Trash Wheel Project

How effective have Trash Wheels been reducing waste in the Baltimore Harbor throughout the years, and the impact of precipitation in trash collection.

Introduction:

  • In Baltimore, Maryland, Mr. Trash Wheel has become a local hero in the battle against water pollution. This innovative device, created by John Kellett, was first installed in Jones Falls Stream in 2014. Since then, it has made a huge impact by intercepting over three million pounds of trash before it can flow into the Baltimore Harbor and then, the Atlantic Ocean. Urban water pollution is a big problem, affecting wildlife, local businesses, and the overall beauty of waterways. Trash in the harbor doesn’t come from people throwing garbage directly into the water, but from land sources. Stormwater washes trash down storm drains, into creeks, and then into the main river feeding the Baltimore harbor. Mr. Trash Wheel works 24/7, collecting garbage from the water. It didn’t take long for the community to notice a big difference. Local businesses and leaders were so impressed that they decided to make Mr. Trash Wheel a permanent part of the harbor. This success led the Waterfront Partnership of Baltimore to launch the Healthy Harbor initiative, aiming to produce and install more trash wheels throughout Maryland. The weather, especially heavy rain, influences how much trash Mr. Trash Wheel collects. By stopping garbage before it reaches the ocean, Mr. Trash Wheel helps keep the environment clean and the harbor beautiful. This project has not only cleaned up Baltimore’s waterways but also inspired other cities and countries to consider similar solutions for their water pollution problems.1 For this project, I decided to study the effectiveness of TrashWheels in reducing waste in the Baltimore Harbor over the years and the impact of precipitation on their trash collection.

Exploratory analysis:

Let’s have a look at the yearly trends of Trash Collected using the variable Volume.

annual_trash <- trash %>%
  group_by(Year) %>%
  summarise(total_volume = sum(Volume, na.rm = TRUE)) %>%
  arrange(Year)
ggplot(annual_trash, aes(x = Year, 
                         y = total_volume
                         )) +
  geom_line() +
  geom_point() +
  labs(title = "Yearly Trend of Trash Collected (by volume)",
       x = "Years",
       y = "Volume (Cubic yards)") +
  theme_minimal()

Figure 1: A line chart of the yearly Trend of Trash Collected

Interpretation:

Over the years from 2014 to 2023, there is a general upward trend in the volume of the trash collected. We could say that as the years go, the Trash wheels collect more trash. We can see a significant peak in 2018, and then a sharp increase from the year 2020 onward. What could explain this ? Maybe the amount of rain ? Maybe it has to do with the Trashwheels themselves ?

Let’s get an idea on the annual precipitation in Baltimore:

But first, some information on the Rain data:

The precipitation data comes from the National Oceanic and Atmospheric Administration (NOAA)2. Using the website’s data access tools, I generated a dataset covering the annual precipitation (in millimeters mm) in Baltimore from 2014 to 2023.

rain <- rain %>% 
  mutate(Date = mdy(DATE)) %>% # Let's get the 'DATE' variable into a date-format
  mutate(Year = year(Date)) # Extracting the 'year' from the date

annual_rain <- rain %>%
  group_by(Year) %>%
  summarise(total_precipitation = sum(PRCP, na.rm = TRUE)) %>%
  arrange(Year)

ggplot(annual_rain, aes(x = Year, 
                        y = total_precipitation)) +
  geom_line() +
  geom_point() +
  labs(title = "Annual Precipitation in Baltimore",
       x = "Year",
       y = "Total Precipitation (mm)") +
  theme_minimal()

Figure 2: A line chart of the annual precipitation in Baltimore

Interpretation

  • There are significant fluctuations in annual precipitation over the years. The total precipitation ranges from around 12,000 mm to over 20,000 mm. However, what is interesting is that there is a sharp peak in precipitation in 2018, where the total amount of precipitation reaches its highest point between 2014-2023.

  • The patterns in annual precipitation appear to align with the trends in trash collection by Mr. Trash Wheel. The peak in precipitation in 2018 likely contributed to the significant increase in trash collected that year. This aligns with the fact that in 2018 the north american blizzard happened 3, causing 2 tornadoes in Maryland. 4

Now that we were able to explain the peak in 2018, what could explain the sharp increase in trash collection from 2020 onward?

