Abstract

Introduction

The objective of this innovation project is to create a floating wetland system for the conditions of Alberta. The main purpose of the system is to filter agrochemicals, focusing on nitrate and phosphate removal in bodies of water through the use of plants and phytoremediation. The goal is to create a system that can not only be cheap and sustainable, but also demonstrate effectiveness over reducing nutrients in water bodies.

Problem

Unfortunately, I could not upload all the images. Please visit my logbook for additional images related to this section.

Designing the System

After months of research and exposure to different fields, doing a project about water is highly beneficial to the world.

Water pollution is a huge problem. This project may focus on agricultural pollution.

A possible project idea is a water filtration system:

objective of the project: create a water filtration system that tackles agricultural pollutants such as pesticides, soil and fertilizer through the use of nanoparticles, chemical catalysts and various microorganisms. 

There are many factors to consider with this innovation.

• What materials are needed?

• How will the different parts of the filtration be separated? By layers? If so, how will I build the layers themselves? 

• How will I test for the purity of the water?

water purity may be tested using the following: 

• PH strips

• Phosphate detection kits

• Pesticide testing kit

• Dissolved oxygen testing kit

Whilst coming up with project ideas, I created a few very rough sketches of what the final product may look like:

Above, is the first sketch I had made of my prototype, though it does need a large amount of adjustments. 

The same day I had also made a slightly more detailed sketch of my initial idea. It was clear from the beginning that some aspects of the prototype needed a lot of edits, the tube mechanism, for example, was something I did not like at that moment.

Later on, I considered the idea to include buckles and rubber to keep the different layers air-tight. Overall, the concept of this design was to create a filtration system people could use at their homes specifically for agrochemicals. 

However, this did not make sense for a variety of reasons. For example:

• One could simply use a multi-purpose filter which would work for agrochemicals

• It would be a cost-effective way to target the source itself.

From the above ideas, two main designs were created:

1. Creating a floating filter system.

2. Creating a filter system that diverts water to another area to be filtered.

In the floating system, the idea is to keep it floating so that it would not disturb aquatic life below. This is one of the most essential aspects to consider. To ensure that the design stays afloat in water, the aerodynamics of it must be considered. Motors may also be included in this design.

In the Water Diversion system, water will be directed to the filtration system and redirected to a cleaner area. It will likely use gravity to collect the water and divert it. It will also make use of many pipes. 

Both systems will make use of the “external testing area” which will be a part of the filtration system where the water is tested for nitrate, PH, turbidity and TDS. This is important to ensure that the system is functioning properly. 

In the end, the floating filtration system was chosen because it will most likely work better for what this is being designed for. This is because the water filtration system is not meant to be a permanent solution, instead, it is a way to purify currently-polluted water. Having flexibility in terms of moving the filter makes it easier to stop the use of it when the time comes. Furthermore, this design has limited interference with underwater wildlife. These are the reasons as to why this design was chosen in the end. However, there are some problems with the floating system:

• Water re-pollution

• Aerodynamics

• Noise Pollution

Originally the plan was to use sensors and an arduino to examine the purity of the water before release but after a discussion with a professional, it was realised that using aquarium test kits is cheaper and more accessible. However, I still want to include AI in the design and I am currently considering using AI to somehow reduce the disturbance to animals. Possibly for aquatic animal behavioural monitoring? 

Another idea which may be included into the final design is a turbine to generate hydroelectric power alongside solar panels. 

This system produces electricity as well which may be beneficial to people. 

This design uses AI, solar panels, a turbine for hydroelectric power generation and aquatic plants for bioremediation and nutrient absorption. The most space is given to the plants because bioremediation is one of the main focuses for the project. There are a few options for the plants that will be used. Some examples that are being considered include: Cattails, Algae, and Bulrush. A mixture of multiple plants will likely be used for biodiversity and to further replicate a natural environment. Though a Raspberry Pi would have to be used, AI can be applied to the project to avoid the filter disturbing the natural environment. This will be done by pattern identification and avoiding areas where many aquatic animals pass through. Solar panels and turbines can be used to generate electricity not only for the system but the energy may also be stored and used for other purposes such as powering appliances in homes. 

However, after some further consideration, it was realised that turbines will not work in this specific system because the most nutrients and agrochemical pollution actually end up in waters that are primarily static. For example, lakes and wetlands. Because of this, it would be immensely difficult to make much significant contributions to electricity generation. 

Later, I decided to remove the AI and motor along with the methods of renewable energy such as the solar panels and turbine. This is because in Alberta, water has strong circulation. So, if the system is deployed in one area, it can purify the entire lake. Also, AI can make the system much more expensive. Removing AI will make the system more accessible to people who could not afford expensive technology.

