Selecting the best Coral Sea imagery (Coral Sea Mapping project)

In this video we discuss the process for selecting the best Sentinel 2 imagery for processing into cloud free composite images of reefs. This video details setting up scripts in Google Earth Engine for processing Sentinel 2 imagery in your own account. This uses the code available from the draft Coral Sea Sentinel 2 image repository (https://github.com/eatlas/CS_AIMS_Sentinel-2-marine_V0). 

Video times:
0:50 - Creating a new code repository in Google Earth Engine and copy over the source scripts from GitHub.
2:31 - Overview of sunglint removal from the images and its limitations. 
5:26 - Dealing with partial scenes where Sentinel 2 swaths don't align well with the image tiles.
6:15 - Classifying images based on cloud cover and sun glint
9:47 - Discussion of various and limitations of image styles (DeepFalse)
11:40 - Limitations of cloud masking and how many images are needed to make good image composites
18:10 - Setting up 02-view-selected-sentinel2-images.js and 03-create-composite-Coral-Sea.js scripts in Google Earth Engine from the GitHub repository.
24:05 - Using 03-create-composite-Coral-Sea.js
26:40 - Choosing the best images to go into the composites.
29:05 - Getting a composite to render to the display and to export to Google Drive
34:00 - Adding additional styles for the composite image to be rendered in (DeepFalse, DeepMarine, ReefTop, Shallow, TrueColour).
37:05 - Exporting the composite image to Google Drive.
41:20 - Deliverables for the final composite 
42:00 - Questions - Running the code in our own GEE accounts should much things up.

This video is a recording of a team meeting of the Coral Sea Mapping project (https://eatlas.org.au/projects-other/coral-sea-reef-mapping) run by the Australian Institute of Marine Science. This project aims to digitise and map the marine features of the Coral Sea including reef boundaries, atoll boundaries, cays, vegetation and beach rock. 
Video recorded: 02 Dec 2021