Let’s have a look at the different TrashWheel’s annual collection of Trash:

best_wheel <- trash %>%
  group_by(Name,Year) %>%
  arrange(Year) %>% 
  summarise(total_volume = sum(Volume, na.rm = TRUE))
  
  ggplot(best_wheel, aes(x = Year, 
                         y = total_volume, 
                         color = Name)) +
  geom_line() +
  geom_point() +
  labs(title = "Total Volume of Trash Collected by Each Trash Wheel Over the Years",
       x = "Year",
       y = "Total Volume of Trash Collected",
       color = "Trash Wheel") +
    scale_color_brewer(palette="Dark2")+
  theme_minimal()

Figure 3: A line chart of the trash collected by each Trashwheel Over the Years

Interpretation

Starting from 2017, we observe a sharp increase in the volume of the trash collected by the Trash Wheels (Fig1). This upward trend can be attributed to the introduction of two additional Trash Wheels in Baltimore Harbor: Professor Trash Wheel in 2017 and Captain Trash Wheel in 2018. These new additions significantly boosted the trash collection capacity, explaining the rise in collected trash starting from 2019. In 2021, the introduction of a third Trash Wheel, Gwynnda, further amplified this effect, resulting in a noticeable sharp increase in trash collection in the following years. Notably, high precipitation levels in 2015 (Fig2) caused Mr. Trash Wheel to collect more trash than the previous year (Fig3). However, the most significant peak occurred during the tornadoes of 2018. It was only after the introduction of the three additional Trash Wheels that a clear trend of sharply increasing trash collection emerged.

Let’s have a look at which one of the three Trashwheels collected most garbage during the tornadoes year:

trash_2018 <- trash %>%
  filter(Year == 2018) %>% 
  group_by(Year,Name) %>%
  summarise(
    total_volume = sum(Volume, na.rm = TRUE)
  )
  
ggplot(trash_2018, aes(x = Name,
                       y = total_volume,
                       fill= Name)) +
  geom_bar(stat = "identity") +
  labs(title = "Total Volume of Trash Collected by Each Trash Wheel in 2018",
       x = "Trash Wheel",
       y = "Volume (Cubic yards)") +
  theme_minimal() +
  scale_color_brewer(palette="Dark2")

Figure 4: A bar chart of the trash collected by each Trashwheel in 2018

Interpretation:

The graph clearly shows that Mister Trash Wheel collected the most garbage in 2018, with a total volume significantly higher than the other two Trashwheels. This can be attributed to its established presence, and maybe it got better during the high precipitation in 2015. The second highest volume of trash collected in 2018 was by Professor Trash Wheel. Although it was introduced in 2017, it made a substantial contribution to the trash collection efforts during the severe weather conditions of 2018. Captain Trash wheel was introduced the same year, and was the one that collected the least trash.

Extra:

merged_data <- left_join(annual_trash, annual_rain, by = "Year")

long_data <- merged_data %>%
  pivot_longer(cols = c(total_volume, total_precipitation), 
               names_to = "Metric", 
               values_to = "Value")

ggplot(long_data, aes(x = Year, 
                      y = Value, 
                      color = Metric)) +
  geom_line() +
  geom_point() +
  facet_wrap(~Metric, # Plot for each "metric"
             scales = "free_y",  # y-scales of the facets should be independent of each other, free-y is for each facet will have its own y-axis range
             ncol = 1) + # vertical layout
  
  labs(title = "Annual Trash Collection and Precipitation in Baltimore",
       x = "Year",
       y = "Metric",
       color = "Metric") +
scale_color_brewer(palette = "Dark2")+
theme_minimal()

Figure 5: Correlation between Trash Collection and Precipitation in Baltimore

Interpretation:

The graph summarizes the previous findings, illustrating the correlation between the volume of trash collected and annual precipitation in Baltimore. The peaks in trash collection volume, especially in 2015 and 2018, align closely with the spikes in precipitation, reinforcing the impact of rainfall on the efficiency of Trash Wheels.

Conclusion:

  • This analysis demonstrates the effectiveness of Trash Wheels in reducing waste in the Baltimore Harbor and highlights the significant impact of precipitation on trash collection. From 2014 to 2023, there has been an upward trend in the volume of trash collected, with notable peaks in 2018 and from 2020 onward. The peak in 2018 correlates with a sharp increase in precipitation due to Tornadoes, suggesting that heavy rainfall contributes to increased trash collection. The introduction of additional Trash Wheels—Professor Trash Wheel in 2017, Captain Trash Wheel in 2018, and Gwynnda in 2021—significantly boosted the trash collection capacity, explaining the sustained increase in collected trash. Mr. Trash Wheel was particularly effective during high precipitation years, especially in 2018, showcasing its efficiency. Overall, the Trash Wheels have significantly reduced the amount of trash entering the harbor, improving water quality and the environment. This looks like a great initiative to start spreading across world’s harbors!

Footnotes

  1. https://www.cnet.com/science/mr-trash-wheel-is-gobbling-up-millions-of-pounds-of-trash/↩︎

  2. https://www.ncei.noaa.gov/cdo-web/search?d atasetid=GHCND↩︎

  3. https://en.wikipedia.org/wiki/January_2018_North_American_blizzard↩︎

  4. www.cbsnews.com/baltimore/news/tornado-caused-damage-in-dundalk-baltimore-mt-airy-friday/↩︎