The design above is the filter that utilises only plants and other abiotic factors such as pebbles, gravel, sand, and charcoal. The sand was moved to the end of the plant section to slow the speed of the water flow and charcoal was kept at the end to absorb any potential toxic chemicals from the plants. 

Why was hornwort, cholera, and bulrush chosen?

For this system, resilient, hardy, and fast growing plants were required. Hornwort is an aquatic plant that grows extremely quickly. It is often used to reduce nitrogen, phosphorus, and its forms in aquariums, furthermore it is very common in Alberta. Bulrush is an emergent plant. It is one of the most common wetland plants in Alberta and is very effective in reducing nutrients in water. Cholera is a type of algae. Though I may face some problems with out-competing bulrush and hornwort, Cholera is the fastest at reducing nutrients in this list. A study was done showing these results: Cholera was able to reduce close to 90% of nitrate at a concentration of 0.25 g/L over the course of eight days

To make the system even more simple and affordable, many aspects were removed. This is the most cost effective design so far. This involves the use of plants such as hornwort and bulrush as well as abiotic factors like rocks and bottles for buoyancy. In this design I also decided to remove the algae because the chances of the algae blooming are high and also have a chance of out-competing the main plants which will affect the system’s ability to reproduce its plants in the lake/wetland. 

It was also determined that, in a real application, the system would have to be deployed from mid-late spring until fall. This is because spring is when farmers use the most fertilizer. A lot of this ends up in waters, requiring additional support of this system. 

Later on, this design was created. It made use of charcoal instead of pebbles. The pebbles were removed to reduce weight. Charcoal is great at reducing many forms of pollutants. Other than this change, most aspects were kept the same. 

Unfortunately, it was impossible to bulrush during December. Because of this I resorted to using a floating plant, duckweed as a replacement. Duckweed is very fast growing and can double the size within two days. This also means that duckweed absorbs nutrients very quickly. However, problems related to the duckweed escaping may be faced with the design preceding this. Also, in the same design, the use of floating plants can drastically reduce the light the hornwort can receive. Because of this, I had the idea of creating various sections made of plastic water bottles to house different plants. Through doing so, the hornwort can still easily receive light. 

The charcoal is placed in the center of nine bottles which have been cut. The hornwort/bulrush is placed around the charcoal. Plastic bottles are used throughout the system which is actually cheaper than the previous design. 

Estimating the cost:

• 24 plastic water bottles cost roughly $3.27. 13 were used for the system. Cost: $1.77

• Charcoal. Cost: $4.17

• 4.17 + 1.77 = 5.94 Total cost for one system: roughly $5.94

Five dollars is extremely cheap. Roughly 100x less money to build a real wetland. 

Final Method

Required Materials

Abiotic components:

• Components of the FWS

  • Abiotic Components

    • Charcoal (8-10 pieces) 

    • 4400 cm^2 of stainless steel woven wire

    • 13  plastic water bottles

    • Several zip-ties

  • Biotic components:

• Bulrush

• Hornwort

• Nitrifying bacteria 

• Testing Setup

  • 2 Plastic tubs with a capacity of at least 29L

  • Any liquid fertilizer that contains a 10-15-10 ratio

  • Water

  • One pack of phosphate aquarium testing kits

  • One pack of nitrate aquarium testing kits

  • Timer

  • One pack of ammonium test kits

  • One pack of ammonia nitrogen test kits

  • One pack of pH test kits

  • One pack of Free Chlorine test kits

  • Access to a computer for data collection

  • Secchi Disk

    • Any disk (a play dough cap was used in this experiment)

    • A permanent marker

    • String

    • Weighted object

• Tools

  • Measuring spoons for measurement 

  • Spoons for dissolving

  • Ruler

  • Pencil

  • Camera or device with a camera

  • Compass

  • Scissors

  • Weighing scale to measure mass

• Safety Equipment

  • Several pairs of gloves

Building the FWS

1. Gather the following equipment:

   1. Charcoal 

   2. 2200 cm^2 of stainless steel woven wire

   3. 2 plastic tubs with a capacity of at least 29L

   4. 13 plastic bottles

   5. Scissors

   6. zip-ties

   7. Phone for a camera

   8. Compass

   9. Plants

      1. Preferably, choosing plants native to one's area is beneficial

      2. For this project hornwort and duckweed was used

   10. Nitrifying Bacteria

2. Assemble the base

   1. Cut 13 plastic water bottles in half so that each will be around 10 cm tall.

   2. Using the bottom part of the plastic water bottles, remove the bottom of each bottle to create a “tube”

   3. Use the “tube” parts of the plastic water bottles and attach 9 of them into a 9x9 grid as such (using a drill to make holes on the sides and zip-ties to attach the bottles together):

      1.  

