Comprehensive Notes on IMOS, AODN, and Marine Data Observation Systems

Sydney Institute of Marine Science (SIMS) Facility and Course Introduction

The Sydney Institute of Marine Science (SIMS) facility is identified in red from aerial/drone views, while the green highlights indicate the public work areas used for demonstrations.
A specific demonstration area is designated for animal tracking where students utilize sensors to understand data collection and results.
Local amenities include an aquarium and a cafe located behind the main facility.

Learning Outcomes

To understand the diversity of data within the Integrated Marine Observing System (IMOS) and the various instrumentation types used for collection.
Developing the ability to formulate and test scientific hypotheses within a network.
Mastering the management and access of large-scale marine datasets.
Achieving a professional standard in presenting data visually, appropriate for Master's level study.
Evaluating and synthesizing expert opinions from the marine science community.

Course Schedule, Modules, and Software Requirements

Introduction to AODN: Currently being conducted in week one by Laurent, a data scientist at the Australian Ocean Data Network (AODN).
Introduction to R (Next Week): Fundamental knowledge of R and R Studio is required as it will be used in every subsequent module. It is strongly encouraged to download these prior to class.
Animal Tracking: Conducted for two weeks by Dr. Uri Niella.
Physical Oceanography: Conducted for three weeks by Dr. Veronique.
Ecological Modeling: Conducted for two weeks by Dr. Christina.
Benthic Ecology: Conducted by the lead instructor for the final two weeks.
Seminars: Afternoon sessions will feature researchers from various Australian universities discussing the current state of marine science in Australia.

Assessment Details and Academic Integrity

Weighting: The course grade is split into $60\%$ for bi-weekly assignments and $40\%$ for the final examination.
Final Exam: Scheduled for May 21. It is an open-book exam administered via Microsoft Teams.
Exam Structure:
- Section 1: Multiple-choice and short-answer questions covering lectures, seminars, and demonstrations.
- Section 2: Long-answer questions focused on core modules.
Plagiarism Policy: Plagiarism is considered serious misconduct and can lead to course failure. Students must cite all sources, use proper references, and paraphrase in their own words rather than copy-pasting. Online submissions are monitored by plagiarism tracking software.

Introduction to IMOS and AODN (Australian Ocean Data Network)

IMOS (Integrated Marine Observing System): A national research infrastructure project primarily funded by the federal government to create a national observing system.
AODN: Serves as the data center for IMOS, functioning similarly to a library that organizes and provides access to massive data collections.
Key Partners: SIMS, SARDI, the Bureau of Meteorology (BoM), CSIRO, and the University of Tasmania (UTAS).
Spatial Scales: Data collection covers Australian waters at both regional and ocean basin scales.
Main Research Themes:
- Multi-decadal ocean change: Heat and $CO_2$ absorption.
- Climate variability and extremes: Events like La Niña and El Niño.
- Continental shelf processes: Connections between boundary currents and coastal ecosystems.
- Ecosystem responses: Monitoring variability from primary producers to apex predators.
- Major boundary currents: The East Australian Current ( $\text{EAC}$ ), the Leeuwin Current (West), and the Antarctic Circumpolar Current ( $\text{ACC}$ ).

