The North Idaho panhandle is home to Fernan Lake Village, a small lakeside community directly outside of the City of Coeur d'Alene. Fernan Lake serves the village and the surrounding area as a hub for fishing as well as swimming, boating, and water skiing. Although Fernan Lake is relatively small, it is the most fished lake per unit area in Idaho and is an important part of the local ecosystem and culture.
The Idaho Department of Environmental Quality has been recording reports of nuisance blooms of algae on Fernan Lake since 2007. The algae blooms have become so detrimental to the lake and its users, the state of Idaho has listed it as an impaired body of water. Since 2011 postings have been put in place warning the public to avoid human contact with the lake water for several months out of the year.
The harmful algal blooms (HABs) that are present in Fernan Lake produce toxins that target the liver, neuromuscular system, and irritate the skin of both humans and animals. The problem is compounded by the inability of conventional water treatment methods to remove the toxins from the lake. Several factors have contributed to the complex problem facing Fernan Lake. Key among them is the impact of cultural eutrophication, which is the abundance of excess nutrients in waters that lead to excessive algal growth thus decreasing water quality as a result of human activities. Road construction, stream restoration, and agricultural activities have led to large amounts of phosphorus being added to the lake. This lowers the ratio of oxygen and nitrogen to phosphorus and provides cyanobacteria with an environment that they are uniquely adapted to thrive in.
Thick cell walls and the ability of the cyanobacteria to 'fix' atmospheric nitrogen allows the algae to take advantage of excess phosphorus that was introduced in part due to cultural eutrophication. This leads to the algae population growing into thick mats on the surface of the water where they are also well adapted to thrive in higher temperatures. These large opaque mats that cover the surface of the lake, in turn, limit available sunlight for other algae species. With their competitors dwindling, this gives yet another competitive advantage to the cyanobacteria and further exacerbates the problem of their extreme population growth.
When the cyanobacteria blooms do decline, the decomposing cells use up oxygen in the water column, contributing to a lack of oxygen in the water which is a condition known as anoxia. This low oxygen state allows phosphorus already stored in the lake sediments to be released back into the water. This process is known as internal loading and contributes additional available phosphorus, leading to even more algae blooms.
In an attempt to curb the HABs, the Idaho Department of Environmental Quality has put in place total maximum daily load limits on phosphorus during the months of August and September, which is intended to keep the phosphorus concentrations below a critical level.
Remediation of the toxic algae blooms requires an understanding of the nutrient dynamics of the lake system. To understand these dynamics the lake was observed for one calendar year between April 29, 2014 and April 29, 2015. Stream monitoring equipment was used to determine the inflow and outflows from the lake and the total phosphorous and sediment entering and leaving the lake. During this period lake water level was also recorded and periodic samples were taken to estimate the total phosphorous and total nitrogen in the water column.
An essential component of nutrient loading is understanding how much sediment is carried into the lake from runoff. To the north east of Fernan Lake is the Fernan Watershed. Precipition that falls on the watershed collects as it runs downhill. As it travels it erodes the land and carries sediment into the lake. The interactive figure below depicts stream flow, a.k.a. discharge, into the lake (orange) and out of the lake (green). As evidenced by the figure the majority of discharge occurs during the short January through May runoff period. This runoff is mostly due to snowmelt. The short window of time presents a management challenge to the reduction of TP and sediment. The locations of these data collection sites are noted in Figure 1.
Water samples collected from the inflows and outflows were used to determine the amount of sediment (total residue) and phosphorous entering and exiting the lake (See Figures 4 and 5). Measured stage is correlated with the rate of discharge and the discharge rate is correlated with total residue and total phosphorous calculations. This means that discharge and nutrient flows can now be predicted from water level at these locations.
Over the observed 1 year period The lake is estimated to have retained 81% of ~1100 kg of phosphorous and 67% over 2000 metric tonnes of sediment from the inflow of Fernan Creek, which has a typical western US snowpack dominated hydrograph. This means that the majority (93%) of the annual flow arrives in a very short period of time in spring.
Cyanobacteria have a competitive advantage when nitrogen is limited, due to excess phosphorous, because many species can obtain nitrogen from the air instead of having to obtain nitrogen from the water. This also means they tend to form scums at the lake surface. This in turn prevents other algae from competing because they are shaded. Furthermore, when the cynanobacteria die their decomposition depletes the water column of dissolved oxygen which allows phosphorous contained in sediment at the bottom of the lake to enter the water column.
Eutrophication is a term that refers tothe presence of excess nutrients in water. The water quality of many freshwater lakes and reservoirs hinges on the nitrogen to phosphorus ratio (Redfield 1958). Phosphorus limitation is the ideal case for a healthy water body. A balanced TN:TP ratio is 7:1 N:P by mass (Redfield 1958). At a ratio greater than 7, in theory, the system is phosphorus limited which limits the growth of harmful algae such as cyanobacteria. However, recent studies have shown that cyanobacteria blooms can proliferate at ratios lower than 75:1 (Harris et al. 2014a). High ratios of nitrogen to phosphorous can result in undesirable levels of nitrates+nitrites (NO3) and ammonia (NH4).
