HotPopRobot

Arushi Nath

On 11 December 2021, MIT organized the Youth Solveathon Workshop to give an opportunity to girls under the age of 18 to come together and discuss real-world problems as well as come up with innovative solutions. The main problem discussed was how to reverse climate issues such as deforestation, air and water pollution, and biodiversity loss.

During the workshop, I had the chance to brainstorm solutions with youth from all around the world. They came from the United States, India, Canada, and China. Listening to environmental initiatives taken by them and their ideas on how to take them forward was very inspiring. Some youths talked about forestry and biodiversity initiatives while others talked about cleaner oceans. Based on these topics, we formed 5 teams. As a group, we carried out research on different climate change-related topics and gave a presentation to the other groups on our proposed solutions. My team chose deforestation. One of the ideas we came up with was creating an App that rewards you every time you plant a tree or plant.

Following the workshop, there was an opportunity to apply for micro-grants to transform ideas into prototypes. Interested participants had to apply online and share details, including their ideas, how they would implement, what materials (hardware and software) would be required, how much the costs would be, and the timeline over which they would implement their solution. This was a great opportunity, and I submitted my application to transform my idea about “Monitoring Tree Density of Local Parks” to increase urban forestry and biodiversity in Toronto.

The idea was well received, and in January 2022, I received a notification that my project was selected for a $600 Prototyping Micro-grant. My solution aimed to raise awareness about the benefits of parks among youth and encourage them to take action to help improve their local parks. To do so, I wanted to calculate how green parks were using their tree density. I could rank different parks based on their tree density. To validate these findings, I would calculate the Normalized Difference Vegetation Index of the parks. I also wanted to compare pollution levels in different parks. This separated my project into 3 parts.

First: calculating tree density. To do so, I turned to satellite images of parks. I was able to access high-resolution imagery from the Planet Labs Dove satellites. With those images, I used i-tree canopy tool developed by the USDA Forest Service to help me find the tree density of 3 local parks: Trinity Bellwoods Park, Stanely Park, and High Park.

Second: validating results. To validate tree density results, I used satellite imagery to calculate the Normalized Difference Vegetation Index (NDVI) of each park. NDVI is an indicator used to quantify vegetation greenness. It is useful in understanding vegetation density and assessing changes in plant health. It does this by looking at the reflectivity of the image in different frequencies. I used the python library Rasterio to extract the four bands (Red, Green, Blue, and Infrared) from the image file and then used the NDVI formula to calculate the average NDVI of my park image.

Third: measuring pollution levels in parks. I created a homemade instrument with a pollution sensor to extract the pollution levels of neighbourhood parks. It contained the PM 2.5 pollution sensor, an Arduino microprocessor, a SEEEDUINO shield to connect the sensor with the Arduino, and a Wi-Fi shield to transmit sensor data collected from the Arduino online to a website.

I created an account with ThingSpeak – an IoT analytics platform service that allowed me to receive data from the Wi-Fi shield, and plot the data at a 20 seconds interval. The entire wireless arrangement allowed me to take my homemade device to different Toronto parks and measure pollution levels by simply creating a mobile hotspot using my mobile phone.

The MIT Solv[ED] Prototyping Micro-grant allowed me to test out my ideas. and evaluate its strengths and weak points to improve it further. I am now more determined and confident about taking the project further. I plan to adopt a two-step approach to expand my project to more parks in Toronto and beyond.

First, I will create more versions of my homemade instrument and give them to other youths so they can find the pollution level of their local parks. Second, I will create a web app to store all information I and other youths have collected. The webapp will be coded in python and include an interactive map of Toronto. This crowdsourcing model of collecting data could then be expanded beyond Toronto.