An analysis of spatial patterns in nutrient concentration in a constructed wastewater wetland system at Oberlin College: While Nitrate Remainsconcentration is Constantuniform, Phosphate Concentration in a Constructed Wetland at Oberlin College is Dependent on Depth and the Presence of Physical Barriers

[Great job of including results in title. Could be stronger if you found a way to start title with statement that emphasizes broad context. E.g. as above

Greta G. Bradford, Kate M. Coury, Emily C. Minerath

Systems Ecology (ENVS 316) 2008

I. Abstract

Natural and constructed wetlands process nutrients and labile organic matter which in excess might damage downstream ecosystems through eutrophication. Little research focuses on the spatial heterogeneity of nutrients within wetlands or what causes this variability [would be nice to include citation here. Did you find any literature on this?]. The Living Machine (LM) is an engineered wetland ecosystem designed to treat and internally recycle wastewater from the Lewis Center at Oberlin College. Previous studies have examined the competence of gravel to adsorb and remove phosphate adsorption in the LM marsh as well as nutrient and water flow in relation to plant communities. We studied the spatial distribution of phosphate and nitrate concentration in the gravel bed of the marsh,: the last stage of LM water filtration. Nitrate is a limiting factor for plant growth and a key component of the nitrogen cycle; any removal of nitrate in the marsh could be due to denitrification, which is expected to occur near plant growth, which provides organic matter. The low level of organic matter limits the amount of denitrification in the marsh (Haineswood & Morse, 2003). Plant uptake could also account for nitrate removal. Phosphate is another limiting nutrient in freshwater plant systems; studies indicate that the marsh is responsible for the majority of LM phosphate removal through adsorption to gravel (Cernac et al, 2004). Our goal was to assess patterns in nutrient concentrations horizontally and vertically throughout the marsh [why? Explain why this is important]. We found that nitrate is not being processed and does not significantly vary across different points and depths. However, phosphate concentration increases significantly with depth, which we attribute to faster flow at the bottom of the bed caused by the larger gravel present at that level, allowing less time for adsorption by gravel [note that there are two complementary mechanisms here – larger size particles means faster flow, but it also means less surface area for adsorption] . In light of these results we suggest future research considering designs that allow for slower flow, investigating the possibility of adding deep-rooted plants, using smaller gravel throughout, or curving the pathway through which the water flows to provide it with additional time for contact.

II. Introduction

Wetlands and the Living Machine

Constructed wetlands are becoming integral to riparian zone protection, municipal water treatment, agricultural wastewater systems and other applications (Sim et al. 2008). Recent engineered ecosystem research has explored the effectiveness of constructed wetlands as wastewater treatment facilities (Sim et al. 2008, Olguin et al. 2008, Meers et al. 2008, Sindilariu et al. 2008). For example, a study concerning wetland processing of organic rich sugarcane molasses stillage focuses on wetlands and wetland plant efficiency (Olguin et al. 2008). In another study, researchers monitored the ability of wetland cells to remove nitrate and phosphate from rural, urban, and municipal sewage in riparian zones. However, few studies have considered wetland nutrient spatial heterogeneity and its causes. [Why might this be important? You need to explain as part of defining a gap in knowledge. Seems to me that the obvious case to make is that if you treat a marsh systems as a black box, in which you only study what goes in and what goes out (sa other studies have apparently done), you may have a good sense of what happens, but you don’t have a good sense of HOW or WHY it happens; spatial data is necessary in order to understand mechanisms an din order to develop information that would allow for optimization of desirable removal functions]. Studies focusing on this subject have the potential to improve management and future design of constructed wetlands.

