explanatory markdown files

.github/workflows/ci-checks.yml

@@ -18,6 +18,8 @@ jobs:
         files: .
         config_file: .markdownlint.yml
 
+
+
   py_formatting:
     runs-on: ubuntu-latest
     steps:

.vscode/settings.json

@@ -122,5 +122,6 @@
       "source.fixAll.ruff": "explicit",
       "source.organizeImports.ruff": "explicit"
     }
-  }
+  },
+  "cSpell.words": ["NFDB"]
 }

.vscode/spellright.dict

@@ -0,0 +1,5 @@
+shx
+prj
+Jupyter
+shapefile
+shp

0_domain_study/README.md

@@ -1 +1,147 @@
 # Domain Research
+
+## Research Question Table
+
+### Research Question 1 (Alona)
+
+**Research Question:**  
+How are the frequency and size of wildfires in North America related  
+to health problems caused by air pollution?
+
+**Research Novelty:**  
+Not really new  
+<https://pmc.ncbi.nlm.nih.gov/articles/PMC3492003/>  
+<https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2828029>
+
+**Expected Data Availability:**  
+Quite available
+
+**Problems:**  
+Although there is a lot of data, the topic is not really new and  
+doesn't seem very promising
+
+**Sources or Brief Explanation:**  
+Studies on PM2.5 in the US  
+<https://cwfis.cfs.nrcan.gc.ca/>  
+<https://www.canada.ca/en/health-canada/services/air-quality.html>  
+<https://ourworldindata.org/wildfires>
+
+---
+
+### Research Question 2 (Alona)
+
+**Research Question:**  
+How are income level and the source of antibiotic prescriptions  
+related in Ukraine?
+
+**Research Novelty:**  
+Seems to be quite new  
+<https://www.sciencedirect.com/science/article/abs/pii/S0196655316310896>  
+<https://academic.oup.com/jacamr/article/6/2/dlae041/7633112?login=false>
+
+**Expected Data Availability:**  
+We might get some data from government statistics or surveys,  
+but haven't checked yet
+
+**Problems:**  
+Prescription and purchase data may be incomplete or unofficial
+
+**Sources or Brief Explanation:**  
+WHO and UNICEF reports
+
+---
+
+### Research Question 3 (Alona)
+
+**Research Question:**  
+How does the intensity of shelling in residential areas of Ukraine  
+affect access to and quality of online education?
+
+**Research Novelty:**  
+Very high
+
+**Expected Data Availability:**  
+Damage data:  
+[ACLED report on Russian targeting in Ukraine](https://acleddata.com/2025/02/21/bombing-into-submission-russian-targeting-of-civilians-and-infrastructure-in-ukraine/)
+
+Education data: Kolibri, Coursera for Refugees, UNICEF
+
+**Problems:**  
+Educational data is tricky to get. We should ask some big platforms  
+and wait a long time. They don’t usually specify regions in Ukraine.
+
+**Sources or Brief Explanation:**  
+UNICEF and UN reports
+
+---
+
+### Research Question 4 (Alona)
+
+**Research Question:**  
+How does the availability of a well-developed public transportation  
+system influence processed food consumption in car-free households in Canada?
+
+**Research Novelty:**  
+High – haven’t found an exact match in literature
+
+**Expected Data Availability:**  
+Not sure
+
+**Problems:**  
+Very area-specific. Requires filtering by income level, car ownership,  
+and food type
+
+**Sources or Brief Explanation:**  
+_(Not specified)_
+
+---
+
+### Research Question 5 (Alona)
+
+**Research Question:**  
+How does non-prescription antibiotic use affect the spread of  
+antibiotic-resistant infections in Ukraine?
+
+**Research Novelty:**  
+Haven’t seen studies on this for Ukraine
+
+**Expected Data Availability:**  
+Some lab data from universities or public health sources.  
+Survey/pharmacy data limited
+
+**Problems:**  
+Black market access to antibiotics; unclear where we can  
+get microbiological data
+
+**Sources or Brief Explanation:**  
+WHO AMR reports
+
+### Research Question 6 (Zaid)
+
+**Research Question:**  
+How effective is Satellite Image Analysis at Predicting Deforestation in Malaysia?
+
+**Research Novelty:**  
+High – not much research on this topic in Malaysia.
+Overall, satellite image analysis is very well-studied, but in the context
+of Predicting deforestation, we could only find very few papers.
+When it comes to predicting deforestation in Malaysia, we found no papers.
+
+**Expected Data Availability:**  
+Highly available.
+Sattelite images are available from many sources, and we are using the
+google earth engine to access them.
+The vegetation index data is also available from Global Forest Watch.
+Other data such as deforestation rates and forest cover are easily
+accessible from government and environmental organizations.
+
+**Problems:**  
+Satellite image analysis can be computationally intensive.
+Satellite images are HUGE. They requires lots of storage and time to work with.
+
+**Sources or Brief Explanation:**  
+
+- [Global Forest Watch](https://www.globalforestwatch.org/)
+- [Google Earth Engine](https://earthengine.google.com/)
+- We hope to use the vegetation index and satellite images to train a model
+  to predict deforestation in Malaysia.

0_domain_study/guide.md

@@ -1,15 +1,69 @@
 # Domain Study: Guide
 
-To do meaningful research in a domain, you need to learn what others already do
-and don't understand in this area. Use this folder to organize your group's
-understanding of your research domain including: your own summaries, helpful
-PDFs, links you found helpful, ...
+To do meaningful research, you must know what others do and don’t understand.  
+Use this folder to collect your group’s knowledge of your research domain.  
+Include your own summaries, PDFs, and useful links.
 