   4. On all of the bottom of each bottles, place metal mesh, cut to measurements of 3 cm x 3 cm

      1. This is to act as something for the plants to “grab” onto, without using soil.

   5. Repeat step “2d” two more times

   6. Gather four additional plastic bottles

      1. Attach four of the bottles to the outside of the 9x9 plastic water bottle gird using gorilla glue

   7. Repeat step “2f” on all of the grids of plastic water bottles.

3. Adding Biotic and Abiotic elements

   1. Firstly, insert the Abiotic elements.

      1. Inside the middle hole with metal mesh on the bottom, add the bag of charcoal

      2. Insert Biotic Elements 

      3. Measuring starting height of the plants

      4. In one system:

         1. Transplant the hornwort into 4 holes of the system as such:

         2.  

         3. In the 4 other holes, transplant the duckweed. 

Setting up testing

1. Fill 2 buckets with about 25.85  litres of water

2. Add exactly 12.5 mL of liquid fertilizer to each bucket

   1. Using a spoon, dissolve the Miracle Gro powder by stirring until the water appears clear

3. Place the buckets in areas with good sunlight

   1. Make a 1 x 2 grid with the buckets, as such:

2. In front of each bucket, add a sign to signify the types of testing (with system, without system, only hornwort, only bulrush)

4. Measure phosphate and nitrate levels using aquarium testing kits

Testing

1. In one bucket, the finished floating wetland system. Leave one bucket empty.

2. Add Nitrifying bacteria to each bucket

3. Make daily observations

1. Record the height of the plant every day

2. Make qualitative observations based on the health of the plants

      4.    After one week:

2. Measure and record the turbidity of the water using the secchi disk made earlier

3. Measure and record the phosphorus and nitrate levels in the water using aquarium testing kits

      5.   Repeat steps “2a - 2b” every week for two weeks

Data Collection 1.0

Quantitative Observations

System deployed bucket:

Control bucket:

Growth of Plants

*For hornwort and duckweed, the numbers refer to the bottle they were kept in:

Qualitative Observations

Official testing began on Dec. 30, 2024 and ended on Jan. 13, 2024.

Data collection 2.0

As part of science, it is important to continue testing innovations to guarantee and prove function. Because of this, the system is being tested again with a few changes:

• Increase volume of water
• Clear buckets
• Increased testing time

Quantitative observations

System deployed bucket:

Control bucket:

Qualitative observations

The project aims to address agrochemical runoff, particularly by exploring nutrients. Excessive runoff can lead to eutrophication and ecosystem degradation. By optimizing both function and cost, this system has the ability to make significant contributions to environmental preservation and sustainable water management. The two main goals of the project were optimizing function and cost. These goals were proven to be achieved through a series of successful experiments and data analysis. This will be further discussed in this section.

Function Analysis 1.0

Firstly, the function will be examined. Through the observation and analysis of various nutrients, it was found that this system was efficiently reducing many different forms of nitrogen and phosphorus efficiently. Secondly, the cost aspect will also be explored to provide further insights into the effectiveness of the system. Analysis of individual aspects:

pH was measured. Before testing, it was expected that pH could provide insights into the nutrients available in the water due to the fact that pH impacts many chemical reactions as well as living organisms in water systems. For example, in highly acidic conditions, certain nutrients may become too soluble, and in basic conditions, essential nutrients may precipitate and become unavailable. Furthermore, processes like nitrification may impact the pH of the water itself, providing important insights. However, very small changes in the water were noticed. In both buckets, the pH measured 7.6, then increased to around 7.8 and decreased to 7.6 once more. This is a change of roughly only 3% throughout the test. This indicates that the water has a stable pH level. Therefore, the pH did not provide much insight into the effectiveness of the system.

Secondly, the turbidity was also measured. There were small amounts of precipitation during the experiment, likely due to the basic conditions of the water. Turbidity was measured to see if the addition of plants would make any changes to the precipitation of the experiment. However, given that the turbidity was caused by precipitation, there were naturally minimal changes in the turbidity measurements. However, in a real river, measuring turbidity may be highly effective in determining water quality and the impact of plant additions. 