To measure the change in nitrogen and phosphorous levels several locations were sampled throughout the year. Recall, the previous data show flow from Fernan Creek decreases substantially after the spring runoff period. In conjunction with this data it suggests that during the summer months the lake exhibits problematic internal loading of phosphorous. Internal loading means that the phosphorous is entering the lake from intrinsic sources such as the sediment at the bottom of the lake.
High external loads of P can accumulate in a receiving water body and lead to legacy effects lasting well into the future, even if the external source is reduced. This legacy effect manifests itself via internal loading, even when external sources are reduced either permanently via remediation, or seasonally when inflows decrease to a minimum. Internal loading of P can occur as a result of changes in redox potential, wind-induced mixing, and metabolic processes of the biological community. Quantifying the internal load which contributes to cyanobacteria blooms is often difficult, but is vital to make optimal management decisions. Fernan Lake receives a large (856 mg·m-2·yr-1) external load of P that enters the lake primarily in the spring. Interestingly, in summer when HABs are prevalent, external inputs are at a minimum, suggesting that HABs in Fernan Lake may be driven by internal loading.
Assessing the internal loading of Fernan Lake requires understanding its unique characteristics. Fernan lake is shallow bathtub-shaped basin with east-west orientation. At its highest water level the deepest location Fernan Lake is roughly 8 meters (see Figure 8).
Many lakes exhibit thermal stratification meaning temperature descreases with depth. However, Fernan Lake is polymictic because of its relatively shallow uniform depth and basin shape. It only weakly stratifies for short (1-2 week) periods in summer (IDEQ 2013, observed data, see Figure 9). During winter the lake is usually ice-covered for approximately three months between December and March.
During this study, the unusually low winter snow pack and early spring rain-on-snow events resulted in an early and short runoff period. This type of hydrograph limits the remediation strategies. Lake managers should start from a wholewatershed perspective to pin point the sources of sediment entering the creek and then proceed to decide on the most plausible management practice. Until a watershed assessment can be performed, nitrogen additions to improve water quality can potentially be a viable option for the short term. Future studies should focus on careful monitoring of the TN:TP ratio and the viability of nitrogen additions as a potential management strategy at Fernan Lake. Overall, without controlling the non-point sources to the creek, Fernan Lake will continue to experience a large influx of P and sediment which will contribute to ongoing cyanobacteria blooms.
Fernan has experienced instances of cyanobacteria since the early 1990’s (Mossier 1993). Over the years the intensity of these blooms has increased. Maintaining high water quality requires keeping oncentrations of nutrients (Nitrogen (N) and Phosphorus (P)) low. Fernan Creek is the main inflow to Fernan Lake and contributes most of the water and sediment to the lake. To remediate Fernan Lake, reducing the influx of P and sediment and treating the inlake sources of P would be the best approach. Controlling the external load will require a wholewatershed approach (see Prescribed Burn). In-lake strategies may also increase water quality, although it is important to note that use of in-lake strategies alone are a temporary measure. Two such strategies are discussed here: Nitrogen Addition and Superoxide Reactor.
The Watershed Erosion Prediction Project (WEPP) has been used to examine areas of the Fernan Watershed that are susceptible to sediment erosion post wildfire. Figure 9 depicts the boundaries of the watershed with the white outline.
The Fernan Watershed is heavily forested. Combustion of vegetation leads to increased risk of erosion after a fire has ran its course. The susceptibility of the watershed to wild fire can be assessed using FlamMap. Figure 10 depicts fire severity over the watershed.
If a wildfire were to occur the resulting sediment loads are likely to be significant and further contribute to the degradation of the lake.
However initiating prescribed burns has been an effective management treatment to mitigate the risk of wild fire and the associated sediment loading. This is because wild fire leads to the complete combustion vegetation resulting in water-repellent soil conditions that are more likely to erode.
Adding nitrogen to the lake when the TN:TP ratio is critically low would likely be the most rapid and the most cost-effective in-lake management. The addition of nitrogen would re-balance the TN:TP ratio and reduce dominance of cyanobacteria. This treatment option would require periodic monitoring of lake nutrient levels to know when addition is necessary and to ensure addition of nitrogen does not result in dangerously high nitrate+nitrite or ammonia levels. Another factor to consider is that nitrogen is considered a pollutant under the Clean Water Act (USEPA 1972) and using nitrogen to treat the lake is potentially controversial.
Another potentially viable solucion is the addition of negatively charged oxygen ions in the form of a superoxide anion using a Kria reactor (Premier Materials 2013). The reactor would increase the dissolved oxygen concentration and enhance/promote oxidation reactions. This technology for lake treatment is fairly novel and limited results are available. The device has been successfully deployed on Grand Lake in Ohio to treat cyanobacteria. A potential advantage of the reactor is that nitrogen and ammonia concentrations drop along with phosphorous (Premier Materials 2013). Laboratory trials have also shown a rapid reduction of the microcystin toxin, suggesting this approach may be a viable strategy. However, the potential drawbask of this technology that it also reduces nitrogen. To-date there has been no consideration of the effects of supersaturation of oxygen in natural waters with respect to its effects on fish or other biological entities. These remain to be discovered, and only then can this strategy be evaluated and compared to the others.