What's Changed

• Test of directly syncing specific tide model subfolders by @robbibt in #86
• 2023 update by @robbibt in #87
• Add AWS CLI to docker image by @robbibt in #91
• Fix styling bug in 2023 uncertain shorelines by @robbibt in #93

Full Changelog: 2.1.1...2.2.0

What's Changed

• Fixed bug with remote offshore tiles failing to process. Updated code to ensure empty arrays use a correct datetime datatype to allow them to be combined with arrays containing valid satellite data.

What's Changed

• Added annual shoreline data for 2022. The 2022 shoreline is interim data that is subject to change, and will be updated to a final version in the following 2023 DEA Coastlines update (in July 2024).
• Added dataset maturity filter. This ensures that the DEA Coastlines analysis is run on only the highest quality "final" maturity satellite data.

What's Changed

• Update Github Actions workflow to push Docker image directly to ECR (based off dea-sanddbox) by @pindge in #78

Full Changelog: 2.0.0...2.0.2

What's Changed

Updates to Python package dependencies to run on latest DEA Sandbox version 2.0.0 image.

Full Changelog: 2.0.0...2.0.1

What's Changed

• Added annual shoreline data for 2021. The 2021 shoreline is interim data that is subject to change, and will be updated to a final version in the following 2022 DEA Coastlines update (in July 2023).
• Improved data coverage over offshore islands and reefs. This includes new shorelines and rates of coastal change over the Indian Ocean, Great Barrier Reef and Coral Sea.
• DEA Coastlines now uses data from the FES2014 global tidal model. This model produces accurate tide modelling across Australia and globally, and is more flexible and easier to install than the older TPXO8 tidal model.
• Inclusion of Landsat 9 data to provide additional satellite data from late 2021 onwards.
• Improved shoreline mapping certainty attributes to provide information about the quality of each individual annual shoreline (compared to the previous approach of identifying all annual shorelines in an area as low or good quality). These annual shoreline certainty attributes include:

• good: Good quality annual shorelines with no low quality flags
• insufficient data: The accuracy of this shoreline may be affected by limited good quality satellite observations at this location. This can lead to noisier and less reliable shorelines.
• unstable data: The accuracy of this shoreline is affected by unstable data at this location. This may be caused by errors in the tidal model used to reduce the influence of tide, the presence of gently sloping intertidal mudflats or sandbars that can lead to inaccurate shoreline mapping, or noisy satellite imagery caused by high levels of cloud.
• aerosol issues: The accuracy of this shoreline may be affected by aerosol issues caused by the 1991 eruption of Mount Pinatubo.

• New rates of change point certainty attributes to provide information about the quality of each rate of coastal change. These new certainty attributes include:

• good: Good quality rates of change points with no low quality flags
• likely rocky shoreline: This coastline has been identified as a probable rocky or cliff shoreline. Rates of coastal change at this location may be less accurate due to noisy shoreline mapping caused by dark terrain shadows.
• insufficient observations: There are insufficient years of good quality annual shoreline data (< 25 years) to calculate reliable rates of coastal change at this location.
• extreme value (> 50 m): This location has been identified as having an extreme rate of coastal change (> 50 metres per year) and should be interpreted with caution.
• high angular variability: This rate of coastal change is unlikely to be accurate due to high levels of angular variability from this point to each annual shoreline. This can occur in complex coastal environments like river mouths, sandbars and mudflats that do not follow linear patterns of coastal change over time.
• baseline outlier: The baseline (i.e. most recent) annual shoreline is itself flagged as an outlier, potentially resulting in inaccurate rates of change at this location.

• Updated coastal change hotspot layers: three coastal change hotspot layers are now included in the product, summarising coastal change within 1 km, 5 km and 10 km windows around each hotspot point. These layers now include attributes providing the median position of all annual shorelines around each hotspot point, which can be used to plot and identify regional trends of coastal change.
• Rates of change points contain a new "uid" ID field containing a unique geohash ID that can be used to more easily track rates of change points as they are updated in future DEA Coastlines updates.
• Rates of change points include a new "id_primary" field containing the ID of their Primary sediment compartment from the Australian Coastal Sediment Compartments framework. This enables points to be easily queried and summarised by coastal sediment compartment.

Backwards incompatible changes:

• The annual shorelines WFS layer dea:coastlines has been renamed to dea:annual_shorelines.
• The rates of change points WFS layer dea:coastlines_statistics has been renamed to dea:rates_of_change.

Full Changelog: 1.2.8...2.0.0

Allow tiles to be generated without an error even if no rates of change points are available (e.g. just shorelines). This may produce additional data over low data areas (e.g. remote oceans) where we may potentially be missing entire years of data.

Full Changelog: 1.2.7...1.2.8

Remove test of broken path after testing it works

Full Changelog: 1.2.6...1.2.7

Test of automatic re-try for tide model access issues

Full Changelog: 1.2.5...1.2.6

• Make shapefile export in continental script optional to reduce run times while testing
• Tweak all time raster thresholds for river and coastal study area extent generation

Full Changelog: 1.2.4...1.2.5