Free chlorine is the available chlorine in a water sample. Chlorine is present in some fertilizers as a micronutrient. It can be in many forms, such as potassium chloride or ammonium chloride. Although they do not serve the same purpose as chlorine disinfection in water, they may behave similarly. This can impact the microorganisms in the water, eventually hurting the plants and other aquatic life. This was an essential aspect to measure due to these reasons. From the graph above, the function of my system could easily be understood. By the end of the two weeks, it can be seen that the control bucket had roughly 1 ppm of free chlorine, yet the deployed system had 0 ppm of free chlorine; the control bucket had 100% more free chlorine. In the graph, it is seen that the system was much quicker than the control bucket. Throughout the testing, the control bucket started at 0 ppm of free chlorine rose to roughly 2 ppm the first week and dropped down to roughly 0.5 ppm by the end of the second week. In the deployed system bucket, the free chlorine started at 0 ppm, rose to 0.5 ppm after the first week, and dropped to 0 ppm after two weeks. The control bucket could only reduce 75% of all free chlorine, but the deployed system bucket reduced it by 100%.

Ammonia nitrogen was not found at all during the whole experiment, likely because of the rapid conversion to nitrification. Possibly, if more water samples were examined, ammonia nitrogen may have been a visible change.

Ammonium was also measured. The graph above shows that the deployed system was slower for this nutrient. This is likely because there were more algae in the control bucket than in the system-deployed bucket. This will be further explored in upcoming sections. The control bucket worked better for this specific nutrient because it peaked at a lower ammonium concentration (50 ppm compared to 100 ppm of the system-deployed bucket) and still was able to remove all ammonium without any sharp increases of the nutrient. However, by the end of the two weeks, the system-deployed bucket was still able to reduce ammonium completely. This further proves the function of the system despite having no algae. 

For nitrite testing, other interesting incidents occurred. From the graph above, it is actually visible that the nitrite levels increased throughout the two weeks. This is likely because nitrite is one of the hardest forms of nitrogen for plants to sequester compared to nitrate, even for algae. Because of this, the nitrite levels can be seen increasing in both the control bucket and the deployed system bucket. However, the nitrite levels in the deployed system bucket increased much more gradually. This can be proven because when averaging the increase of nitrite in each bucket, it was found that the control bucket increased roughly 3.57 ppm per day. In the system deployed bucket, it increased by around 1.43 ppm per day. 

Nitrate testing went almost exactly as predicted. As seen in the graph above, the bucket with the deployed system most quickly reduced nitrate with the least extreme spikes in nitrate amounts. The deployed system was also much faster than the control despite the control having more algae. Nitrate levels are likely one of the most important nutrients to measure because they provide direct insight into the efficiency of the plants' and microorganisms' health as nitrate is the easiest nutrient for plants to sequester. Because of this, high nitrate levels can also lead to a greater likelihood of algal blooms and eutrophication. 

Phosphorus is the easiest identifier of whether an algal bloom will occur. The most common type of phosphorus in fertilizers is orthophosphate. It is extremely difficult to find test kits for primarily orthophosphate so a test kit to identify phosphate was used. The graph above, provides the best insight into the function of the system, showing the reduction of the phosphate in comparison to the algae and bacteria in the control bucket. This can be proven because the control bucket, had an average reduction of 28.57 ppb per day while the control reduced only about 17.86 ppb per day. The system was approximately 60% higher.

Function Comparison Between Various Nutrients

Although the above graph is not a great way to compare phosphate, the peak heights give insight into the efficiency of the system compared to the control bucket and algae. In the graph, the higher the bar, the worse the performance. This is because the bucket with the highest peak has the most amount of nutrients. This graph proves that the control bucket was only able to perform better in sequestering ammonium. This is because algae prefer this form of nitrogen over even nitrate. This is because nitrate has a higher energy requirement for assimilation. Nevertheless, the system was still able to completely reduce ammonium despite having much less algae. 

Average Growth of Hornwort

The growth of hornwort was the least successful aspect of this project. As seen in the graphs above, the height of the hornwort reduced and broke apart by the end. This was due to the lighting situation. The hornwort was kept in buckets like these:

As seen in the image above, the box is completely black which lets almost no light inside of the bucket. A solution to this would be using a clear bucket next time. However, it was very difficult for me to find the same size bucket for an affordable price. Throughout the second week, I had to add some more hornwort into the system. 

Growth of Duckweed: 

The growth of duckweed is very hard to measure because of the large amount of individual fronds making it very difficult and time-consuming to measure each one individually. Because of this, an estimation kwon as a grid-based estimation was used. In this method, a grid is placed on top of a photograph of the duckweed and it is estimated how much duckweed fills the individual squares of the grid. The number is then divided by the total number of squares, in this case 900. Below is the rough estimate of the amount of duckweed throughout the two weeks. For some days the images were lost. EST signifies that it is just a rough estimate based on past trends.  

