Geog0027 Coursework (2020-2021)

Course Tutors

Prof. P. Lewis

Prof. M. Disney

Dr Qingling Wu

Department of Geography

University College London

Welcome to GEOG0027 Coursework documentation

Contents:

• 1. Introduction
  • 1.1. Introduction
  • 1.2. Purpose of the practical
  • 1.3. Overview of task
  • 1.4. Summary
    • 1.4.1. Data processing
    • 1.4.2. Modelling
  • 1.5. Details
  • 1.6. Further Reading
• 2. ENVI Setup
  • 2.1. Introduction
  • 2.2. Desktop@UCL
  • 2.3. Stand-alone installation of ENVI
    • 2.3.1. Downloading and Installing ENVI
  • 2.4. Set up a UCL VPN connection
    • 2.4.1. For Windows:
    • 2.4.2. For Mac:
  • 2.5. Using ENVI for GEOG0027 Classification (Skip this step if you wish to use python-based GEEMAP)
    • 2.5.1. For Windows
    • 2.5.2. For Linux (Unix) or Mac
  • 2.6. Summary
• 3. Modelling Setup
  • 3.1. Setup options
    • 3.1.1. Using R or Rstudio
    • 3.1.2. Which option should I choose?
    • 3.1.3. Installing R and Rstudio on your own computer
    • 3.1.4. Installing data packages
    • 3.1.5. Option 1
    • 3.1.6. Option 2
    • 3.1.7. Option 3
    • 3.1.8. Option 4
    • 3.1.9. Option 5
  • 3.2. Running Rstudio
    • 3.2.1. Select the .Rproj file
    • 3.2.2. Set working directory
    • 3.2.3. Required files
    • 3.2.4. Load and run UrbanModelData.R
    • 3.2.5. Save the R file regularly
    • 3.2.6. Binder care
    • 3.2.7. Write a report
• 4. Google Download
  • 4.1. Download
  • 4.2. Getting the data archive
  • 4.3. Visualise time series
  • 4.4. Visualise spatial data
• 5. Classification with ENVI
  • 5.1. Data processing summary of requirements
    • 5.1.1. Part 2a: Data Exploration and Classification Theory
    • 5.1.2. Part 2b: Time Series Classification
  • 5.2. Extraction of land use extent
  • 5.3. Obtaining Landsat data
  • 5.4. Classifying the data: manual method
  • 5.5. Two ways to batch process large datasets
  • 5.6. Automated processing with ENVI IDL
  • 5.7. Loading the automation software
  • 5.8. Pixel counting
  • 5.9. Summary
    • 5.9.1. Part 2a: Data Exploration and Classification Theory
    • 5.9.2. Part 2b: Time Series Classification
• 6. Classification with GEE
  • 6.1. Jupyter setup
  • 6.2. First map
  • 6.3. Examining the time series
  • 6.4. Select ‘Shenzhen’ as the area of interest (AOI)
  • 6.5. Load Landsat data collections from GEE
  • 6.6. K-means unsupervised classification with GEE
  • 6.7. Link unsupervised clusters into Land Use Land Cover (LULC) classes
  • 6.8. Cluster areas
  • 6.9. Repeat for multiple years
  • 6.10. Export results to excel (.csv) files
• 7. Modelling
  • 7.1. Summary of requirements
  • 7.2. The model
    • 7.2.1. Pre-requisites
    • 7.2.2. The Seto model
  • 7.3. Software
  • 7.4. Download data files
  • 7.5. Using Rstudio
  • 7.6. Using R
  • 7.7. Modelling in R
  • 7.8. Experimentation
    • 7.8.1. The sort of questions you should ask of the data / results
    • 7.8.2. Additonal experiments
  • 7.9. Summary
• 8. Project Write up
  • 8.1. Assessment
  • 8.2. Requirements
    • 8.2.1. Part 1: Introduction (20%)
    • 8.2.2. Part 2: Data processing (60%)
    • 8.2.3. Part 3: Modelling (15%)
    • 8.2.4. Part 4: Discussion and Conclusions (5%)
  • 8.3. Write up
• 9. Project Advice
  • 9.1. Selecting Landsat scenes
  • 9.2. (i) Data Gaps
  • 9.3. Range of years to use
  • 9.4. Data for Modelling

Indices and tables

• Index

• Module Index

• Search Page