The National Emissions Inventory (NEI) is an estimation of air pollutant emissions conducted once every three years by the Environmental Protection Agency (EPA). The estimations are comprised of submissions from State, Local, and Tribal air agencies as well as supplemented data developed by the EPA. Sources of air pollutants are categorized as either point, nonpoint, onroad, nonroad, and fire that distinguish similarities in data sources and processing methods. Commercial Marine Vessels (CMV) are categorized as both a nonpoint and nonroad source. Emissions estimates of CMVs are derived from a model that combines automatic identification system (AIS) data with vessel attributes from the Clarksons database and performs calculations based on scientific research.

I co-developed an R computer program that models emissions from CMVs that is currently used for EPA's emission inventory of class 1 and 2 vessels. I currently serve as the technical lead for this initiative where my responsibilities include executing yearly runs of the model, performing quality control checks, communicating problems to scientists, implementing solutions, and updating documentation with model outputs. I also spend time updating the functionality of this program to make it as user-friendly as possible.

The Toxic Releases Inventory (TRI) is a resource developed by the Environmental Protection Agency (EPA) to provide easily accessible information on chemical releases and pollution prevention activities reported by facilities in the United States. The Where You Live section of the TRI National Analysis contains a web map that showcases a summary of TRI releases at various geographic levels for the most recent reporting year as well as US Census population information and Risk-Screening Environmental Indicators (RSEI) model outputs.

The workflow for processing and publishing the Where You Live analysis was performed manually up until the 2021 reporting year where roughly 140 hours were annually allocated to this initiative. I automated most of this workflow within an ArcGIS Notebook which drastically improved processing time. The ArcGIS Notebook allows the analysis to be executed in segments, which is imperative since this workflow is reliant on the delivery and retrieval of data from other contracting firms.

The Toxic Releases Inventory (TRI) is a resource developed by the Environmental Protection Agency (EPA) to provide easily accessible information on chemical releases and pollution prevention activities reported by facilities in the United States. The EPA aims to provide comprehensive and up-to-date geographic information on TRI facilities through a REST Service. This REST Service contains a map server with web GIS layers representing the location and release amounts of TRI facilities from the five most recent reporting years.

This REST Service automatically refreshes biannually with an ArcGIS Python Toolbox that I developed with code recycled from another refresh process. This ArcGIS Python Toolbox pulls in the latest TRI release data from the EnviroFacts API. It is currently operational and scheduled to be executed every year on the last day of July and October. It is featured on well known applications such as EJScreen.

The Environmental Protection Agency (EPA) has an initiative to target facilities that are non-compliant with their standards. There are numerous methods for determining whether a facility is not in compliance such as the use of sensors and establishment of monitoring stations. One method that is currently gaining traction is the use of aerial surveying and satellites to estimate emissions from a facility's plume. Since plume raster files are not tagged to a specific facility, there is a need to perform spatial analysis to identify a potential source of these plumes.

I created the Plume Data Extraction repository on Github that helps scientists and analysts link plumes to facilities. This is done by creating a buffer around a facility (or list of facilities) and extracting all plumes located within it. This repository is comprised of two R markdowns that obtain plume data from Carbon Mapper and NASA EMIT, which are two separate databases containing plume data

The Regional Coastal Integrated Information Management System (RCIIMS) is an online platform for hosting metadata that supports climate analysis and planning activities for six pilot countries (Dominica, Grenada, Haiti, Jamaica, Saint Lucia, and St. Vincent and the Grenadines). The platform has been created under the Caribbean Regional Track of the Pilot Programme for Climate Resilience (PPCR) and is currently operational at rciims.mona.uwi.edu. RCIIMS utilizes the CKAN open-source data portal software as the platform for developing the information management system.

I served as a junior developer for this project where I tested the installation of CKAN and its configuration settings in a virtual machine with a Linux environment, wrote JavaScript code to enhance search bar functionality, developed a Python script to batch import the initial metadata onto the portal, and created custom extensions to meet functionality requirements of our clients at the University of West Indies.