Management strategies

Although the sea will rise uniformly across a region, the way in which the rise will be experienced along the coastline will not be uniform in 3 main ways: 

First, take a short drive along a coastline and you are likely to pass numerous coastal configurations. Some locations will be more elevated, others will be lower, some resistant to erosion, and others vulnerable to erosion. All of these different configurations may exist within a few hundred metres of coastline. Likewise, the impact of sea-level rise and storm surges will be felt differently from one location to the next. One location may be very vulnerable to impacts of the sea, but take a short walk along the coast and you may find yourself in another place that is not at all vulnerable.

Second, what we as humans have placed behind the shoreline creates a myriad of complex relationships and responsibilities that are not easily managed. The example below at Hallett Cove could be replicated at many locations around the state.

Third, the way in which a coastline is orientated makes it more or less vulnerable to storms. In the Hallett Cove region, the largest storms come from the southwest and strike the landforms in a particular manner.

It is true that increased heat will be a driver of sea-level rise globally, primarily by thermal expansion and loss of mass from ice sheets. It also true that rising ocean temperature is likely to impact marine ecosystems across regions. However, the focus of this paper is to consider the impact of sea-level rise upon coastal environs.

The main point here is, that this impact will be experienced within localities, and the strategies we employ to deal with the sea-level rise problem must be fine grained and local.

While the coastline south of Adelaide was being pounded in the May 2016 event, Port Parham in the north of Adelaide remained dry, but only just so. If the level of water had been just 10 cm higher, the impact would have been significantly different.

In the context of gradual sea-level rise this example demonstrates how the impact of rises will be experienced within localities over time. A sea-level rise of 0.3 m may not be a problem in a particular location, but a sea-level rise of 0.4 m may be the point when that location becomes vulnerable to inundation by storm surge, or even from routine tides.  

Increases in sea level will also mean that storm events felt rarely now, will be more experienced much more frequently in the future. For example, a storm surge water level currently expected to occur on average only once every 100 years on the Adelaide coast would occur on average once every 5 years after 0.3 m of sea-level rise.

Tipping points are not just confined to areas of low elevation nor just to extreme events. Another little-known fact is that 38% of the South Australian coastline is made up of cliffs of varying erodibility. These cliffs have been quietly sitting there for the last 7,000 years with very little change occurring in the ocean². However, sea-level rise will bring routine tidal actions to the base of many of these cliffs, and some of them will erode quickly, not slowly. Where and when is not well known. Therefore, while the sea level in the region will rise uniformly, the timing and nature of impacts will be different for every location.

² Bourman, Murray-Wallace and Harvey 2016, Coastal Landscapes of SA, University of SA Press, pp 27–28.

The most effective way to identify these tipping points is by way of scenario modelling. In this modern era highly accurate modelling within a digital environment can identify when and where locations are likely to be vulnerable over time. It is important to remember that these are just projections of plausible futures rather than predictions of a certain future.

Scenario modelling is essential to:

• Understand the likely impacts upon localities over time.
• Guide land-use planning to ensure that new development is not placed in locations at risk both now and in the future.
• Provide a warning where existing development may be at risk over time.

Two examples of scenario modelling are shown here.

As part of my job I meet many people around the state, often when I am on the beach. I also have presented at about 30 town-hall and ‘tin-shed’ meetings, the latter a reflection on the nature of some of the smaller towns. At one such meeting I recall presenting for only a short time before one of the elected members exclaimed that my ‘flood mapping was causing property values to go down in the area’. Other comments I have heard include:

I am not afraid of what the sea can do, I am most afraid of what the council might do in forcing me to relocate.

Another resident in a community meeting said she was ‘most afraid of what the council might do because her house was her only asset.

The reason for this fear is that retreat (usually called planned or managed retreat), is most favoured by urban planners and imposed on the landowner by governments. However, recent history throughout Australia has shown that these proposals usually fail. Where ever an attempt has been made to implement planned retreat, community backlash and hostilities have developed between governments and the landowners³.

The reaction of property owners is one problem associated with publishing the scenario modelling. In some places around South Australia scenario modelling suggests that some locations are not likely to be viable places to live by the end of the century if the sea-level-rise projections are realised.

But on the other hand, people need to be warned, especially in the act of purchase.

In 2015, I was inspecting a levee bank when a man on a neighboring property popped his head over the fence and asked: ‘Do you know anything about flooding in this area?’. He had just purchased the house. On 9 May 2016 it had water over the floor.

In another location a man who attended a community meeting had just purchased within the township. He said that he was not aware of any sea-level rise problems in the area prior to purchase. The first house was under a land-management agreement, so the warning should have been adequate. It is worrying that the second man had little knowledge. Without adequate warning councils are likely to be liable should new owners discover that they have purchased properties that are likely to become increasingly vulnerable to actions of the sea.

Finally, while opinions still vary widely about climate change, over the last few years the trend has been for people to be more accepting that ‘things are different than they were’, especially after the storms events of 2016. The prevailing mood tends to be: ‘What can we do about the threat?’.

³ Western M 2016, Review of coastal hazard policy contained within the SA Policy Planning Library, pg 1.

