Choosing questions and aims

[johann-wagner]

You must choose questions and/or aims that are:

• likely to be things you can investigate using the dataset(s) you have chosen
• go further than merely summarising the dataset(s) you have chosen
• are different from things we have covered directly in the course so far
• can be realistically addressed within the scope of the final project (you are expected to put around 30 hours of work into this project - 10 each from weeks 9, 10, and 11)

Within those constraints, please follow your interests and be creative.

In your report you must first state and then justify your question(s)/aim(s). The justification must include background research on what has been done before in this area.

Here are a couple of intentionally generic examples. I offer these to help you start thinking. But I want you to be creative - you don’t need to have questions/aims that look like this!

Example 1: testing for correlations

Questions: Is trait x correlated with traits a, b, and c? Are these correlations similar across taxa?

Justification: Many studies have suggested that trait x should correlate with traits a, b, and c because [briefly explain, with references]. These correlations haven’t been examined in clades y and z [e.g. Eucalyptus, Acacia; provide and breifly describe references to prove this], but database q [e.g. AusTraits] allows us to test this.

Example 2: building something useful

Aim: to build a shiny app / R package for task x (e.g. data cleaning) o dataset z

Justification: Task x is crucial for Exploratory Data Analysis because… [explain why with references], but is challenging for dataset z because [provide reasons]. I will write a shiny app / R package to facilitate task x on dataset z. Similar work has been done in packages a, b, and c [provide references and briefly describe how your work will be different and/or build on that work].

Examples 3-∞

Do not let the above examples constrain your thinking here. Read a few papers and/or blog posts, or just dig around in some datasets and see what looks interesting to you. If you’re not sure if your questions/aims are appropriate, just ask one of the teaching team and we’ll help you figure something out.

[johann-wagner]

I'm going to focus on invasive animal species in the ACT, specifically rabbits and foxes using The Atlas of Living Australia data.

Technical features:

• RShiny Dashboard
• Spatial visualisation

[johann-wagner]

Potential Questions:

Does rabbit observations change once a biocontrol program is released?
Create a time series of rabbit observations across time and see if there are any changes after these years:

• myxomatosis (an infectious rabbit disease), released in 1950
• European rabbit fleas (as a vector of myxomatosis), released in 1968
• rabbit haemorrhagic disease, released in 1995.

Resources on rabbit biocontrol programs:

• https://theconversation.com/dont-underestimate-rabbits-these-powerful-pests-threaten-more-native-wildlife-than-cats-or-foxes-168288
• https://education.nationalgeographic.org/resource/how-european-rabbits-took-over-australia/

Are fox and rabbit observations negatively correlated?

• Population dynamics, diet and aspects of the biology of feral cats and foxes in arid South Australia
• The short-term response of feral cats to rabbit population decline: Are alternative native prey more at risk?

What is the relationship between rabbit observations and weather patterns?

• Are there more rabbits when there is wetter weather as there is increased plant growth/food for rabbits?

Create an RShiny Dashboard to produce the above visualisations for several Australian invasive animal species for the different states and territories:

• Statistics / General Info on selected invasive animal species (let this change based on what species have been checked in checkbox) (Let's calculate number of observations in ATLA, type of animal, other general info)
• Frequency table of selected invasive animal species by state and territory
• Time series of annual observations filter by state and territory (dropdown) and by several Australian invasive animal species (checkbox)
• Spatial visualisation filter by state and territory and by several Australian invasive animal species
• Maybe create a download dataset button

This would be essentially an RShiny Dashboard that creates visualisations and statistics for Australia's top 10 invasive animal species using The Atlas of Living Australia data.

[johann-wagner]

Final Idea:

Research Question:

• Develop an RShiny app that showcases the spatial and temporal (monthly) occurrence of the top 5-10 invasive animal species in Australia by state/territory and by regionality (urban vs. regional)
• Visualise and calculate the proportion of the selected invasive animal species observed in an urban area
• Create a download option for users to download the cleaned and tidyed dataset for the selected invasive animal species for a particular state/territory

Dynamic options:

• Dropdown of invasive animal species
• Dropdown option for state/territory

Let the visualisation and the statistics vary by invasive animal species

Visualisations:

• Spatial leaflet map of data points with urban overlay (change based on state/territory)
• Circular monthly time series of observations over the year

Statistics:

• Proportion of invasive animal species in urban (change based on state/territory)

Create a download dataset option

• Download the dataset for the selected invasive animal species for a specific state/territory

Optional:

• Add a text box space for a variety of facts about the invasive animal (insights from the visualisations & background research)
• Species count by source (citizen science, professional survey)
• Species count over time (Note: it is likely the count simply increases every year, due to more observations from users, rather than a change in actual species populations)

[johann-wagner]

Whiteboarding the RShiny Dashboard design

[johann-wagner]

Idea inspiration:

• RShiny Dashboard layout
• https://labs.ala.org.au/posts/2023-04-03_highlighted-time-series/post.html
• https://labs.ala.org.au/posts/2023-05-16_dingoes/post.html

[johann-wagner]

TO DO:

• In your report you must first state and then justify your question(s)/aim(s).
• The justification must include background research on what has been done before in this area.

[DS4B-ANU]

Good stuff, looking forward to the final RShiny app!

Fred