Change course outline - Githubissues

salvafern commented 2 years ago

1. Tell a story: @lauramarque

Someone collect observations of sharks/cetaceans/sea turtle and wants to fit a species distribution model
- Do you have enough data? Where can I find more?
- From what region of the world?
- How do you make sure you are using the same species name than your peers?
- How do I link with environmental data? Where can I find this information?
"In this workshop, we will answer all these questions" "and apply this to your own research!" "you will have an analysis-ready dataset"

2. Show what we will do in the workshop with a visual aid: Connection of the different projects. @whaleshark99

Explain quickly the connections but we will go in detail about every project

3. Exercises: Content: @salvafern Outline: @whaleshark99

Exercises 3.1. Standardize your data: Get lifewatch data and standardize to MR and WoRMS rethink the wording of standardize

Explain the 3 projects: show MRGID and AphiaID @lauramarque
Show code examples
"This can be applied to your own dataset!"
Input: LifeWatch occurence data
R packages: lwdataexplorer, worrms, mregions2
Output: dataset

Exercise 3.2. Get EurOBIS data to complete your dataset

Input: dataset from ex. 1
R package: eurobis, explain what is EurOBIS @lauramarque
Wrangle lifewatch and eurobis data to be in the same data.frame: species, lon, lat, time
recommend use of darwin core terms

3.3. Get seabed habitats and bio-oracle data

Input: none
R packages: (emodnetWFS (seabed habitats)), sdmpredictors (Temp, pH, salinity, ...)
Output: dataset with environmental info

3.4. Link occurrences dataset to environmental data and get an analysis ready dataset.

Input: datasets from exercises 2 & 3
Output: combined dataset

lottepohl commented 2 years ago

Just to get it clear again:

1. Tell a story:

Someone collect observations of sharks/cetaceans/sea turtle and wants to fit a species distribution model

We will provide example observations of sharks/cetaceans/sea turtle that people can work with that don't have their own dataset, right?

"In this workshop, we will answer all these questions" "and apply this to your own research!" "you will have an analysis-ready dataset"

And people that indeed have own data that they want to work on can do so.

lottepohl commented 2 years ago

Here are draft versions of visual aids.

1. Overview on the databases and products overview_databases_whitebg

2. Full pipeline of the workshop pipeline_whitebg The pipeline could be sliced into pieces for each exercise, of course.

Feedback welcome :)

salvafern commented 2 years ago

Hey looks great!

Some comments:

The data we will get from EMODnet Seabed Habitats won't be raster but vector (polygons). The EMODnetWFS package does only retrieve vector. We are working on the raster bit (WCS) that will be ready later this year. See this link
The EurOBIS occurrences will be actually appended below the lifewatch data :) not right-joined.
Don't use species but taxon.

@LennertSchepers any extra comments?

thanks!!

lottepohl commented 2 years ago

@salvafern noted :)

next versions below: overview_databases_whitebg_2

pipeline_whitebg_2

salvafern commented 2 years ago

Maybe I prefer the jumping guy after all 😅

Besides that all good for me!

lottepohl commented 2 years ago

Well, we can also just leave it out completely :D Or maybe we come up with another idea in the coming days, it's not the most relevant thing on the visual...

salvafern commented 2 years ago

Yes good for me

salvafern commented 2 years ago

updated for now https://vlizbe.github.io/ebr-2022-data-combine

lifewatch / ebr-2022-data-combine

Change course outline #2