By Dr. Mark Luttenton
Higgins Lake has changed substantially over the past 30 years. The addition of nutrients at an accelerated rate and the introduction of numerous aquatic invasive species (AIS) have pushed Higgins Lake to a condition that is unnatural for this type of Michigan inland lake. The fact that Higgins Lake has reached this condition is no surprise; several previous studies outlined steps that needed to be taken to prevent this outcome. Schultz and Fairchild (1984) clearly stated that increasing population density, aging septic systems, and decreasing nutrient binding capacity by the regional soils would result in impacts to Higgins Lake due to domestic sewage. Subsequent studies have confirmed that nutrient levels in nearshore groundwater has reached unnaturally high levels (Minnerick, 2001; Martin et al. 2014).
There does seem to be general agreement that the addition of nutrients, particularly phosphorus, has resulted in changes in Higgins Lake and is a primary concern. However, there is an ongoing debate regarding the need for a wastewater facility that has been proposed to serve the Higgins Lake community. Planning for the sewer system has been based in part on several studies that have developed nutrient loading estimates for Higgins Lake. However, models estimating the same parameter (e.g., nutrient loading) for the same system often return different results. This fact is not surprising given that nutrient loading estimates are the result of various parameters used in the model and various assumptions made by the author(s).
When developing a model, the values used for model parametersand assumptions made are based on the information at the time the study is conducted. Subsequently, new information becomes available which allows models to be refined and updated, often providing estimates (in this case nutrient loading) that are typically more representative of actual conditions. Because several modeling studies (and more recent empirical data) developed for Higgins Lake over the past 30+ years have estimated different loading rates, a critical evaluation of the available nutrient loading estimates in light of new information may offer some clarity and new insights.
One factor to clarify is the long-held thought that phosphorus (P) is the primary limiting nutrient in freshwater lakes. The idea is partially based on historic conditions, traditional lake monitoring programs, and common lake management practices. Today we know that the concept may no longer be applied in absolute definitive terms, and this is clearly the case for Higgins Lake. Schultz and Fairchild (1984), Akey (2019), and Luttenton (unpublished 2020, 2021) have all shown that sufficient phosphorus is being added to Higgins Lake that nearshore areas are not simply P limited, but are co-limited by phosphorus and nitrogen depending on the time of year, particularly later in the summer. Shultz and Fairchild (1984) noted that creating nitrogen limitation in the nearshore area will promote growth of nitrogen fixing organisms such as cyanobacteria. Indeed, that is exactly what we see in the nearshore region. In addition, because so much of the phosphorus has accumulated on the bottom of the lake, we are now seeing nitrogen-fixing cyanobacteria become common in the off-shore areas during the summer.
Even though Higgins Lake exhibits both P and nitrogen limitation, controlling P loading to Higgins Lake is absolutely essential and will require implementing several best management practices. Schultz and Fairchild (1984) provided several recommended actions to reduce nutrient loads. Many of those still apply.
To develop their recommendations, Schultz and Fairchild (1984) evaluated the sources of nutrients entering the lake. For example, they evaluated the input from numerous road ends and confirmed that while some were high, many contributed relatively little to Higgins Lake. Many road ends have been modified since 1984, but reevaluation of some road ends is likely warranted. Schultz and Fairchild (1984) also note that the contribution of surface runoff varies with precipitation. Fortunately, because the soils in the Higgins Lake watershed allow precipitation to infiltrate very rapidly, nutrient loading by surface runoff from forested areas should not be considered an important source. Lastly, they recommended banning lawn fertilizers within 100 yards of the shoreline to protect shallow nearshore areas. Currently there is an ongoing campaign to reduce the use of lawn fertilizers which are considered to be a risk.
Shoreline erosion does present an additional source of nutrient loading to Higgins Lake. High water levels that result in shoreline erosion reduce the amount of shoreline vegetation that may have the capacity to take up nutrients before they enter the lake. Perhaps more important is that the condition leading to shoreline erosion (high water levels) will potentially raise groundwater levels. Higher groundwater levels will reduce the distance between septic tanks and groundwater thereby decreasing the capacity for soil-septic effluent interaction. Depending on the rise in the water table due to higher lake levels, nearshore septic systems may be inundated.
