REMOTE SENSING/ GEOGRAPHIC INFORMATION SYSTEMS
NASA/NOAA FLOOD MAPPING PROJECT
Objective: Todevelop a 3D flood map of Hurricane Matthew’s impact in settlement areas using UAV imagery and LiDAR data.
Method: Applied spatial statistics to estimate flood depth with DEM and flood extent using Arcpy.
Outcome: Identified flood risk levels in settlement areas, aiding emergency response and future adaptation strategies.
3D flood risk map with building footprints, and roads for Greenville, NC
Estimated floodwater extent from post flood aerial imagery.
3D flood risk map with building footprints for Grifton, NC
Objective: To identify plantation farms most at risk of infestation from nearby swamps.
Method: Conducted proximity analysis to classify farms into high, neutral, and low-risk categories based on distance.
Outcome: Provided a risk assessment framework to support mitigation efforts.
MAPPING HISTORY PROJECT
Farm planation infestation risk map
Statistics chart showing the number of farm plantations at each risk level.
URBAN GREEN PROJECT & SPATIAL MAPPING
Objective: To support community led green initiatives through spatial mapping.
Method: Planted trees, created walk paths in public parks, and mapped projects using ArcGIS Pro and Field Maps.
Outcome: Improved urban green spaces and created a geospatial record of initiatives.
Tree planting initiative; Watson and Basset street, New Haven , CT
Tree and shrub planting initiative; Kensington Park, New Haven , CT
Park management and restoration; Fairmont Park, New Haven , CT
GREEN DEVELOPMENT ANALYSIS
Objective: Investigated the development trends in New Haven to inform the creation of a Green Ordinance for sustainable growth.
Method: Analyzed data from the City Plan Commission (CPC) and the Board of Zoning Appeals (BZA) to examine neighborhood-level development trends.
Outcome: Identified development patterns and their implications for New Haven’s green initiatives and zoning regulations.
Map showing counts of development project applications per census block
Special exceptional approvals per neighbourhood over time
Site plan approvals per neighbourhood over time
NASA ENVIRONMENTAL JUSTICE PROJECT
Objective: To assess urban heat and tree canopy cover in New Haven for better urban planning.
Method: Performed block-level analysis of street trees and compared temperature and NDVI with built environment data in New Haven, CT.
Outcome: Provided geospatial insights to guide tree planting initiatives by analyzing urban heat and tree canopy distribution.
Building counts and mean temperature (avg. 2022) per census block.
Number of buildings per block.
Comparative analysis of street trees and building counts per census block.
SPATIAL ANALYTICS
Objective: To estimate building heights using 3D point cloud data.
Methodology: Processed LiDAR point cloud datasets in ArcGIS Pro.
Outcome: Generated accurate building height estimates for campus structures.
Lidar Point Cloud Analysis
LiDAR 3D point cloud for buildings
Fire Emergency Response Optimization
Objective: Analyzed fire truck accessibility and response times in Accra, focusing on Katamanto Market.
Method: Mapped optimal fire truck routes and evaluate fire station coverage against NFPA standards.
Outcome: Identified gaps in coverage, showing fire stations exceeded the NFPA standard distance, potentially affecting response times.
Fire truck routes analysis
Response Time analysis
Objective: To identify difficult-to-enumerate areas for the 2021 Census for the 2021 Population and Housing Census.
Methodology: Analyzed and combined building footprints, road networks, and population counts in ArcGIS Pro to determine enumeration difficulty levels.
Outcome: Helped allocate resources to challenging areas for better census planning.
Census Enumeration Optimization
Results for 1 enumeration area.
DATA COLLECTION AND ANALYTICS
Objective: To create a comprehensive datasets for training and evaluating AI algorithms to predict post flood depth.
Methodology: Utilized ArcGIS Pro and Arcpy to process and analyze aerial imagery and 3D point cloud.
Outcome: Generated flood datasets with 10,000 256 by 256 tiles.
Flood depth data for AI algorithms
Data creation Flowchart
Real-Time Spatial Data Collection & Monitoring
Objective: Designed an efficient field data collection system for gathering ground truth data on 600 buildings in Ghana.
Method: Used ArcGIS Pro, Online, Field Map, and Information dashboard to design, develop and deploy data collection and monitoring pipelines for collecting and monitoring field data in real-time.
Outcome: Improved efficiency, data organisation, real-time monitoring, and preliminary analysis of collected data.
Objective: To estimate and analyze building infrastructures with AI across cities in Africa.
Methodology: Processed 60,000 OSM building footprints into raster masks using QGIS, ArcPy and R for model training.
Outcome: Created building density datasets for training deep learning algorithms.
Urban Africa Project
ArcGIS Field Map app
Information Dashboard for real-time monitoring and preliminary analysis
OSM building footprint in 30 sq.m grids
Created raster mask showing building area coverage per 30 sq.m
3D Visualization of building footprint, 30 sq.m grids, and raster mask.