Responses to storm damage: research evidence for early warning and evacuation planning

This post was contributed by Dr Sue Brooks, Senior Lecturer in Physical Geography in Birkbeck’s Department of Geography, Environment and Development Studies.

Few people can have failed to notice the high flood incidence over the past few weeks across the UK. One particularly newsworthy event was the storm on 5-6 December 2013 that brought widespread coastal flooding into the southern North Sea and spread destruction across Europe.

Flooding on the North Norfolk Coast following the storm of 5-6 December, 2013 (photos courtesy of Mike Page)

Flooding on the North Norfolk Coast following the storm of 5-6 December, 2013 (photos courtesy of Mike Page)

Photo courtesy of Mike Page

Photo courtesy of Mike Page

The worst in 60 years?

The 5-6 December storm forced evacuation of over 10,000 homes in Norfolk and Suffolk, flooded over 1, 400 properties and caused houses to collapse into the sea. It appears to have been the most serious storm for 60 years, following the notorious storm surge of 1953 when over 2000 people lost their lives. But it could happen again next week, next month or next year – so we need to heed the lessons from both events.

Lessons from 1953

So how do these two storm events compare, why did the earlier event lead to such greater human cost and what can new research on the 2013 event tell us?

The 2013 and the 1953 storms arose in similar ways involving deep low pressure systems tracking down the North Sea. But the 1953 event, with eight metre waves, also had devastating consequences because it struck over a weekend. Early warning and evacuation planning were piecemeal, with the radio warning ceasing to transmit from midnight on the Saturday and the offices set to receive telegrams being shut for the weekend. People had very little warning. In 2013, in view of lessons from 1953, the sea encountered strengthened and heightened flood defenses, better early warning and better co-ordinated evacuation planning.

How our research is providing new lessons for early warning and evacuation planning

Following the 5-6 December storm we wasted no time recording its coastal impact. A team of researchers from Birkbeck and the University of Cambridge documented water levels around the coast of East Anglia. We used debris drift lines (we had to be quick as the tidy up was swift), erosional notches in banks and cliffs (these are preserved for longer), water marks on buildings and anecdotal evidence from local residents to position the water levels.

Drift lines and water lines on buildings in the immediate aftermath of the storm recording water elevations at Burnham Overy Staithe, Norfolk (photo taken on 6th December, 2013 by T Spencer)

Drift lines and water lines on buildings in the immediate aftermath of the storm recording water elevations at Burnham Overy Staithe, Norfolk (photo taken on 6th December, 2013 by T Spencer)

Water line evidence on public toilet block at Brancaster Beach, Norfolk following 5-6 December storm surge (photos taken on 6th December, 2013 by T Spencer)

Water line evidence on public toilet block at Brancaster Beach, Norfolk following 5-6 December storm surge (photos taken on 6th December, 2013 by T Spencer)

Brancaster Beach Sign

We used a Global Navigation Satellite System (GNSS) accurate to 20mm to determine the precise water elevations. We found 2013 water levels very similar to those of 1953, and in places 1953 levels were exceeded. More exciting was to find considerable variation in the water level around the coast, about two metres difference in North Norfolk.

Why was the water level so variable?

Water levels recorded in tide gauges in sheltered locations show how the surge level became progressively higher and occurred later as the storm moved southward (see diagram). Surge models can predict this effect and can thereby generate early warning and evacuation procedures against imminent flooding. However, they can only operate at this broad regional scale.

Recorded and predicted water level at locations along the east coast of UK, 5th-6th December 2013 (based upon data from the National Tidal and Sea Level Facility, provided by the British Oceanographic Data Centre and funded by the Environment Agency)

Recorded and predicted water level at locations along the east coast of UK, 5th-6th December 2013 (based upon data from the National Tidal and Sea Level Facility, provided by the British Oceanographic Data Centre and funded by the Environment Agency)

Strong, persistent winds generate high waves, also a factor in coastal damage. Waves were four metres at the Blakeney overfalls wave rider buoy (10 km offshore). Waves are considerably modified by the coastal setting. Along the North Norfolk Coast they were able to add significant force on top of already elevated water levels.

The collapsing cliffs at Covehithe, Suffolk (research will quantify the land loss in the 5-6 December, 2013 storm surge) (photo taken on 11th December, 2013 by SM Brooks)

The collapsing cliffs at Covehithe, Suffolk (research will quantify the land loss in the 5-6 December, 2013 storm surge) (photo taken on 11th December, 2013 by SM Brooks)

Further around the coast in Suffolk we found evidence for cliff erosion, arising from wave action at the cliff base (Brooks et al., 2012). Our initial findings suggest that water levels and wave action reached almost four metres ODN (Ordnance Data Newlyn, which approximates to mean sea level), producing notching and cliff collapse. Loss of land and homes through cliff retreat is irreversible and cliff retreat can continue long after the surge event has happened.

What lessons can we take forward from 2013?