If there is a true desire to improve and protect Higgins Lake, both historic and current data clearly indicate that reducing nutrient loading needs to be a priority. Fortunately, because of the range of previous studies and ongoing annual lake assessments, we have a good grasp of the issues that Higgins Lake faces and the actions required to address these issues. Although there are ongoing efforts to reduce nutrient loading due to several sources, the need to reduce nutrient loading to groundwater from septic effluent was recently punctuated by information on nitrate levels presented by the Central Michigan Health Department. Those data show concentrations in some residential wells now exceed the EPA drinking water standards for nitrate and a significant number of wells have levels that are of concern. High nitrate concentrations in drinking water can cause severe complications during fetal development and in young children, and recent information suggests a link to various abdominal diseases. As large as the concern is about the health of Higgins Lake, human health has now emerged as an additional and, in some cases, an even greater concern due to the high concentrations of nitrates.
Our understanding of Higgins Lake has benefited from several studies that model the amount of P entering the lake from septic tank effluent. The phosphorus loading estimates from septic effluent developed in 1975, 1980, 1984, and 1992 ranged from 570 pounds per year to 1,435 pounds per year. Each of the modeling studies used values for model parameters that were available at the time, and included assumptions that were generally considered acceptable.
In 2007, Huron Pines developed a phosphorus loading estimate that was significantly higher than previous estimates, although the estimate was revised downward due to a reported conversion error (I have not seen the report so I can’t comment on specific errors). Compared to the average of the earlier studies, the revised Huron Pines estimate (4,965 pounds per year) was noted to be as much as five times higher than previous estimates. The difference between the Huron Pines loading estimate and the previous loading estimates is simply the amount of area and the total number of septic systems included in the model. The Huron Pines estimate incorporates the area 1000 feet from the lake, whereas the previous studies only included homes within 300 feet from the lake.
Widening the zone that contributes to nutrient loading accounts for two major factors that have emerged since the early 1990’s: 1. The demographics of the Higgins Lake area has changed significantly since the early 1990s, and 2. the more recent study conducted by Martin et al. (2014). Martin et al. (2014) used empirical data to determine which parameters have the greatest influence on nutrient loading associated with groundwater. They included: 1. Groundwater flow velocity (direct measurement), 2. Hydraulic gradient (a groundwater elevation map at 400m (1312.3 ft) and 1000m (3280.4 ft) from the shoreline), 3. Septic count (estimate of septic systems within a 100m buffer zone (328 feet) along the upgradient groundwater flowpath at 400m and 1000m upslope; (flowpath is the direct line of groundwater flow from upslope to downslope to a groundwater sample site in Higgins Lake); 4. Septic flux (hydraulic gradient multiplied by septic count to estimate septic system contributions from the specific flowpath length (400m or 1000m)); and 5. Parcel count and parcel flux were calculated in the same way as septic count and septic flux.
Primary findings of Martin et al. (2014) are:
FOR GROUNDWATER –
1. The Higgins Lake area has high to very high groundwater flow rates based on USGS criteria.
2. TP concentrations were significantly related to hydraulic gradient at 400m and 1000m.
3. TP concentrations were significantly related to septic flux at 400m and 1000m.
4. TP concentrations were significantly related to parcel flux at 400m and 1000m.
5. Ammonia concentrations were significantly related to groundwater flow velocity.
6. Ammonia concentrations were significantly related to septic count at 400m and 1000m.
7. Ammonia concentrations were significantly related to parcel count at 400m and 1000m.
FOR NEARSHORE HIGGINS LAKE SURFACE WATER-
1. TP concentrations were significantly related to groundwater flow velocity.
2. TP concentrations were significantly related to hydraulic gradient at 400m and 1000m.
3. TP concentrations were significantly related to septic count at 400m and 1000m.
4. TP concentrations were significantly related to septic flux at 400m and 1000m.
5. TP concentrations were significantly related to parcel flux at 400m.
6. Ammonia concentrations were significantly related to all parameters (groundwater flow velocity, hydraulic gradient, septic count, and septic flux) at 400m and 1000m.
