Team Zero E-Mission’s FIRE Summit Presentation

Sohum Belhe, Aadit Shah, Aidan Vu

Research Question

What is the impact of opening electric vehicle (EV) battery recycling plants on PM2.5 pollution levels in surrounding areas?

Data Wrangling

Outcome variable

Our outcome variable is air quality before and after opening an electric vehicle battery recycling plant. Specifically, we examine the amount of NO2 emissions in surrounding areas.

Our data comes from the NASA NOx emissions database. The dataset obtained from NASA Earth Data for our research is namely : OMI/Aura NO2 Tropospheric, Stratospheric & Total Columns MINDS Daily L2 Global Gridded 0.25 degree x 0.25 degree V1.1 (OMI_MINDS_NO2G) at GES DISC. This dataset contains daily global measurements of nitrogen dioxide (NO2) in the Earth's atmosphere. The data is gridded at a resolution of 0.25 degrees latitude by 0.25 degrees longitude and is available from 2004 to present. In our research, we exclusively utilized data specific to the United States, employing the filtering options available on the website.

In the code below, we combined our own research of EV battery recycling plant data, and queried the data from the NASA database. We processed this data, extracted it, and compiled it in a dataframe.

library("tidyverse")
library("terra")
library("raster")

factories<-read.csv("Factory Data.csv")
factory_plot <- vect(factories,geom=c("Longitude", "Latitude"))
bf<-buffer(factory_plot, width=10000)

# starting index
i<-2 

df3<-c()
for (i in 2:501) {
  print(i)
  n<-nc_open(paste0(path, f[i]))  
  r<-brick(paste0(path, f[i]), varname="ColumnAmountNO2")
  sr <- as(r, "SpatRaster")
  rp<-project(sr, "+proj=longlat +ellps=WGS84 +datum=WGS84 +no_defs ")
  
  df<-extract(rp, bf, fun="mean")
  
  df2<-data.frame(df, filename=f[i])
  
  names(df2)<-c("ID", "NO2", "Filename")
  
  df3<-rbind(df3,df2)
}

Treatment variable

Company	Location	Latitude	Longitude	Opening Year	Opening Month	Process
Li-Cycle Holdings Corp.	"Rochester, New York"	43.19823767	-77.67825568	2022		Hydrometallurgy
Li-Cycle Holdings Corp.	"Gilbert, Arizona"	33.33090907	-111.7293214	2022	May	Hydrometallurgy
Li-Cycle Holdings Corp.	"Tuscaloosa, Alabama"	33.22030374	-87.64053777	2022	October	Hydrometallurgy
Redwood Materials	"Carson City, Nevada"	39.20098488	-119.7371803			Hydrometallurgy
American Battery Technology Company	"Reno, Nevada"	39.44152083	-119.7641779	2023		Hydrometallurgy
Glencore Recycling	"San Jose, California"	37.31085584	-121.8696997
Ascend Elements	"Covington, Georgia "	33.61454146	-83.83656686	2023	March	Hydrometallurgy
Inmetco	"Elwood, Pennsylvania"	40.85721477	-80.27311203			Pyrometallurgy

Our treatment variable is the opening of the lithium ion battery recycling plants. We mapped the data, collecting the longitude, latitude, and opening dates from separate factories chosen from the table shown above, our map is shown below.

install.packages("tidyverse")
library("tidyverse")
install.packages("raster")
library("raster")
install.packages("simplermarkdown")
library("simplermarkdown")
install.packages("terra")
library("terra")
install.packages("tidyterra")
library("tidyterra")
factories<-read.csv("Factory Data.csv")
factory_plot <- vect(factories,geom=c("Latitude", "Longitude"))
s <- vect("state/cb_2018_us_state_20m.shp")
crs(factory_plot) <- crs(s)
plot(factory_plot)
plot(s, add = TRUE)

Below is our factories mapped on a map of the US.

factories<-read.csv("Factory Data.csv")
factory_plot <- vect(factories,geom=c("Longitude", "Latitude"))
s <- vect("state/cb_2018_us_state_20m.shp")
#f<-project(factory_plot, "+proj=longlat +ellps=WGS84 +datum=WGS84")
crs(factory_plot) <- crs(s)
plot(factory_plot)
plot(s, add = TRUE)

Preliminary Results

Display a figure showing how the treatment variable impacted the outcome variable.

df<-read.csv("nasa_no2.csv")

open1<-as.Date("20220101", format='%Y%m%d')
open2<-as.Date("20220501", format='%Y%m%d')
open3<-as.Date("20221001", format='%Y%m%d')

df2<-df %>%
  filter(!is.na(NO2)) %>%
  filter(ID<=3) %>%
  mutate(date=paste0(substr(Filename,6,9), substr(Filename, 11,14))) %>%
  mutate(date=as.Date(date, format='%Y%m%d')) %>%
  mutate(PrePost=ifelse(ID==1 & date<open1, "Before", "After")) %>%
  mutate(PrePost=ifelse(ID==2 & date<open2, "Before", PrePost)) %>%
  mutate(PrePost=ifelse(ID==3 & date<open3, "Before", PrePost))

ggplot(data=df2, aes(x=factor(ID), y=NO2, fill=PrePost)) +
  geom_boxplot() +
  theme_bw() +
  labs(title = "NO2 Before and After Opening", x = "Factory ID", y = "NO2 Value", fill = "Opening Status")