From the graph above, it is clear that the overall growth of the duckweed was substantial. The growth accelerated in the second week, likely due to the conversion of nutrients. However, there was one outlier. On 01/07/2025, it was found that the amount of duckweed was approximately 63% less than the previous day but returned to normal levels throughout the week.

There are many reasons this may have happened. One possible reason is environmental stress, as the temperature during this week suddenly changed from −11°C to +9°C within 72 hours. This is a roughly 182% difference and could have disrupted the duckweed. Other factors, such as extreme changes in pH levels, nutrient imbalances, or even human error during estimation, may have contributed. It is worth noting that this is an estimate, and perfect accuracy can not be expected.

Function Analysis 2.0

The system worked well during the first test. However, before using it in the real world, it is necessary to conduct extensive studies and research to confirm its function and to make any possible improvements. Because of this, the system was tested in water buckets for a second time. This time I changed a few aspects:

• Clear buckets were utilized, allowing plants to receive more light
• The volume of water was doubled
• Increased the testing time by one week
• water samples were tested for nutrients every three days instead of every seven.

This new setup allowed us to explore the potential of the system even further.

One noteworthy observation is that there was much more periphyton growth in the system-deployed bucket rather than in the control bucket. This will be further discussed in this section.

Throughout the three weeks, fluctuations in the pH were seen, but they were generally small. Near the end of the three weeks, it was seen that the pH increased, and on average, the deployed system bucket had more basic water. This is likely due to microbial respiration, which releases CO₂, creating basic conditions. However, the small drop in pH on days 18 and 21 in the system-deployed bucket was likely due to the death of periphyton. When periphyton dies, it releases organic acids, causing the pH to drop.

The rise in free chlorine in the system-deployed bucket is likely due to the decomposition of the periphyton. As periphyton breaks down, it turns into small organic matter, which then interacts with the chlorine present in water and creates forms of chlorine that are much harder to break down. Because there was much less periphyton in the control bucket, free chlorine was not observed in this bucket.

There was a noticeable ammonia spike on day 9 when both buckets experienced ammonia levels of 0.5 ppm. The ammonia seemed to get converted in the system-deployed bucket but was still present in the control bucket, indicating that ammonia was not fully converted or used. This likely happened either because microbes in the control bucket were slower at converting ammonia or the periphyton may have utilized it. The reason lower periphyton means slower nitrification (ammonia to nitrite to nitrate) is because it is a biofilm that creates a habitat, and more periphyton allows for more nitrifying bacteria.

The control bucket had much more ammonium throughout the testing. By day 21, the control bucket had 100 ppm of ammonium, and the system deployed bucket had none. This means that the control bucket had 100x more than the system-deployed bucket by the end of the three weeks. This is the opposite of what had happened during the previous tests. Last time, the control bucket had less ammonium. This is because many forms of algae prefer ammonium over any form of nitrogen, and because there was more periphyton in the system-deployed bucket, there was also less ammonium.

\\

In the system-deployed bucket, nitrite and nitrate experienced large spikes, only starting to reduce near day 18. The reason for this is likely the decomposition of periphyton as well. As it decomposes, periphyton releases nutrients back into the water because its cells break down and release stored nitrogen or phosphorus. As microbial activity gradually stabilizes, decreases in nitrate levels can be seen. Additionally, nitrification happened much faster in the system-deployed bucket, as seen in the ammonia graph. This means that in the control bucket, ammonia was not being converted to nitrate or nitrite as quickly, likely due to less periphyton and bacteria.

Similar to nitrate and nitrite, phosphate levels had spiked in the system-deployed bucket, eventually reducing near days 16-18. This is likely due to periphyton decomposition as well.

As the cells break down, they may also release phosphorus compounds, increasing phosphate levels in water. After the bacteria gradually stabilized, it was evident that the phosphate levels decreased as well. This fluctuation in phosphate levels shows a strong relationship between periphyton, bacteria, and nutrient cycling in the ecosystem. Understanding these interactions is crucial for maintaining water quality in aquatic environments.

Function Analysis 2.0

This graph shows the highest amount of a specific nutrient throughout the testing. From this graph it appears that the system only performed well for controlling ammonium and ammonia. However, due to the lighting situation, this graph is an unfair comparison. It would be better to compare the nutrients over time instead for more reliable results. This will be further discussed.

Growth of Duckweed/Hornwort/Periphyton:

Hornwort:

Unfortunately, even with clear buckets the light was not enough. In the wild, plants require roughly 2 000 lux to thrive, in my conditions, it was receiving only around 500-600 lux for most of the day which is not enough. Because of this, it was still required to replace the hornwort, even with clear buckets. As a result, data on the growth was not collected.

