The threat of sea level rise to the nuclear industry

In this week's post I will be taking an alternative view on the relationship between nuclear power and the environment, considering how the environment threat of sea level rise (SLR) could hinder the safe operation and development of future plants.

Nuclear power plants are built close to the coast to reduce the costs of transporting vast quantities of water needed for cooling inland Source: Guardian, 2012

At present few nuclear power stations are waterproofed and many in the US and Japan in particular have no sea wall (Lipscy, 2012). Substantial sea level rise could put more locations at danger and heighten the need for protection. 

Likely SLR? 
The extent of SLR is key in quantifying the risk to the nuclear industry. The recently published fifth IPCC report projects a rise of between 0.33 -0.63m by 2100  for the middle emission scenario. However this projection is far from precise and not universally accepted; some semi-empirical models have predicted SLR up to twice as high by the end of the century (IPCC, 2013). 

Confidence in the IPCC projections have been reduced by the fact that model estimates  of the various contributions to the budget (upper four blue entries) have been consistently below the observed rates, However as can be seen in the error bars there is uncertainty in both measures and some palaeodata (see Siddall et al, 2009) has supported the modesty of the IPCC projections Source: IPCC, 2007

Calculating the future sea level budget is by no means easy, it requires accurate estimates of  thermal expansion and related ocean heat-content change, by far the largest store in the climate system (Church et al, 2011). However perhaps the biggest tripping point is in the likely contribution of the Greenland and Antarctic ice sheets over the next century. This is something which the IPCC has played down, however there is a growing body of evidence to suggest that rapid dynamical changes are happening. Take Church (et al, 2011) study, introducing a terrestrial storage term and using higher estimates of sea-ice contribution it was able to accurately close the sea level budget for the period 1972-2010, unlike the IPCC. Significantly, using an energy balance model, it found that since 2004 the ice sheets contributed to sea level at a combined rate of 1.3mm yr^-1, a level approaching the combined rate of upper ocean thermal expansion and glacial ice, not represented in IPCC models. Many believe that the Greenland and Antarctic ice sheets are an important component in closing the sea-level budget, particularly since 1993, and could as a result lead to additional sea level rise into the 21st century. 

IPCC projections of global mean sea level rise over the 21st century across all representative emission pathways Source: IPCC, 2013

How big a threat? 

Given the lack of constraints on future SLR there is uncertainty with any risk assessment. At present, according to a study by Lipscy (et al, 2012) nuclear power in most areas appear safe (see below). However if greater than expected SLR is realised in combination with an increase in storm surge magnitude and an investment in plant construction worldwide, the graph below could change significantly. 

Difference between average wave height of largest recorded event and maximum plant/sea wall height in metres. A number above zero means the plant and sea wall both lie below the average wave height of an historical incident implying inadequate protection, as can be seen most plants at present have large negative values associated with them Source: Lipscy et al, 2012

In America many have voiced concern about the safety of their current generation of nuclear power plants especially in light of Hurricane Sandy which caused the emergency shutdown of three of it's nuclear power plants. 

"We learned that they (the region’s nuclear power plants) just barely made it through safely during Sandy, but that does not imply that future storms, when combined with continued sea level rise, could not cause serious problems,” said Klaus Jacob, seismologist at Columbia University, source: Magill, 2013

Barely is perhaps an exaggeration, in truth 24 plants out of the 34 in the storms path continued to operate at 100%, but the east coast power plants are at particular risk. Climate change will not lead to uniform sea level rise everywhere, spatial variations exist in terms of circulation and thermal expansion. Significantly a recent study by Sallenger (et al, 2012) found rates  3-4 times higher than the global average along a 1000km long hotspot on the North American Atlantic coast. 

The video below summarises some of the factors contributing to the additional sea level rise along the east coast. 

What should be done? 

In response to Fukushima and in the face of the uncertainties associated with global as well as regional SLR many have touted a move away from purely probabilistic thinking when considering adequate sea wall height and plant height. There are many problems in the estimation of very low probability events, with growing non-stationarity of the climate to consider,  and the need to extrapolate the magnitude of  long return period events from a short instrumental record (NRC,2013). The risk is not solely dependent on spatial location of course, Lipscy (et al, 2012) found that more vulnerable plants were associated with the operations of larger utilities. They tend to have more diverse operations and thus pay less attention to the safety of specific plants, especially older plants nearing decommission.  

Ultimately I believe the industry needs to consider a ‘worst case’ scenario based on upper estimates of SLR rise for a particular plant's expected operating time. Government regulation is needed to ensure synchronous review of climate projections with planned plant construction. As a result there will likely be a greater need for more than adequate protection, all of which will add to cost and serves to make nuclear power a less attractive preposition.