The Living Machine (LM) is one such engineered wetland ecosystem that utilizes a series of tanks and a gravel marsh to purify waste water for reuse in the Lewis Center at Oberlin College. The LM is housed in a solarium in the building, allowing light to reach the plants in the LM's open aerobic tanks and marsh bed. Water first enters the LM via a series of closed anaerobic tanks to decompose and settle most solid waste. The largely solid-free water continues through closed aerobic tanks to allow for bacterial processing of organic matter and nutrients. The water then flows into open aerobic tanks, which resemble hydroponic plant tanks and are built into the gravel marsh. In the open aerobic tanks more nutrient processing occurs and nutrients are absorbed into suspended plant roots. Water passes into the floor of the solarium, a meter deep gravel marsh, home to a myriad of plant species [vague. How dense are these plants?], which acts as a final nutrient filtration stage. Lastly, water is treated with UV light to kill bacteria before it is reused as toilet water.

[Your figure of the marsh would be useful here]

Living Machine Marsh

The marsh is a rectangular basin (11.58m x 5.16 m x 0.98 m) filled with gravel ranging in diameter from 4cm -10cm which acts as the floor of the LM. Water entering the marsh still contains elevated levels of nitrogen and phosphorus. The gravel at the base of the marsh is larger than the gravel visible at the surface, which causes water in the bottom of the marsh to travel at a higher velocity (McConaghie, 2003). The water drains east to west due to a 2.0% gradient. Set into the marsh are the open aerobic tanks, the bases of which go all the way to the floor of the marsh, causing water to meander through the marsh at varying velocities. Though other waste processing systems do not necessarily contain a gravel marsh stage, other facilities use similar structures to facilitate phosphate adsorption and denitrification, making a study of the LM marsh applicable for comparison [cite examples of other such treatment plants that you are discussing here].

Nutrient Dynamics

As a key nutrient for flora and fauna, nitrogen is an important element to consider when studying ecosystem dynamics. The nitrogen cycle is a biogeochemical cycle via which nitrogen passes through solid, aqueous, and gaseous phases. It involves a complex series of transformations between that occur under both aerobic and anaerobic conditions, organic and inorganic forms of nitrogen, mostly mediated by microbes in the soil, water and sediments. Elevated nitrogen levels cause eutrophication in aquatic systems (Ricklefs, 2007).

In the case of the LM, nitrogen enters the anaerobic tanks in the form of labile organic material from toilets and, through ammonification, is transformed into dissolved ammonium (NH4+) ion. Subsequently it is converted into nitrite and then to nitrate through the processes of nitrification in the aerobic tanks (2NH4+ + 3O2 à 2NO2- + 2H2O + 4H+ + Energy; 2NO2-+O2 à 2NO3- + Energy). In tThe marsh is designed so that, water diffuses slowly enough through the gravel bed to maintain the anaerobic conditions conducive to denitrification [If you are going to include the reactions for nitrification, seems like you would want to also include reactions for denitrification]. However, low levels of labile organic matter in the LM march limit denitrification, the process via which nitrates and nitrites are reduced to nitrogen gas (Haineswood & Morse, 2003).

The phosphorous cycle, though simpler than the nitrogen cycle, is equally important for ecosystem balance [what does “ecosystem balance” mean?]. For organisms, phosphorous is a key element of ATP and ADP, nucleic acids, cell membranes, bones, and teeth, making it readily present in organic waste (Ricklefs, 2007). Weathering of soil parent material releases phosphorus for root uptake by plants [not clear what weathering of parent material has to do with your particular study; try to focus information so that only info that bears directly on your work is included]. Phosphorus does not have a gaseous phase, so it will remain in the plant body until decomposition, at which point it is returned into the soil in the form of soil organic matter [OK, but what does this have to do with the LM, a system in which decomposition takes place in the water? In what form does P occur as a dissolved inorganic ion?]. Phosphorus also acts as a limiting factor in aquatic systems. Anthropogenic sources of phosphorus and nitrogen, such as chemical fertilizers or unprocessed human and animal waste have led to an overabundance of limiting nutrients which can lead to algal blooms and extreme disturbance regimes in aquatic ecosystems, contributing to harm of fisheries (Herdendorf, 1987), and endangering human health. For this reason, waste processing and removal of both phosphate and nitrate has a direct impact on the health of downstream ecosystems, and has attracted scientific attention. Though the LM does not have a 'downstream' due to water recycling within the building [in the introduction, you have not indicated that water is recycled yet], knowledge of how phosphate behaves in the marsh is applicable to similar systems where water is not recycled, (de-Bashan, 2003). Our research has thus focused on finding analogous stages [what do you mean by stages?] within the LM for aquatic ecosystem nutrient processing. Freshwater systems, like the LM marsh, act as sinks for phosphorous and nitrogen. Phosphorus concentration drops when it passes through the marsh, as reflected by regular influent and effluent testing conducted by the Living Machine Crew [citation?].