Duckweed:

The growth of duckweed was measured using the same estimation method as before. A grid was placed on top of an image and then analysed for how much space it would take up. For example:

From the line chart above, it is clear that the duckweed grew. Though it is worth noting that it was replaced on day 5, the growth was still noticeable. From day 6 to 21 the duckweed grew roughly 85%, around 5.64% per day. In the wild, these numbers are usually 2-10% per day so this is in the normal range, even with inadequate lighting. 

Periphyton

For periphyton growth, the Excess Green Index (ExG) was used to measure. This is a method that quantifies the amount of green light relative to red and blue light. Because periphyton contains chlorophyll which reflects green light, ExG is a very useful tool for measuring periphyton and other plants. This method is especially useful as it is possible to analyse the amount of periphyton simply with images instead of needing specialized equipment. ExG values will be calculated throughout a few days in the three weeks. The formula used is the following: 

ExG = 2G - (R+B)

G = green light intensity, R = red light intensity, B = blue light intensity

First, RGB values throughout six different areas of the bucket were collected and the formula was applied to each pixel. Then, the average ExG value was determined.

From the  graphs above, it can be seen that there is more ExG value in the system-deployed bucket. This indicates a higher chlorophyll concentration in aquatic plants as well as periphyton. The ExG value in the control bucket is about -6.333 and 3.833 in the system-deployed bucket by the end of the three weeks. This is roughly 160% more periphyton by the end of the testing period. This indicates that there was more plant growth (including periphyton growth) in the system-deployed bucket. I believe this was because of the light exposure. When the lux levels were measured, it was found that the system-deployed bucket received three times more light than the control bucket. Since clear buckets were used this time, the light difference impacted the growth of periphyton much more. However, this hypothesis must be further tested. 

How did excess periphyton growth impact performance?

My prediction of the rapid increase in nitrogen levels is likely due to periphyton decomposition. As periphyton dies, it releases stored nutrients back into the water, temporarily increasing nutrients in the water. Because of this, it may take more time for the system to fully stabilize nutrient levels. However, based on the data, it already appears that the system is working for nutrient stabilization. This is also something that may happen in the real world so this was a very interesting mistake. However, this experiment should be replicated even more times with the same amount of light to get more accurate results.

Cost Analysis

Traditional floating wetland systems are already highly cost-effective, typically costing around $35. However, this project introduced several improvements to traditional designs. For instance, soil was not used in this system, allowing it to house both aquatic and floating plants, which traditional floating wetland systems do not support. Additionally, it makes use of additional materials such as activated charcoal to reduce nutrients in water. This adaptation helps to mimic a natural wetland ecosystem more closely. Despite these enhancements, the system has an estimated cost of $5.85, making it approximately seven times less expensive than traditional floating wetland systems. This significant reduction in cost makes the system highly affordable when scaling it for broader implications. Additionally, floating wetlands are far more affordable than constructing large artificial wetlands, which cost an average of $ 18,000, reducing the cost by around 3000x. By leveraging these cost benefits, the system offers a practical and affordable solution.

Limitations and Future Improvements

Challenges were faced in many aspects of this project. One of the biggest challenges was figuring out lighting. Due to inadequate light, the hornwort could not grow and consistently had to be replaced during the second week of the first testing period. Even with clear buckets, the light was hard to control as the system-deployed bucket received 3x more light, causing excess periphyton growth. However, this hypothesis should be tested under more controlled environments. 

Another major limitation was the inaccessibility to emergent plants. This test was done during winter, and during this time, no shops sold bulrush, a plant I was originally planning to use. When replicating this experiment, it will be a huge benefit to include bulrush or any other emergent plant native to Alberta.

Human error may also be a potential challenge because some errors in measurement may have occurred, such as adding too much fertilizer or bacteria. A weighing scale would have helped with this problem, but currently, I do not have access to a weighing scale that can measure such small amounts.

The accuracy of the test kits may have also impacted the results. It is almost impossible to identify the exact quantity of a specific nutrient or pollutant in the water just by looking at coloured test kits. Sensors would have made this much more accurate.

Another improvement would be using the system in real freshwater ecosystems to see how it performs under natural conditions. The system could also be optimized to filter pesticides, herbicides, and other agrochemicals. 

The objective of this project was to create an affordable and effective system capable of removing agricultural runoff from freshwater ecosystems. Since then, this project evolved into a practical floating wetland system. The system has proven to be highly effective, as seen in the data. For instance, phosphate reduction in the system-deployed bucket was approximately 60% faster than in the control bucket (28.57 ppb/day versus 17.86 ppb/day). Nitrate levels were also reduced significantly faster, with the highest nitrate concentration in the control bucket being 30% higher than in the system-deployed bucket. Even in the second test with light levels increasing periphyton growth, the system was still able to reduce nutrients near the end of the three weeks. These results further validate the functionality of the floating wetland system.