To acknowledge uncertainty does not suggest that the science that underpins sea-level rise projections is not sound. The CSIRO report, Sea Level Rise and Allowances for Coastal Councils around Australia provides an excellent overview from a myriad of sources to demonstrate that the increases in sea level, especially since 1970, can only be explained by anthropogenic contributions. And the data from the tidal gauges and satellite measurements is clear, the rate of sea-level rise has accelerated in this century when compared with last century.

A greater understanding for the previous rises does provide a basis for greater confidence in projecting future change. However, this does not negate the problem of compounding variables inherent in the climate models. For example, due to the natural variability of our climate identifying the exact contribution of climate change to sea-level rise is difficult. Long periods of data collection and analysis are required, usually in the order of 20 years. Rapid ice melt is not factored into the models. Feedback loops within climate are only now being more understood. And the greatest variable of all is how the global community might change its production of green-house gases.

In 2015, the IPCC produced a range of sea-level rise scenarios to account for the various trajectories that greenhouse emissions may take. The figure below depicts the projected sea-level rise trend using a high-emission scenario. The insertion of the sea-level rise policy bench marks for 2050 and 2100 demonstrates that this scenario has been adopted by the state government.

The first half of the century is more certain. In local waters the rate of sea-level rise has averaged 4–5 mm per year since 1992 (with a very small acceleration observed in the satellite data). At this current rate of sea-level rise we are on target for the projected rise of 0.3 m by 2050. The projections for the second half of this century are far less certain. At some stage in the future an escalation in the rate of rise must occur if the projections for 2100 are to be realised.

What is the antidote for uncertainty?

The short answer is: the collection and analysis of accurate data over time.

If the projections for the end of the century are to be realised, then an escalation in the rate of sea-level rise will be identified. Global monitoring will detect this rise, but rates will vary around the world due to natural climate variability. Local data will be essential to make the political case for change. As a general rule people have a higher trust for local data than they do of global data⁴.

Initially, South Australia did have two SEAFRAME gauges as part of the Bureau of Meteorology’s highly accurate network of tide gauges around Australia: one at Thevenard, and another within Gulf St Vincent at Port Stanvac. After providing 18 years of data the gauge at Port Stanvac was decommissioned in 2010 to make way for the desalination plant.

⁴ Townsend, Murray 2017, presentation to SEGRA conference in Port Augusta, Department for Environment and Water.

Accurate monitoring of sea-level rise in local waters will provide the data necessary to identify when the rate of change is escalating, either due to thermal expansion, or from rapid ice melt.

Over the coming decades the sea-level rise ‘picture’ will become clearer and the projections more accurate, but in the meantime the precautionary principle should apply for any substantial new development. The life spans of buildings and infrastructure are long and new development, such as subdivisions and urban infrastructure that we implement today, is likely to remain at the end of this century and beyond.

But how to deal with existing urban settlements is problematic in the face of uncertainty.

Currently, as a society we are tending to make the case for change based on projections, what we could call a ‘predict and respond’ model of adaptation. We continually use the projections to point to the future and appeal for change. However, because of the uncertainty described above, this model of adaptation is not likely to have sufficient impetus from which to make real political and social change in the way in which we are interacting with the coastline.

The recommendation is, that we should move to a ‘monitor and respond’ model of adaptation.

What should we monitor?

In addition to monitoring sea-level rise in our local region, we should monitor coastal change. How fast is the shoreline eroding? How fast did it erode in the past? Where are the locations that are experiencing their first touch of erosion from rising actions of the sea?

However, before coastal monitoring programs can be activated a baseline must be created. If a baseline is not created, then the question is, ‘monitor, in relation to what?’.

• Creating baselines

There are a number of ways to create a baseline within urban coastal environments, but the most effective way in this current era is to create a digital baseline. The 2 main technologies utilised today to create digital models are LiDAR and photogrammetry.

Without digital models we cannot produce scenario modelling from which to create a picture of plausible futures, nor create the necessary baseline upon which to measure coastal change over time.

Historical flood and tidal analysis can provide a baseline understanding of how the sea is currently interacting with the coastline. A study of the history of shoreline recession may provide an indication the rate of erosion over time. All of these facets can be modelled within a digital environment.

• Monitoring change over time

Monitoring changes in the land forms over time can take place by direct observation, photographic comparisons, or sophisticated means such as recapturing digital data and making comparisons within a software environment.

Monitoring how the sea behaves in localised regions now will aid in understanding how it may behave in the future. Monitoring of data can be achieved by direct observation, photographs, videos, and interim modelling with cost-effective tide gauges.

Analysing the impact of extreme events provides an improved understanding as to how actions of the sea are impacting localities around the state. The public are usually willing to provide copious amounts of photographs, videos and anecdotes about extreme events. In some locations, people even mark the height of floods over a long period of time.

In summary, the imperative is to identify a current baseline, and then collect and analyse accurate data over time.

A ‘monitor and respond’ approach to adaptation brings together all of the necessary facets:

• Develop strategies that are fine grained enough to manage adaptation within localities.
• Use scenario modelling to identify plausible futures.
• Track sea-level change within local waters.
• Establish coastal baselines.
• Monitor change on coastlines over time.

And finally, taking a ‘monitor and respond’ approach to adaptation is more likely to bring the community along the adaptation journey, rather than a fear-based approach which tends to focus on a dire future.