III. Goals and Hypotheses

Our goal is to determine spatial patterns of phosphate and nitrate distribution by studying water samples taken at diverse points and depths throughout the marsh. We [future tense does not make sense since the study is completed] will measured concentrations of phosphate and nitrate both horizontally and vertically. The study will provides a cohesive and detailed view of how nitrate and phosphate are being processed, which has not yet been produced [does it accomplish this?]. This will yield insight into marsh management and may suggest locations for additional plant growth most conducive to increasing nutrient removal as well as decreasing water flow. Spatial patterns in the marsh have been researched in depth by James McConaghie in relation to water flow and nutrient dynamics. Our study of nutrient accumulation and spatial heterogeneity within a wetland system may ultimately lead to more efficiently constructed wetlands, improve management practices, and present a basis for future LM studies. [I sense that this last paragraph is taken from your proposal, which is fine, but you need to recast the text to report on what was done rather than what you plan to do]

As water is pumped into the marsh from the clarifier we hypothesized that the phosphate concentration will would consistently decrease toward the effluent as it is adsorbed onto the gravel. In areas of high flow we predict there will be higher phosphate concentrations relative to areas of low flow (Mitsch et al. 1995). Towards the bottom of the marsh bed the gravel increases in size, allowing for a faster flow. This larger gravel has a smaller surface area to mass ratio than the smaller gravel at the top of the marsh. These characteristics of the deep marsh level will tend to decrease phosphate adsorption and cause the water at the bottom to have a higher concentration of phosphate. Behind the open aerobic tanks, which retard water flow, we predict that phosphate concentrations will be significantly lower than areas in front of the tanks, where flow velocity is higher.

A small decrease in nitrogen levels throughout the marsh could be caused by denitrification. Though the marsh is anoxic, denitrification will be limited by the low level of labile organic matter (Haineswood & Morse, 2003).

IV. Materials and Methods

We attempted to quantify spatial nutrient patterns in the LM marsh through a single collection of data from multiple locilocations. To obtain water samples we used an existing grid of PVC pipe sampling ports that were drilled into the marsh, dividing it into meter squared numbered columns and lettered rows (Figure 1). Column 10 is closest to the influent; 1 is closest to the effluent. Rows lettered A to E describe the width. Each intersection, 27 total, contains 3 PVC pipes cut to 7.62 cm (shallow), 76.2 cm (medium), and 91.4 cm (long) inserted to allow sampling at each depth (McConaghie, 2008). We sampled only once to create a snapshot of a moment in time in the LM marsh. We were aware before the experiment commenced that because waste level input from the toilets is not a constant; there will be spatial variation in our data due to variations in phosphate and nitrate input that occur over time. Our study will, however, provide a picture of nutrient levels in the marsh at a high resolution, to address the behavior of the system at a given time.

Figure 1. Schematic of the Living Machine marsh. Water flows right to left. Each orange dot represents a sampling site consisting of a long, medium, and shallow tube and identified by column (number) and row (letter). The open aerobic tanks and the clarifier reach down to the bottom of the marsh, obstructing water flow. [Very nice diagram]