hazards-coastal

Code and plots for 'coastal and estuarine flooding' and 'coastal erosion and shoreline change' NCRA hazards

Description

GitHub repository for ACS Coastal Hazard Team to store, track and develop code.

Indices considered by the hazard team:

• Mean Sea Level (Essential Climate Variable)
• Flood days
• Flood extent of 1% AEP extreme water level
• Change in frequency (multiplication factor) of storm tide levels (flooding metric) and total water levels (erosion metric).

Products:

Global Warming Levels are unsuitable for understanding sea-level rise (SLR)-related hazards due to SLR continuing long after temperatures stabilise.  We selected several SLR increments (SLRI) relative to the IPCC AR6 baseline of 1995 – 2014 for the National Climate Risk Assessment as follows:

• 0.06 m: Approximately 2011-2030 mean, the "present-day" period used in NCRA
• 0.1 m: Approximate projected SLR at 2030
• 0.2 m: Approximate projected SLR at 2050
• 0.38 m: Approximate projected SLR GWL 2.0, assuming GWL 2.0 is reached in 2090.
• 0.6 m: Approximate projected SLR at GWL 3.0, assuming GWL 3.0 is reached in 2090.
• 1.0 m: Higher-end projected SLR consistent with jurisdictional planning benchmarks used in Australia (c.f., Dedekorkut-Howes et al. 2021, DOI: 10.1080/14693062.2020.1819766)

In terms of the colors:

• 🟢 The data is available in its final official form
• 🟡 The data creation is currently in progress and available soon
• 🔴 The data processing has not yet started
• ⚪ Not intended for delivery/not applicable

In terms of the methods:

• MSL: Regional sea level projection datasets are provided at yearly (2020 - 2150) under 5 SSPs (119, 126, 245, 370, 585) in NC format for Australian region (105E-165E, 5S-50S), rather than using the sea level rise increments.

• Flood days: These were calculated using results from Hague & Talke (2024) [https://doi.org/10.1029/2023EF003993]. This provides 1000 different future realisation of flood days, based on different ways of storm surges and tides coinciding. This assumes no change in future tides and storm surges, based on homogenised sea-level data and IPCC AR6 SLR projections. The original SLR projection source was reapplied to the data to re-express the results in terms of GWL, as follows:
  • Identify years in projection timeseries (for different SLR scenarios) where SLR equals the increment of interest
  • Use the flood days estimate for all scenarios to estimate an average 10th, 50th and 90th percentile estimate from.

• Flood extent: Refer: https://github.com/AusClimateService/ncra_coastal_hazards/blob/main/Inundation_layers.html. The SLR increments are approximated using Year 2020 for SSP1-2.6 (0.06 m), 2030 for SSP1-2.6 (0.1 mm), 2050 for SSP2-4.5 (0.20 m), 2090 for SSP1-2.6 (0.38 m), 2090 for SSP5-8.5 (0.6 m), and 2100 for SSP5-8.5 (1 m). The underlying DEM has fundamental vertical accuracy of 0.3 m and horizontal accuracy of 0.8 m. The datafiles (geoJSON) are presented by Local Government Area. Further information on the methodology is provided in files for each LGA (available on NCI).

• MF: All approaches use the formula of Hunter (2012) [https://doi.org/10.1007/s10584-011-0332-1] MF = exp(SLR/lambda), where lambda is the Gumbel scale parameter. Three different estimates of MF are obtained by fitting the Gumbel distribution to annual maxima of three different datasets:
  • quality controlled but unhomogenised tide gauge data from GESLA3 (Haigh et al. 2022) [https://doi.org/10.1002/gdj3.174]. These are files of MFSLR_GESLA3_MMALL_SLR<>.csv
  • storm-tide reanalysis of Colberg et al. (2019) [https://doi.org/10.5194/nhess-19-1067-2019] (see column 'SWL_MFSLR_<>' in files of format MFSLR_ACSSH_MMALL_SLR<>.csv)
  • storm-tide reanalysis of Colberg et al. (2019) with addition of empirical estimate for shoreline wave setup from O'Grady et al. (2019) [https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2018JC014871] (see column 'MTWL_MFSLR_<>' in files of format MFSLR_ACSSH_MMALL_SLR<>.csv)
  • For further information, please see: https://github.com/AusClimateService/ncra_coastal_hazards and the compiled markdown page https://htmlpreview.github.io/?https://github.com/AusClimateService/ncra_coastal_hazards/blob/main/Extreme_water_level_hazards.html Both of which have an accurate description of the data and code for reference.
  • Difference data is provided at /g/data/ia39/ncra/coastal/MF/difference_files. The names of the files are the associated additional SLR relative to the 2021 - 2030 baseline period (0.06 m above IPCC AR6 baseline).

(Table was last updated 11201m 30/07/2024)

Index/metric	SLRI data	SLRI 2D map	SLRI change data	SLRI change map	(Notes)
MSL	🟢	🟢	⚪	⚪	Data present as annual under SSP, not SLRI Data is at /g/data/ia39/ncra/coastal/MSL
Flood days	🟢	🟢	🟢	🟢	Data is at /g/data/ia39/ncra/coastal/flood_days
Flood extent	🟢	🟢	⚪	⚪	geoJSON files at LGA scale information only. Data is at /g/data/ia39/ncra/coastal/flood_extents
MF	🟢	🟢	🟢	⚪	Data is at /g/data/ia39/ncra/coastal/MF

Example of Climate Risk Overview information

The below plot shows that minor flooding will occur frequently in the future. Under 0.2 m SLR, we estimate a 50% chance that these sea levels will occur 38 days per year on average. Under 0.6 m SLR, these these minor flood levels will occur daily at some locations with 90% probability. Flooding will be chronic at many other locations in eastern Australia.

FAQs

(a work in progress, contact ben.hague at bom.gov.au if you have a question that is not answered here)

• What are the limitations of the indices and data?

Some limitations that apply to all indices that follow from the fundamental assumptions that underpin their production. For example, we only consider the effect of mean sea level rise on future hazards, we do not consider any changes in sea level variability. This means we assume that storm surge magnitudes and tidal ranges do not change from the historical period.

The MF calculations assume that a Gumbel distribution is the most appropriate for characterising the relationship between sea level heights and frequency. A consequence of this is that the factor by which sea level extremes increase is frequency is the same regardless of flood severity - the factor by which the frequency of the present-day 1% AEP increases is the same as the once-a-year level, for example.

• What are some of the key insights?

Present-day sea level extremes and associated flood impacts will become very frequent with future sea-level rise, especially in locations in estuaries, harbours, and bays where the effects of breaking waves on water levels are minimal.

Sea level extremes and flood heights will increase approximately in line with increases in mean sea level. For example, the 1% annual exceedance probability (AEP) flood level under 0.2 m SLR will be 0.2 m higher than the 1% AEP level today.

• How are the percentiles calculated?

For flood days, these represent the effect of different ways that storm surges and tide can coincide on flood day estimates under the SLR increment considers. They are a measure of aleatoric/stochastic uncertainty.

For MF, these represent different amounts of SLR, based on the underlying SSP scenarios identified above. For example, the 0.20 m SLR increment is the 50th percentile in the 0.20 m SLR increment files, but the 10th and 90th percentiles are 0.17 m and 0.26 m respectively. They are a measure of epistemic (SLR response to climate system changes) uncertainty. See: https://github.com/AusClimateService/ncra_coastal_hazards for more information.

Authors and acknowledgment

Hazard team:

• Julian O'Grady (CSIRO, lead)
• Ben Hague (Bureau of Meteorology, alternate lead)