Projects
MAPC Multi-Hazard Research
Heat and Air Pollution are frequently expressed as separate aggregations up to the census tract or town average. However, there is significant variation within these aggregations, even between adjacent streets. These microspatial inequities were identified within Boston in a paper from OBrien & Mueller (2023). The impetus of this project was to verify how these differences are expressed on a regional scale and use information about nearby infrastructure to identify drivers of heat and air pollution. Preliminary findings indicate that these microspatial inequities remain significant throughout the region, adding additional nuance to how we understand and address environmental hazards across a wider variety of community types.
I merged street and parcel data by heavily simplifying street names in both datasets and joining the closest matching street. I used this along with the structures dataset to determine the effective street height and width which determines the air flushing rate that affects air pollution exposure. The streets are then used to extract the satellite data on carbon monoxide emissions, land surface temperature, albedo, tree canopy, and impervious surface percentage to examine the drivers of heat and air pollution along with some other factors. Putting these factors into a mixed effects model allows me to understand how local and nearby factors affect the livability of each street.
The final models and findings are currently in the drafting process, but some things are very clear. The microspatial inequities that O'Brien & Mueller found in Boston continue to persist throughout the region. Streets are no longer the clear highest source of variance, but street level infrastructure is the biggest determinant of conditions across scale. Heat and air pollution are also rather strongly correlated together to create overall high hazard areas. This is illustrated by the first two maps showing neighborhood level exposure to each and the within-town correlation between heat and air pollution. Recent studies also show that the combination of these two hazards together significantly increases mortality rates (M. Stafoggia et al, 2023). This combination is also negatively correlated with income, indicating that poorer residents suffer much higher exposure levels than their wealthier counterparts.
Click the link below to view my iSUPER retreat poster on this topic:
iSUPER Poster
Boston Food Desert Study
I currently live in a part of the city that is about a 20-minute walk away from the nearest grocery store. This combined with the lack of tree canopy on that walk led me to use the MBTA rapid transit system to take the train to a grocery store located very close to a station. The travel time is about the same, but it's much cooler and easier. I wondered what other means are available to bridge the gap between where people live and where they buy fresh food.
I used MassGIS data on roads and grocery stores to examine the state of food deserts in Boston. The road data was leveraged to create a network in ArcGIS with a time cost for traversal based on the average walking speed of a middle-aged adult. I could then use grocery stores and MBTA stations as nodes to create service areas. The gaps in service could then be used to apportion census data and approximate the proportion of residents that are out of reach of grocery stores without a faster mode of transportation.
The results were pretty encouraging for the state of Boston food access. The vast majority of Boston residents live within a 20-minute walk of a grocery store, and most of the gaps were in unpopulated areas such as parks or the airport. However, the gaps that did exist should be considered when planning new bus routes and transit expansions where they are not already served.
Click the link below to view my presentation with the rest of the analysis and more maps:
Project Presentation
BlueBikes Equity Analysis
BlueBikes has been a resounding success of a program. Ridership numbers have increased consistently, even through the pandemic. However, it's clear that BlueBikes is not equally successful across all neighborhoods and segments of Bostonians. I had the opportunity to talk to a group of high school students primarily from Roxbury about BlueBikes. Most of the students had heard of BlueBikes, but none of them knew anyone who used the system. One student remarked that it seemed pointless to rent a bike when they already had a car that could take them anywhere.
This led me to investigate which demographics were using BlueBikes. BlueBikes releases anonymized data about trips taken each month. My goal was to focus on commuter behavior as that is one of the best use cases for BlueBikes and it helps me make some assumptions for analysis. To do this, I filtered the data to weekdays and BlueBikes subscribers. Only rides started between 7:00 and 9:59 or rides ended between 4:00 and 6:59 would count towards each station's total ride count. The aim was to only count ridership from riders' "home" station which I would use to create a model with various census demographics.
The resulting model found a significant relationship with income, GINI (Inequality Measure), and the population proportion Ages 18-34. This affirmed BlueBikes' popularity among college students and hinted at an economic barrier to BlueBikes usage. This study was done before the significant price decrease for income-eligible Boston residents, and the network of stations and bike lanes has only continued to expand. I hope these programs will continue to encourage wider adoption.
The outcome of this project was a guided lesson intended to demonstrate how to extract insights from open data sources. Check out the lesson here:
BlueBikes Guided Lesson
Eastie Farm Radius Map
Community Supported Agriculture (CSA) programs allow people to connect directly with farmers to subscribe to the harvest of one or several partnered farms. This model helps farmers with money upfront at the beginning of the season and helps them sell some less well-known regional crops. It also brings people in to share in the risks of farming so bad harvests from weather or insects will be less likely to put the farm out of commission.
I am a long-time CSA subscriber at my neighborhood's local community organization called Eastie Farm. This organization is a hugely important community asset with a "pick your price" style of CSA shares allowing more fortunate residents to support their neighbors. I love having access to fresh and local food and learning to cook with produce that I never used before through their CSA program which has grown massively over the last few years. They also organize many local events to educate and strengthen community ties.
When a staff member sent out an email asking for people with GIS experience, I was quick to respond and offer myself up to help. They wanted a map to highlight their CSA partner farms and show where they all are. We eventually landed on the map on the left here with a combination of individual and aggregated farms to highlight individuals where possible but prevent farms from overlapping.
I'm looking forward to continuing to work with Eastie Farm whenever they need to create maps in GIS. I'm also looking forward to continuing to support this wonderful organization and local farmers.
If you aren't already, consider joining a local CSA. See if there are any close to you:
Local CSA List