-This folder is different from `/notes` because it contains _only_ information
-about your research domain. When deciding what goes here, ask yourself this
-question: _Would someone need to know this to understand our research?_
+This folder is not like `/notes` — it contains _only_ research domain info.  
+Ask yourself: _Would someone need this to understand our research?_
 
 ## README.md
 
-Use this folder's README to document all the notes and resources in this folder.
-Someone shouldn't need to read through _everything_ to find what they need.
+Use this folder’s README to describe the notes and files inside.  
+Someone should not need to read everything to find key info.
+
+## Chosen Question
+
+How effective is Satellite Image Analysis at Predicting Deforestation  
+in Malaysia?
+
+### Why this question?
+
+Malaysia has vast forests, but deforestation is a serious issue.  
+Global Forest Watch shows a 31% drop in tree cover since 2000.  
+It is important to track and predict deforestation early.  
+If satellite image analysis works, it can support conservation.  
+It could also be a low-cost tool for environmental protection.
+
+### Satellite Image Analysis
+
+This means using satellite images to study Earth’s surface.  
+We can analyze land use, vegetation, and environmental changes.
+
+Image resolution matters. High-res images give better results  
+but need more storage and processing time.
+
+We plan to use free Landsat images from Google Earth Engine.  
+They have 30m resolution and are updated every 16 days.  
+This offers a balance between quality and efficiency.
+
+There are 3m daily images from PlanetScope, but they are costly.  
+They also need too many resources for our timeframe.  
+They may help future studies track trees at an individual level.
+
+### Rough Plan
+
+- **Data Collection**:  
+  - Get satellite images and vegetation data from GFW and Earth Engine.  
+  - GFW makes it easy to find vegetation index data.  
+  - Google Earth Engine provides free satellite images.  
+    We need to create accounts and request API access.  
+    We also need to write code to extract and process images.  
+  - Gather other data from public sources and institutions.
+
+- **Model Training**:  
+  - Use k-means clustering with 5 time-based clusters.  
+  - Combine vegetation index and image data to detect forest loss.  
+  - Add more features as the project progresses.
+
+- **Model Evaluation**:  
+  - Check how well the model predicts real deforestation.  
+  - Compare predicted and actual results to assess performance.
+
+### Resources
+
+- [Global Forest Watch](https://www.globalforestwatch.org/)  
+- [Google Earth Engine](https://earthengine.google.com/)  
+- [Google Earth Engine API Docs](https://developers.google.com/earth-engine)  
+- [Interesting Paper](https://arxiv.org/pdf/1803.02489)  
+- More resources will be added as we find them.

0_domain_study/related_research_topics/forests.md

@@ -0,0 +1,102 @@
+# Forests
+
+- Earth surface is divided into large geographic areas called **Biomes**.  
+
+- Each biome is characterized by distinct climate, vegetation, and animal life.
+
+![alt text](image.png)
+
+![alt text](image_1.png)
+
+- **Forests** are a biome where treas are the dominant life-form.
+
+- There are three types of forests:  
+
+1. **Boreal forests:** also known as taiga,
+ are found in the subarctic regions and are
+ characterized by cold temperatures and coniferous trees.
+
+![alt text](image_2.png)
+
+2.**Temperate forests:** located in mid-latitudes, experience four distinct
+seasons and can be either deciduous (trees that lose their leaves seasonally) or
+coniferous.
+
+![alt text](image_3.png)
+
+3.**Tropical forests:** near the equator,
+ are known for their warm temperatures, high
+ rainfall, and exceptional biodiversity.
+
+![alt text](image_4.png)
+
+- Forests can be **dry** or **humid**
+- Forests can be:
+
+1. **Primary** or old growth:
+
+- are those that have not been significantly disturbed by human activity.
+
+2.**Secondary**:
+
+- are areas where forests have regrown
+ after being cleared or disturbed by human activities or natural events.
+
+![alt text](image_6.png)
+
+## Some relevant terms
+
+**Tree cover:**  
+
+- Encompasses all vegetation taller than 5
+meters, including natural forests and tree crops (like rubber, oil palm, etc.).
+- May be expressed as a percentage of the area covered by tree canopies, or in
+absolute terms (e.g., square kilometers).
+
+**Tree canopy:**  
+
+- Refers to the layer of branches and leaves formed by trees, effectively
+creating a "roof" over the ground.
+
+**Tree canopy density, canopy cover or crown cover**:  
+
+- is the proportion of ground area covered
+by this tree canopy, usually expressed as a percentage.
+- It essentially measures how much of the
+ground is shaded or covered by tree foliage when viewed from above.
+
+![alt text](image_7.png)
+
+**Tree cover loss:**
+
+- removal or mortality of that tree cover, regardless of the cause.
+Can be :
+
+1. human caused: eg. Deforestation.  
+2. due to natural causes: eg. Wildfires
+
+**Some measurement units:**  
+
+- Hectare(ha) : 10,000 square meters.
+- Kha : Kilo hectares.  
+- Mha : million hectares.  
+- Gha : Global hectares.
+
+- A global hectare represents the biologically productive area (cropland,
+grazing land, forest, fishing grounds, and built-up land) needed to provide the
+resources consumed and absorb the waste generated by a population, but with the
+productivity of each land type adjusted to the world average for that year.
+- **Ecological Footprint:**  
+Global hectares are the primary unit used to express an ecological footprint, which
+is a measure of human demand on the Earth's ecosystems.
+- **Example:**  
+If a country's ecological footprint is 4.5
+gha per person, it means, on average, that
+each person in that country requires the
+equivalent of 4.5 hectares of biologically
+productive land and water, with the average
+world productivity, to provide for their resource needs and waste absorption.
+
+![alt text](image_8.png)
+
+## Forests in Canada