  • Water elevation differences affect a property’s flood risk.
  • Homes and businesses need to have information on their specific vulnerability.
  • Models provide general predictions of timing of surges and open sea water levels.
  • We don’t currently consider how surges interface with the coastal setting encountered.

From 2013 we have now learnt that in coastal settings such as North Norfolk and Suffolk, the barrier islands, dunes and gravel spits, interspersed with tidal inlets with marshes and mudflats and separated by eroding cliffs make for huge variability in the potential for the sea to ingress and flood land, as well as to cause cliff retreat through wave action.

Our results show there are still lessons to be learned that could help prevent future societal and environmental damage that accompanies storm surges.

The team’s initial assessment is published in Nature and further information is available here.

Share
. 1 comment . Category: Social Sciences History and Philosophy . Tags: , , , ,

Who pays the cost of flooding?

This post was contributed by Dr Diane Horn, Department of Geography, Environment and Development Studies and former visiting scholar at Old Dominion University’s Climate Change and Sea Level Rise Initiative, and Michael McShane, associate professor of finance and co-director of the Emergent Risk Initiative at Old Dominion University. It was originally published by the Pilot.

The U.S. National Flood Insurance Programme is facing rough seas  ahead. The programme is about $26 billion in debt after Hurricane Katrina in 2005 and Sandy in 2012. Worse, a large percentage of US property owners are heavily subsidized and do not pay full, risk-based rates. In addition, flood insurance is required only for certain property owners in high-risk flood zones.

No private insurance scheme would survive in a market where only the high risk buy insurance and do not pay risk-based rates.

The insurance programme has survived for 45 years because it can borrow from the U.S. Treasury when flood payouts are more than the amount paid in by policyholders, paying back the loan with interest in years where there aren’t many floods.

This worked pretty well until the 2005 hurricane season and the massive payouts after Katrina. Since then, much of the programme’s income has gone to cover interest on the debt, with little available to pay the debt down. The situation has resulted in calls for fiscal responsibility and the end of subsidies for all high-risk properties.

In this case, however, the “solution” has caused more problems. The Biggert-Waters Flood Insurance Reform Act of 2012, which aims to phase out these subsidies, has created another set of angry constituents – property owners who are experiencing big increases in their flood insurance premiums.

Bipartisan bills in Congress would delay the move toward risk-based rates. The programme finds itself between a rock and a hard place: those who are calling for fiscal responsibility and the end of the subsidies, and those who are having a hard time affording the new rates.

Ongoing research, part of the Climate Change and Sea Level Rise Initiative at Old Dominion University, is looking at how other countries deal with flood losses in the search for solutions to this dilemma.

Our research started by comparing flood insurance in the United States and the United Kingdom, where flood insurance is completely handled by private insurers. In the U.K., flood insurance is included as part of standard homeowners’ policies. This avoids one of the big problems in the U.S., where even people who live in the floodplain don’t buy flood insurance.

Including flood damage as part of a single policy would solve another U.S. problem: determining whether wind or water caused the property damage. Hurricane Katrina shone a light on this problem. If wind caused the loss, it was covered by the homeowner’s insurance policy. If storm surge caused the loss, the flood insurance programme policy would pay.

However, the use of private insurance in the U.K. doesn’t solve the problem of subsidies. Just like in the U.S., British property owners who live in the floodplain pay a lower rate than they should because flood insurance is subsidized. The difference is that in the U.K., subsidies come from other policyholders; the government doesn’t pay anything toward the cost of flood damage.

The U.K. is moving to a new flood insurance system, however. Under the new scheme, policyholders outside flood-prone areas will pay around $17 per year to make the rates more affordable for the high flood-risk policyholders. Rates will still be high in flood-prone areas but not so high as to be unaffordable, at least not initially. The system is designed to end subsidies in 20-25 years, unlike the U.S., where the national flood insurance subsidies are ending over a five-year period.

In the U.S., some are calling for private insurers to take over the flood insurance programme. However, you would be hard pressed to find a private insurer interested in offering flood insurance. Private insurers could not survive in an insurance market where only those most at risk buy flood insurance; they would need to charge even higher rates for those high flood risk policyholders than the national programme charges. The premium shock would be even worse, and Congress would be under even more pressure to step in and reduce rates.

Floods aren’t going to go away, and whenever a flood occurs, someone has to pay to repair the damage. Even with flood insurance, most policyholders don’t pay a price that reflects their true risk, and most flood insurance policies are subsidized to some extent. The real question is who pays the subsidy.

The U.K. experience shows that leaving everything to the private sector doesn’t work, but the U.S. experience shows that leaving everything to the public sector doesn’t work, either. We need to come up with the right mix of contributions from government, individuals and insurance companies – before the next Katrina or Sandy comes along.

Dr Horn and Dr McShane’s paper Flooding the Market” was published in the November edition of the journal Nature Climate Change.

Share
. 1 comment . Category: Social Sciences History and Philosophy . Tags: , ,