IMOS Observing Facilities and Instrumentation

Deep Water Moorings: Measure variables like $CO_2$ and currents. Often use $ADCP$ (Acoustic Doppler Current Profilers) at the seafloor to measure velocity and direction across water columns via three beams.
Argo Floats: A global program with thousands of floats. These instruments are approximately $2\,m$ long. They sink to $1000\,m$ for nine days, sink further to $2000\,m$ to collect a profile, and then resurface to transmit data via satellite. Standard measurements: Pressure ( $P$ ), Temperature ( $T$ ), and Salinity ( $S$ ). Bio-Argos measure biological properties like $pH$ and nitrates.
Ocean Gliders: Autonomous underwater vehicles ( $AUV$ ) that can be remote-controlled to follow specific courses or cross-sections of currents, unlike drifting Argo floats.
Animal Tagging and Tracking:
- Seals: Equipped with head-mounted instruments using epoxy to record foraging behavior and profile data ( $T, P, S$ ) up to depths of $80\,cm$ or more.
- Acoustic Tagging: Arrays of detectors (e.g., between Mainland Tasmania and Maria Island) pick up pings from tagged marine animals. This data is part of a global project.
Coastal Wave Buoys: Partnership with BoM to measure over 50 different wave parameters.
Ocean Radar: Pairs of poles send radar echoes to determine the state of the sea (height and direction) over a range of approximately $20\,km \times 20\,km$ .
Satellite Remote Sensing: Provides over 30 years of Sea Surface Temperature ( $\text{SST}$ ) data.
- Composites: 1-day, 3-day, or 6-day composites are available. Multi-day composites reduce cloud coverage interference (white areas on maps), which is critical for biology-focused research.
Ships of Opportunity ( $SOOP$ ): Instruments placed on commercial vessels (e.g., Spirit of Tasmania) or fishing nets (e.g., $\text{FishSOOP}$ ) to measure turbidity, salinity, and temperature.

Data Management, Metadata, and Formats

Metadata: Crucial for defining data. It includes Project Metadata (PI, location, time range) and File-level Metadata (instrument serial number, units, variable attributes).
NetCDF Format: A binary format essential for climate and marine science. It is self-describing, containing both metadata and data in one file.
Conventions: Follows the Climate and Forecast ( $CF$ ) convention for standardized variable names (e.g., temperature vs. sea surface temperature).
Quality Control (QC): Data contains QC flags ranging from $1$ (good data) to $9$ (bad data).
Data Modes:
- Near Real-Time: Often automated QC only; includes basic de-spiking.
- Delayed Mode: Expert-controlled and validated by scientists.
Time Standards: Data is always recorded in UTC (Coordinated Universal Time), denoted by a $Z$ (e.g., $2010\text{-}01\text{-}01T00:00:00Z$ ).
Time Conversion: $\text{Local Time} = \text{UTC} + \text{Time Zone Offset}$ .

Future Directions and Cloud Optimization

New Portal: AODN is transitioning to a more efficient portal and cloud-optimized data formats.
Zarr and Parquet: These formats allow "chunking" of data. Users can query specific points in a 30-year gridded dataset without downloading tens of gigabytes of data. This allows for rapid time-series analysis at specific coordinate points.
Open Source: AODN code is hosted on GitHub. Python and Jupyter Notebooks (via Google Colab) can be used to query these cloud-optimized datasets directly.

AODN Portal Workshop and Practical Navigation

Portal Navigation Steps

Step 1: Select data collection (filters for parameters, organizations like IMOS, or platforms).
Step 2: Subset data temporally or geographically using a bounding box or polygon.
Step 3: Download data in formats such as CSV or NetCDF.

Workshop Examples and Highlights

Case Study: Rottnest Channel Swim: Exploring current speeds during the 2014 event near Perth ( $\text{WA time is UTC}+8$ ). If swimmers arrive at $11:30\,AM$ , the UTC time is $03:30\,AM$ .
Argo Data: A specific float (e.g., number $5901625$ ) shows repeated cycles of profiles over time.
Ship of Opportunity Tracking: Using Continuous Plankton Recorders ( $CPR$ ) to find the northernmost latitude ( $-10.5^{\circ}$ ) where phytoplankton abundance was measured.
AUV Viewer: A specialized tool ( $av.aodn.org.au$ ) for stereoscopic seafloor images collected by autonomous robots.
Squidle+: A platform for crowd-sourced or expert annotation of images from the seafloor.

Questions & Discussion

Question Regarding Phytochemical Measurement: A student asked how satellites measure phytochemicals.
Response: Satellite products often measure chlorophyll coverage as a proxy for phytoplankton abundance, essentially using color data to determine biological distribution.
Technical Note on Web Portal: The instructor noted the current portal is over a decade old and can be "finicky" with certain searches (e.g., searching "SST" may yield different results than "Sea Surface Temperature"). The new portal is expected within a month.