Kimia Chenary
Spatial analyzer. Data Scientist. Researcher
Greetings! I’m Kimia Chenary, a researcher passionate about addressing urban challenges. My work harnesses both big and small data alongside innovative computational methods such as programming and open-source tools to promote smart, sustainable, and equitable urban systems. I have had the privilege of collaborating with esteemed scholars from Japan, Australia, and New Zealand, gaining valuable experience and insights through these partnerships.
I am deeply committed to tackling city-related problems and focusing on topics that offer novelty, enhanced by advanced methodologies, ensuring the research is valuable for future generations and cities.
My research involves:
1. Collecting and managing big and small data using data science techniques. 2. Analyzing the current situation through spatial analysis methods, including Python and Google Earth Engine. 3. Forecasting future scenarios using machine learning or simulation techniques. 4. Modeling interactions within urban systems, encompassing urban environments, transport networks, and human movements.
I am particularly interested in exploring the interactions between various factors within cities to forecast, optimize, and enhance urban functionality for smarter living.
Programming Skills
Python | R | SQL | JavaScript
Open-source tools
Google Earth Engine (GEE) | Spyder & Jupiter Notebook | AnyLogic |JOSM | PostgreSQL
Software Skills
ArcGIS | QGIS | Space Syntax| AutoCAD | Photoshop
Projects
Climate change visualizations using Google Earth Engine (GEE) and Python
Urban Heat Island using Google Earth Engine (GEE)
The map depicts the mean land surface temperature (LST) of Mashhad city, Iran, from 2013 to 2021. The data source is a Landsat-8 TOA (Top-of-Atmosphere) image collection processed using Google Earth Engine (GEE) at a 100-meter scale.
All NDVI values fell within the same range, from -0.5 to 0.7. Typically, areas with low NDVI values represent built-up areas and bare soil, while vegetated areas have moderate values. According to the boxplot, districts 6 (c) and 10 (d) exhibited the highest median NDVI, while districts 11 (e) and 1 (b) had lower median NDVI values.
Visualizing Temperature’s Data using Altair Python library
Land Surface Temperature (LST)
The data was collected and exported from Google Earth Engine (GEE) before being visualized using the Altair Python library. Interestingly, the maximum temperature between 2020 and 2021, which coincided with the COVID-19 pandemic, was lower than in previous years.
Air Temperature
The data was gathered from free air data source and was visualized using Altair Python library. For more informations please visit : https://www.meteoblue.com/
LST, NDVI, and Housing Price Distributions across 22 Tehran Districts
This plot visualizes the distribution rank across 22 Tehran districts for three variables: Land Surface Temperature (LST), Normalized Difference Vegetation Index (NDVI), and Housing Price.
Network Analyse using Python
Network Closeness using Momepy Python library
Closeness centrality is a concept in graph theory that measures a node’s average distance to all other nodes in the network. In this context, nodes likely represent street intersections, and edges represent connecting streets. A higher closeness centrality value indicates that an intersection is, on average, closer to other intersections.
Network Edge Bearing using OSMnx Python library
Street edge bearing was used to classify urban patterns in Mashhad. This chart visualizes the distribution of street orientations across all districts in the city.
Network indexes
This figure depicts network indices calculated using SpaceSyntax for the Sareban neighborhood in Bojnourd city, Iran.
Traffic flow modeling using AnyLogic
During my undergraduate studies, I created a basic model of traffic lights’ cotrol and operating traffic flow in a street using AnyLogic: https://drive.google.com/file/d/1AB_y-NpQ4RkZrg37bEdhMbL1L51FP2vX/view?usp=sharing
spatial analyse using Image processing
NDVI Distribution (OpenCV Python library)
The distribution of vegetation in a city cannot be reliably calculated solely based on the black and white range of an image. While darker areas in a black and white image might suggest more vegetation, other factors like soil type and shadows can influence the image.
Shade distribution (OpenCV Python library)
The distribution of shade and bright colors in Yazd’s vernacular architecture is surprising. This is because these traditional buildings are designed to create distributed shade, resulting in a relatively even balance of shade and sunlight throughout neighborhoods.
Covid-19
This map depicts the spread of the COVID-19 virus across Iran. It likely originated in Qom province before reaching the capital city, Tehran. The map was created using geospatial libraries GeoPandas and Matplotlib.
Spatial Analyse using Software
Site Analysis
This figure analyzes various urban design indexes for the Sareban neighborhood in Bojnourd city, Iran. ArcGIS was likely used for spatial analysis, while SketchUp and Photoshop are better suited for 3D modeling and image editing, respectively.
Site Selection using ArcGIS
This figure presents a series of maps created to identify the most suitable location for a new city near Bojnourd, Iran.
Work expreineces
Online teaching
I have an online teaching session. You can find the details here: https://girs.ir/author/kimiachenary/
House Design
