ENVS563 – Geographic Data Science 563.1 Programmed Map (40%)

联系我们： 手动添加方式: 微信>添加朋友>企业微信联系人>13262280223 或者 QQ: 1483266981

ENVS563.1 SUMMER RESIT ASSESSMENT 202526
ENVS563 – Geographic Data Science
563.1 Programmed Map (40%)
Mapping Population Density and Change in Colombia
For this assignment you will use population raster data. You will assign raster values to regions of Colombia (national and departmental level) to map average population density. You will then create a map of the population change between 2015 and 2025. Hints are included to help in case you get stuck. This assignment will be evaluated on technical data processing, map design abilities (assemblage), and design overall narrative.
Software: You can carry out the assessment using R or Python. Remember that you are required to do both assignments for this course in the same coding language.
Evaluation
The assignment will be evaluated based on three main pillars, on which you will have to be successful to achieve a good mark:
1. Data Processing: Your proficiency in handling and manipulating data will be a fundamental aspect of the assessment.
2. Map assemblage: This includes your ability to master technologies that allow you to create a compelling map.
3. Design and narrative: Your success in designing an appealing map with a compelling narrative will play a pivotal role in your overall evaluation.
Data
Gridded Population of the World from WorldPop (1km resolution). You need to download 2015 and 2025 Population Count layers for Colombia. Note that each dataset has related metadata.
National boundary of Colombia (Level 0) and Departmental boundaries of Colombia (Level 1). These can be downloaded from GADM (Colombia, levels 0–1), Colombia’s national statistics agency, or the Humanitarian Data Exchange (HDX), which hosts OCHA-verified Colombia administrative boundaries.
1. Download the data
1.1. Use the Gridded Population of the World from WorldPop. You will need the Population Count in GeoTIFF format for the years 2015 and 2025.
1.2. You will also need the national and departmental administrative boundaries of Colombia, which you can download from GADM, DANE, or HDX.
2. Import and Clip
2.1. Import and explore the data by plotting it.
2.2. Clip your density rasters for 2015 and 2025 to keep data just for Colombia.
ENVS563.1 SUMMER RESIT ASSESSMENT 202526
2.3.
Hint: It is easier to first clip to extent and then mask pixels outside the polygons.
3. Zonal statistics – population density
3.1. Calculate the average population density for each department (regional level 1) of Colombia for the year 2025. The national boundary (level 0) is just for illustration purposes. If you prefer to focus on a subset of departments rather than all 33, for example the Caribbean Region: this is an alternative — La Guajira, Cesar, Magdalena, Atlántico, Bolívar, Sucre, and Córdoba. THIS DOES NOT MEAN YOU NEED TO FOCUS ON THESE DEPARTMENTS; you could choose others (e.g., the Andean Region, the Pacific Region, or the Amazon Region).
Hint R: You will need to use either the extract() function from the raster package or the exact_extract() function from the package exactextractr. Check the options, for example for extract() the options fun=, na.rm=, sp.
Hint Python: Use zonal_stats() from rasterstats to extract the population density values for each region. Make sure to handle no-data values appropriately.
3.2. Map 1: Your task is to create a map of population density by department (level 1). You should include this map in your assessment submission.
4. Categorical Change in Colombia’s population
4.1. Your task is to create one thematic map of the categorical changes in population in Colombia. You need to compute the difference between the raster population layers and create another raster where each pixel represents the change in population.
Hint R: The overlay function of the raster package allows you to calculate the difference between two rasters (R).overlay_raster <- overlay(popraster2015, popraster2025, fun = function(x, y) y – x)Hint Python: In Python, you can calculate the difference between two rasters using the rasterio and numpy modules.
4.2. Visualise your new raster, which could be called population_change_2015_2025 or similar. Create a map with 4 discrete categories: Decline, Neutral, Growth, and High Growth (Map 2). You should include this map in your assessment submission.
Hint R: The function summary() allows you to calculate the summary statistics for your raster; the function reclassify() classifies groups of values to other values, e.g., -100 to 0 becomes 1; 0 to 200 becomes 2, etc.
Hint Python: In Python you can reclassify groups of values using numpy's logical operations or numpy.digitize.
Submit
An HTML version of a .qmd document with R integrated code, OR an HTML version of a Jupyter Notebook with Python code.
In this document there will be:
Your annotated code.
Maps. This should include two maps, Map 1 and Map 2, which can be combined as one final output.
500 words which should address the following questions: 1. What value is associated with each pixel
2. Which CRS are you using Justify your answer.
3. What is Map 1 saying about population density in Colombia
4. Which categorical divisions (Decline, Neutral, Growth, and High Growth) did you choose for Map 2, and why
* If you are working in R, to make sure your html is self-contained, please add the following in the preamble of the qmd file.
—
title: "Assignment Title"author: "Author Name"format:html:self-contained: truetoc: true
—
Submit online via Canvas.
If you experience any difficulty in submitting your assessment, please contact the Student Experience Team (envsci@liverpool.ac.uk) as soon as possible and in advance of the deadline for further advice and support.

ENVS563 – Geographic Data Science
563.1 Programmed Map (40%)最先出现在KJESSAY历史案例。