Term projects are worth 25% of your course grade, and have the appearance of labs though larger, and with less direction as to the speific implementation.
Important dates
Wednesday November 15th, Project Proposals
December 5th Noon, send presentation pptx to Matt
December 6th Project Presentations
December 11th Project Report Due (Hard deadline no extensions)
Grading Rubic
Project Proposal | 3%
Jupyter Notebook | 15%
Presentation | 3%
Final Report | 4%
Project Overview
The projects provide an oportunity to demonstrate coding ability, scientific communication, and answer an environmental question. There is two primary project streams.
A) Large Scale remote sensing question: For this project option you should choose a question that covers a large geographic area, and includes remote sensing data. A speific option for this would be to investigate the correlation between NDVI and Temperature / Rainfall. To answer this question you could pick 3 regions (Data has good availability in Canada for environmental factors, talk to Matt for more details). For each of these regions download the Rainfall and Temperature data as tabular data. In your notebook use python to agregate this data. Then in your notebook get some remote sensing data you can calculate NDVI from, determine NDVI for each reagon for each time period, and Calculate the averages. Finally make some plots demonstrating the correlations between rainfall, temperature and NDVI.
B) Use interactive notebooks to allow an end user without GIS experiance to explore a dataset. This project option will be more technically challenging to implement, however complexity of expected analysis will be dramatically reduced. A speific example here could be to make a cencus data explorer. Using Ipython Widgets make a notbook that allows users to examine the canada census data, such a project would see the creation of a leaflet map that would display the cencus data, allowing users to select how granular to view the data (Province, Division, or Sub division) and should have multiple styling options (ie pick different fields and compare year to year, or in a single year look at correlations between fields).
C) Choose your own adventure, propose an idea of a problem you would like to solve to Matt, to varify complexity of project is within reason.
Project Components
Proposal
Your proposal document should be 1 – 2 pages in length, and will be graded for its writing and communication. The required parts to this document are
- Executive Summary
- Data sets needed (why were they chosen?)
- Proposed Methods
- Expected report formating
- Hypothosis of results (if applicable, project A or C if non interactive)
- Desciption of end user intecation (project B, or project C if interactive)
Jupyter Notebook
This will be the bulk of your course grade and is the code that performs analysis. Code should be commented, and logically laid out (we don’t do science alone, could someone else work with your code if needed?) (Optional lecture on code styling November 24th).
All graphs produced should have titles, logical colour schemes, appropreate units, etc.
Be correct! (Your analysis should not be producing fake data).
Quality of code will have higher expectations that in lab. Once you get your project working, restart the kernel run it from the top make sure it is still working. Go back and clean up code, restart the kernel make sure it still runs, repeat if needed to make code clean.
Due with final report (though will need to be working enough to get results for project presentations).
Presentation
Prepare a presentation for the class.
Each student will provide a 10-15 minute presentation to the class. (Order will be randomly determined by random number generator).
Each presentation will consist of
- Cover slide (Single slide, with participant name, and project title)
- Introduction to project ( ˜1 minute, what question does your project answer, can be longer if background is needed)
- Description of data used, and why it was chosen (˜1 minute)
- Methods (˜2-3 minutes, what steps were taken for analysis)
- Results ( ˜2-3 minutes, this should include matplot libs, or if interactive a demo (can be outside of ppt))
- Future Work (˜1 mintue, what would make this project better if only you had time and resources)
- Lessons Learned (˜2-3 min, are there any skills you gained during the project, concepts that turned out to be important, problems you solved the dard wway before learning the easy way? That can be about GEOG 250 as a whole if you wish)
- Questions (˜1-3 min, leave a placeholder slide for this; answer any questions from classmates or Matt) You are encouraged to ask at least one other student a question at the conclusion of their presentation
Final Report
In terms of expectations, imagine you were hired to perform this analysis, what would you send to your client? This document should be easy to read, well formated, and clearly describe what you did, and what the outcome was. Two options here:
- Make a word document just as you would for any University paper, making sure to embed any plots generated.
- (You must use this option if making an interactive notebook) Use markdown cells inside of your Jupyter notebook such that you get a report that includes all of your code (Will be explained further on 24th). This is a great way to leverage the notebook file format.
Your final report should include
- Introduction
- Data sources
- Methods
- Results
- Conclusion (what are the implications of the results)
- If you have any sources, bibliography (If you take code from a fourm such as stack overflow make sure to site the author of the post you used.
Expectations
Final projects are expected to be completed individually, and you should not be asking other students for project speific help. Working on other students projects would be academic misconduct for both students involved.
The project timeline include 15hr of class time before the presentations, and projects should be anticipated to take 30-50hr to complete (you will not be able to complete homework with scheduled class time alone). As a rought timeing guide
- Proposal under 3hr
- Jupyter Notebook 22-39hr
- Presentation Prep 2-3 hr
- Final Report 3-5hr
Get help when you need it;. There is an expectation that you continue to attend class time to work on projects unless you are ahead of schedule. Matt will be available by e-mail and to book office hours as needed. My schedule is quite flexable for the remainder of the semester and happy to meet as much as needed outside of class time, however out of respect for time it is expected that meetings out of class time be in addition to scheduled lab time, not instead of.