This project has many applications. Firstly, eutrophication is a problem globally, affecting over 30% of lakes and reservoirs. Excessive algal blooms create hypoxic zones, making these areas inhabitable to aquatic life.  Additionally, the use of plastic bottles helps to reduce waste. With over 300 million tons of plastic produced annually and only 9% being recycled properly, repurposing plastic can help to improve sustainability across the globe. The cheap price also allows people living in first-world countries to gain access to cleaner water. To improve this project, it should be tested in better conditions with more controlled variables such as controlling light lux. Furthermore, additions such as zeolite and emergent plants will further improve the functionality and efficiency of the system. 

• Atwood, D., Paisley-Jones, C., Biological and Economic Analysis Division, Office of Pesticide Programs, Office of Chemical Safety and Pollution Prevention, & U.S. Environmental Protection Agency. (2017). Pesticides Industry Sales and Usage 2008-2012 Market Estimates. In U.S. Environmental Protection Agency. https://www.epa.gov/sites/default/files/2017-01/documents/pesticides-industry-sales-usage-2016_0.pdf
• Biology Dictionary. (2017, June 5). Phosphorus Cycle. https://biologydictionary.net/phosphorus-cycle/
• Clean Water. (n.d.). Wetlands Work. https://www.wetlandswork.org/wetland-benefits/clean-water#:~:text=Since%20the%20flow%20of%20water,of%20creeks%20and%20rivers%20downstream
• Encyclopædia Britannica, inc. (n.d.). Denitrifying bacteria. Encyclopædia Britannica. https://www.britannica.com/science/denitrifying-bacteria
• Environmental Protection Agency. (2012, March 6). 5.6 phosphorus. EPA. https://archive.epa.gov/water/archive/web/html/vms56.html
• Floating wetlands for low-cost wastewater treatment. (2024). Springer Nature Communities. https://communities.springernature.com/posts/floating-wetlands-for-low-cost-wastewater-treatment
• Gonzalez, P. (2021, April 14). Understanding phosphorus. Chem Inc. https://www.cheminc.com/post/understanding-phosphorus
• Government of Canada, Canadian Centre for Occupational Health and Safety. (2024, May 10). Pesticides - General. https://www.ccohs.ca/oshanswers/chemicals/pesticides/general.html
• Government of Canada, Canadian Centre for Occupational Health and Safety. (2024b, May 10). Pesticides - Health Effects. https://www.ccohs.ca/oshanswers/chemicals/pesticides/health_effects.html
• How much do artificial wetlands cost. (2024). Bing. https://www.bing.com/search?q=how%20much%20do%20artificial%20wetlands%20cost&qs=n&form=QBRE&sp=-1&lq=0&pq=how%20much%20do%20artificial%20wetlands%20cost&sc=12-36&sk=&cvid=28777706FBB145C084D97A35A2250F17&ghsh=0&ghacc=0&ghpl=
• Indicators: Phosphorus | US EPA. (2024, May 30). US EPA. https://www.epa.gov/national-aquatic-resource-surveys/indicators-phosphorus
• Mays, C., McLoughlin, S., Frank, T. D., Fielding, C. R., Slater, S. M., & Vajda, V. (2021). Lethal microbial blooms delayed freshwater ecosystem recovery following the end-Permian extinction. Nature Communications, 12(1). https://doi.org/10.1038/s41467-021-25711-3
• Mitchell, C. (2024, May 18). What Is Bioremediation and How Does It Work? Investopedia. https://www.investopedia.com/terms/b/bioremediation.asp
• Nitrogen - Water Quality Guide - DataStream. (n.d.). DataStream. https://datastream.org/en-ca/guidebook/nitrogen
• Nitrate in drinking water. (2024, July 9). Health Information and Services. https://info.health.nz/keeping-healthy/environmental-health/nitrate-in-drinking-water#:~:text=Nitrate%20is%20a%20 naturally%20 occurring,or%20smell%20in%20drinking%20water
• Nutrient Removal — The Wetlands Initiative. (n.d.). The Wetlands Initiative. https://www.wetlands-initiative.org/nutrient-removal#:~:text=These%20naturally%20occurring%20processes%20adsorb,the%20atmosphere
• Nutrients and Eutrophication | U.S. Geological Survey. (2019, March 2). https://www.usgs.gov/mission-areas/water-resources/science/nutrients-and-eutrophication
