2014; Khare et al. 2015; Mainali & Chang, 2018; Pratt & Chang, 2012). Agricultural inputs can cause watershed and local scale influences on N and P (Chen & Lu, 2014). Agriculture croplands have fertilizer used on them which are enriched in N and P. This N and P source can then be transferred to waterways through run off from these fields ultimately contributing to HABs (Chen & Lu, 2014; Khare et al. 2015; Mainali & Chang, 2018; Pratt & Chang, 2012). Urban areas are another contributor to N and P levels because of the low permeability surfaces in urban areas lead to runoff into nearby water bodies (Chen & Lu, 2014; Mainali & Chang, 2018; Pratt & Chang, 2012). These two types of land cover are the primary sources of N and P in waterways, however there are other land covers that act as sinks for N and P (Basnyat et al. 2000; Chen & Lu, 2014; Pratt & Chang, 2012). Forests have been shown to decrease the N and P levels of the nearby water bodies as they can absorb the N and P before these nutrients reach the water (Basnyat et al. 2000; Chen & Lu, 2014; Pratt & Chang, 2012).
Factors other than land cover can also influence the N and P levels in a waterbody. One of these factors is non-point source pollution (Fanelli, Blomquist, & Hirsch, 2019). This is pollution coming from a known single source, such as an effluent pipe (Fanelli et al. 2019). Although these sources are large contributors to N and P levels, in the Lake Erie watershed point source pollution has been drastically decreased in recent years (Fanelli et al. 2019). Causes of point source pollution have been identified and removed through improvement of infrastructure (Fanelli et al. 2019). Although point source pollution has largely been reduced, the non-point sources of pollution from agriculture and otherwise still contribute significant amounts of N and P to the waterways, which continue to cause HABs. Precipitation and temperature do not directly increase or decrease N and P levels. However, the effects of agricultural cropland are increased by rainfall as this can cause large amounts of run off to flow into the rivers, increasing the N and P levels (Pratt & Chang, 2012). Soil type can also have an influence on N and P levels as different types of soil have different erodibility (Colborne et al. 2019). Soils that have higher erodibility can cause there to be increased soil inputs into water bodies, and thus increased P and N inputs (Colborne et al. 2019). The final factor that can affect N and P levels in water bodies is slope. Areas with high slope have higher rates of erosion leading to increased sediment transport and run off, which increases N and P levels (Chen & Lu, 2014; Colborne et al. 2019).
Current Knowledge & GIS Importance
There has been a multitude of research done regarding Lake Erie and its watershed including impacts, factors, and consequences of eutrophication (Schindler et al. 2016; Scavia et al. 2014; Bentanzo et al. 2015). There is also a collective understanding that the problem is exacerbated by various factors interacting within the watershed at the field scale. These include differences in time and technique of fertilizer applications, excessive soil erosion, increased connectivity from the drainage network to the lake, excessive P soil tests, and stratification of P at the surface (Reutter et al. 2011). Studies involving spatiotemporal variability and environmental factors have been researched (Tian et al. 2017; Bridgeman et al. 2006), however, there remain gaps. Specifically, there is a lack of research contributing to the understanding of collective factors interacting to influence N and P levels and their relative importance in determining N and P levels (Reutter et al. 2011). In order to bridge this gap, studies must go beyond the plot scale and include year-round collection from different locations with various runoff conditions as well as record keeping of agricultural practices within the drainage area.
Given the vast, interconnected, and constantly changing traits of watersheds, it is therefore critical to spatially analyze their current and future states. The field of geomatics provides a multitude of tools and models that can be used to better understand, analyze, and predict variables of differing spatial and temporal scales. In the context of the eutrophication of Lake Erie, predictive GIS-based models would assist in understanding areas of the watershed that have minimal to no monitoring and/or sampling. This can help in the identification of future areas of research and the prioritization of policy, planning, and best management practices.