The state and federal partners have invested heavily in restoration efforts, amounting to nearly $1.5 billion of state and federal funding in 2019 alone. I am interested in seeing whether those efforts have paid-off? The indicator I picked up is the "Water Temperature", for example, have the summer or the winter water temperature be higher than usual?

What it does

  1. Data Collection

I chose "Water Temperature" as the indicator. I used the CMC and CBP water quality data set provided by Booz Allen Hamilton. For the water temperature, all of the filtered data sets for the analysis and visualization are from the CMC dadabase. (Link to the raw data)

  1. Data Preprocessing
* Year: I used Python pandas to trim date and time elements from the Date parameter, and filtered 2015 - 2020 as the time window I am interested in.
* Season: I categorized May to August as  Summer , and November to February as Winter.
  1. Exploratory Data Analysis
* Methodology:

1. Stationality: 

    Use AF, PACF and Augmented Dickey–Fuller test to check stationality. The water temperature is stationary time-series data.

2. Summary Statistics: 

    After plotting water temperature within 2015 to 2020, I found the temperature range for the winter has something interesting to look at. Especially after 2017. 

3. Geospatial Maps: 

    I used the 5-year water temperature for the color scale, so I would be able to see whether a certain range of temperature increases in certain area. In 2020, there are several unusual high temperature points.

How I built it

Data Visualization

I created a seasonal overview dashboard in Tableau, using year, season and state parameters to invite end users to take a look at the data set. Additionally, With Tableau's geo-spatial mapping feature, I was able to check on the land use in the above region, The Yellow Breeches Creek in Pennsylvania, which is a common fishing resort.

Challenges I ran into

  • Not all the states in the Chesapeake Bay area have comprehensive data in the 5-year window.
  • Most data sets are in large magnitude, and hope to be able to access data sets in the FTP in the near future

Accomplishments that I'm proud of

This project identified the unusual high water temperature in 2020 winter in the Yellow Breeches Creek in Pennsylvania, which is a common fishing resort. Additionally, I designed a Tableau interactive dashboard to invite the end users to check out on the Year-on-Year seasonal water temperature in the Chesapeake Bay area, and drill in to see the red-flagged regions with unusual high water temperatures.

What I learned

Throughout this project, I learned how to pre-process large quantity of geo-spatial data, creating maps with Python geo-pandas libraries, and be able to design a Tableau dashboard on it. I was highly rewarded by the sense of achievement from learning and implementing new things.

What's next for Case Study on Chesapeake Bay Water Temperature


Due to the time limit, I had not yet got time to dig into the residents occupation, conductivity, recreation and harvesting products in the Yellow Breeches Creek in Pennsylvania. From the satellite iamges, I saw most of the land use there is in agriculture. I assume agriculture and fishing might be the major land and water use there. I should have checked out on the data collection time in the specific region, as well as weather condition. I had not checked on the air temperature and should have mapped a confusion matrix to check on the correlation between the air temperature and the water temperature.


In conclusion, I was super thrilled to be able to touch on this topic and have a chance to look at the data sets, read the background documents, and work with my great teammates in the 2020 Hack the Bay Hackathon. Since the state and federal restoration efforts and funding have poured in since 2019, in the summer in 2020, the water temperature did not show high deviation from the previous year, though the winter temperature had several outliers. The top 2 highest degree are in Pennsylvania, and third highest is in Maryland. However, the highest one is 43 degree Celsius, while the second and the third are 26 and 24 degree Celsius. On the "Yellow Breeches Creek, Pennsylvania Fly Fishing Reports & Conditions" website, it did mention about high water temperature, but it stated the water quality is good. As a result, I propose the 43 degree Celsius in the winter might not be a typo from data collection. I look forward to expand this case study further in the days to come.

Built With

Share this project: