Structure

To mantain the repository organized and make it easier for other people to understand it, I tend to keep the following simple structure:

• scripts are placed in code.

• datasets in data

• documentation in docs

Sometimes, it is convenient to have sub-directories (folders) for data. For instance, we could have data/raw storing raw data, and data/processed storing the datasets we already pre-processed.

Goals

• Familiarize ourselves with GitHub

  • Always use git pull before you start working;
  • git pull --rebase may be helpful on solving conflicts.
  • Note: I put .Rproj files on the .gitignore. Thus, if you use Rstudio projects, you will need to create your own project locally. That should not affect how things are working.

• Coding style: we should try to follow coding conventions to make it easier for others to read each others code. I know not all people are "tidyverse" enthusiasts, however, their style guide is still quite useful. Check this link.

• Discuss the repository structure. Some of the options are:

  • Working on separate scripts (pros: avoids conflicts; cons: it makes it harder to organize the repo.)

R as a GIS

In the file code/read-dt.R, I wrote a simple code to read the data we will work with. Next, I used the sf package to transform the data.frame into a spatial object.

The tasks I planned are:

1. Understand what is going on in the code/read-dt.R file.
2. What are the lines 17 to 22 doing?
3. What is a Coordinate Reference System (CRS)?
4. Trying to use maps and descriptive statistics to understand the data and come up with some interesting research questions.
5. Other "spatial" packages to check:
   • Mostly for raster data: terra, raster;
   • Interactive maps: tmap, mapview, leaflet,
   • Storing data efficiently: geoarrow

New tasks

1. Practice git and GitHub
2. Length of cruise survey and sights
3. data processing

P.s.: I had to guess the CRS for this dataset. The epsg:4326 is usually a good guess.

Data info

From: https://oceaninformatics.ucsd.edu/datazoo/catalogs/ccelter/datasets/262

General Effort Protocols

Observers are always on effort when ship moving at 5+ knots and weather permits (on all transits, both on and off transect). There should always be 2 observers present. Observations should begin as soon as the ship has left station. Observer on port side scans from bow to 90 degrees left, from ship to horizon. Observer on starboard side scans from bow to 90 degrees right, from ship to horizon. Binoculars should be used for scanning approximately half of the time to scan from ship to horizon. Observations occur until Beaufort 6, rain, or fog reduces visibility to 0.5 nmi or less. While off effort due to weather, at least one observer watches for changes in weather and opportunistic sightings.

Sightings Protocols

Sighting numbers are cumulative throughout the cruise. Sightings are considered on-effort when: 2 observers are present; Beaufort sea state 5 or less, visibility greater than 0.5 nmi; Ship is traveling along prescribed CalCOFI trackline; Ship speed is greater than or equal to 5 knots. Species ID may be confirmed via Big Eyes. Photos of sightings may be taken to verify species ID, though photo effort should not interfere with visual effort.

The species codes are available in the following link

https://oceaninformatics.ucsd.edu/datazoo/catalogs/ccelter/datasets/255