7. Ammonia concentrations were significantly related to parcel flux at 400m.
Based on the study by Martin et al. (2014) expanding the zone of influence from 300 feet to 1000 feet as noted in the Huron Pines study is valid and reveals a much greater issue than previously recognized. Indeed, the previous estimates of nutrient loading into Higgins Lake are very likely significant underestimates of the true nutrient loading rates due to septic effluent. Assuming that nutrients in septic effluent generated more than 300 feet from the lake will never reach the lake, or are significantly reduced due to soil adsorption capacity could underestimate loading rates by as much as two thirds. Based on Huron Pines (2007) and Martin et al. (2014), the need to control septic effluent from beyond 300 ft is critical.
Beyond the empirical data reported by Martin et al. (2014), there is additional information that supports expanding the zone that contributes to nutrient loading, specifically incorporating new soil information. Obviously, soil type plays a critical role because soils have the capacity to adsorb nutrients such as P, albeit at varying rates. Less P will be added to the lake the better a soil type can bind P. The nutrient binding capacity of soil is based on a complex of chemical interactions, water infiltration rates, groundwater flow rates (e.g., Martin et al. 2014), distance between the nutrient source and the groundwater, and how long the soil has been subject to the addition of nutrients. Ultimately, even soils with relatively high nutrient binding rates will realize a significant drop in binding capacity.
Because the early modeling studies selected a 300-foot boundary, they limited the primary soil types used in the models to those immediately adjacent to the shoreline which were listed to be primarily in the Grayling and Rubicon Sand series, among a few others. Using these soil types in the model was apparently based on the information available at the time, primarily low-resolution soil survey maps from 1924 and 1927. The significance of using Rubicon sand in the model is that it has a high P binding capacity and Grayling sand is listed as having a medium to low binding capacity. Reviewing current USDA-NRCS (Natural Resource Conservation Service) soil maps for the Higgins Lake area indicates that Grayling and Rubicon sands are not the only soil types within the 300-foot boundary around the lake. For example, along the Northeast shore of the North basin, Graycalm sand accounts for approximately 43% of the soil immediately adjacent to the shoreline. Grayling sand totals less than 1% and Rubicon sand is completely absent. Graycalm sand has very low P binding capacity compared to Rubicon sand. Assuming that Rubicon and Grayling sands are the dominant soils within 300 feet of the entire shore would result in an underestimate of nutrient loading.
Perhaps even more important is that Rubicon and Grayling sands account for less than 6% of the soil within the Higgins Lake watershed. In comparison, Graycalm sand totals over 50% of the soil in the watershed. Because Graycalm sand has a much lower P binding capacity, much less of the P associated with septic effluent will be retained in the soil and much more will be carried to the groundwater.
Even in areas with soils that have high P binding capacity, the amount of P that will be retained is not infinite. As P binds to particles in the soil, the total binding capacity is diminished over time. Consequently, less P is retained in the soil and more P is transported to the groundwater and into the lake. In areas with several tiers of septic systems, the issues may be more severe. The soils between the lake and nearest septic system may have reduced binding capacity due to the system nearest the lake. This allows effluent from the septic systems farther from the lake to pass through with far less phosphorus (and other nutrients) being removed from the effluent.
High nitrate levels in private wells, high P in nearshore groundwater and surface water, and biological changes occurring in Higgins Lake all indicate that septic effluent is having a major impact within the Higgins Lake watershed. Conditions will continue to become more severe if septic effluent continues to be released into the surrounding soils. Even strict ordinances will likely do little to improve conditions. For example, Limno Tech (1992) found that 75% of nearshore residents had their septic tank pumped at least every 5 years. But nutrient levels in both the groundwater and nearshore surface water are much higher than normal. It appears that regular septic tank maintenance is not sufficient to address the elevated nutrient levels in groundwater.
Taken together, the data presented by Martin et al. (2014) and Minnerick (2001), a reevaluation of soils within the watershed, and the changes in regional demographics strongly suggest that nutrient loading to groundwater from septic effluent is a long-term and persistent issue. In addition, nutrient loading due to septic effluent has very likely been underestimated by earlier modeling studies. These conditions have resulted in an unnatural increase in nutrient loading to Higgins Lake which has resulted in substantial ecological changes to the lake.