• OpenAI. (2024). ChatGPT (September 11 version) [Large language model]. https://chat.openai.com/
• Parker, J., Pell, J., Tim, & (Chris), P. I. (2024, June 1). 12 best plants that Clean & Filter Pond Water. Pond Informer. https://pondinformer.com/plants-that-clean-water/
• Phosphorus Cycle | Southwest Research and Outreach Center. (n.d.). https://swroc.cfans.umn.edu/research/soil-water/phosphorus-cycle
• Pollutants from agriculture a serious threat to world's water. (n.d.). Newsroom. https://www.fao.org/newsroom/detail/Pollutants-from-agriculture-a-serious-threat-to-world's-water/en#:~:text=Agricultural%20water%20pollution%3A%20Numbers%20of%20note%201%20Irrigation,sprayed%20into%20the%20environment%20every%20year.%20More%20items
• Runoff. (n.d.). https://education.nationalgeographic.org/resource/runoff/
• Sayadi, M. H., et al. "Removal of nitrate and phosphate from aqueous solutions by microalgae: An experimental study." (2016): 357-364.
• Sposito, G. (2024, August 16). Soil | Definition, Importance, Types, Erosion, Composition, & Facts. Encyclopedia Britannica. https://www.britannica.com/science/soil/Soil-erosion
• Stewart, R. E. (2024, August 31). Fertilizer | Definition, Types, Plant Nutrients, Application, & Facts. Encyclopedia Britannica. https://www.britannica.com/topic/fertilizer
• Taziki, M., Ahmadzadeh, H., Murry, M. A., & Lyon, S. R. (1970, January 1). Nitrate and nitrite removal from wastewater using algae. Latest TOC RSS. https://www.ingentaconnect.com/content/ben/cbiot/2015/00000004/00000004/art00005#:~:text=Microalgae%20have%20a%20high%20capability%20to%20remove%20combined,opportunity%20for%20efficient%20recycling%20of%20nutrients%20in%20wastewater.
• The Editors of Encyclopaedia Britannica. (2009, February 20). Agrochemical | Pesticides, Fertilizers & Herbicides. Encyclopedia Britannica. https://www.britannica.com/technology/agrochemical
• Wetland. (n.d.). https://education.nationalgeographic.org/resource/wetland/
• Wetlands Purify - DNREC. (2024, August 26). DNREC. https://dnrec.delaware.gov/watershed-stewardship/wetlands/purify/#:~:text=Wetlands%20purify%20our%20water%20by,the%20kidneys%20of%20our%20watersheds.%E2%80%9D
• What Are Algae Blooms? (2019, November 22). Sciencing. https://sciencing.com/algae-blooms-8266246.html
• Why are there so many toxic algae blooms this year (cnbc.com). (2016, July 26). https://www.cnbc.com/2016/07/26/why-are-there-so-many-toxic-algae-blooms-this-year.html
• Zumdahl, S. S. (2024, August 2). Ammonia | Definition & Uses. Encyclopedia Britannica. https://www.britannica.com/science/ammonia
• Wikipedia contributors. (2024, August 6). Thiocarbamate. Wikipedia. https://en.wikipedia.org/wiki/Thiocarbamate
• Wikipedia contributors. (2025a, January 17). Carbamate. Wikipedia. https://en.wikipedia.org/wiki/Carbamate#cite_note-J%C3%A4ger-1
• Wikipedia contributors. (2025b, January 19). Organophosphate. Wikipedia. https://en.wikipedia.org/wiki/Organophosphate
• Wikipedia contributors. (2025c, February 4). Organochlorine chemistry. Wikipedia. https://en.wikipedia.org/wiki/Organochlorine_chemistry
• Cost of constructed wetlands systems - Brown and Caldwell. (2020, August 4). Brown and Caldwell. https://brownandcaldwell.com/papers-and-reports/cost-of-constructed-wetlands-systems/
• Cost of constructed wetlands systems - Brown and Caldwell. (2020, August 4). Brown and Caldwell. https://brownandcaldwell.com/papers-and-reports/cost-of-constructed-wetlands-systems/
• Admin. (2023, June 13). What is Eutrophication? -  Definition, Causes, Classification, Effects and FAQs on Eutrophication. BYJUS. https://byjus.com/chemistry/eutrophication

I would like to aknowledge Hannah Porter from the University of Calgary who consistently supported me and provided me with feedback throughout the project. Without her knowledge, this would not have been possible. Furthermore, I would also like to thank my teachers, Mrs. Burkell, Mr. Bykovskikh, and Mr. Stranzinger who helped to further refine and improve my project. Finally, I would like to thank my parents who helped me with gathering materials and conducting my experiments.