Hurricanes Impacting Hawaii Hurricane Iniki MFE 658 Lecture 3b Hurricane Impact

• Hurricane’s in Hawaii • Hurricane hazards • Hurricane forecasting • Hurricane risk assessment Hurricane Neki

1 2

Some Common Comments Hawaii Hurricane Impacts

• No hurricane has made landfall on Oahu. • Only Kauai gets hit Tropical Cyclone – The Big Island and Maui were struck by a hurricane in 1872 tracks within 200 – Dot 1959, Iwa 1982, and Iniki 1992 all impacted Kauai miles of the • Mountains protect us Hawaiian Islands – If so, why don’t the mountains of Puerto Rico or Taiwan protect since 1949. them? • There is no Hawaiian word for hurricane – No “Hawaiian Term” actually is not a surprise, since words such as Hurricane and Typhoon arise from local words for the winds observed. – David Malo (1843) defined five different Kona Winds.

3 4 Hawaii Hurricane Climatology Central Pacific Hurricane Climatology

Note: more hurricanes occur in the central Pacific during strong el niño years.

Why?

Number of hurricanes per month in the central Pacific. Hurricane tracks in the central Pacific from 1949-1998

5 6

Human Impacts More Iniki Impacts

• Economy - • 90% of structures on Kauai affected • To-date cost of Iniki • 14,118 damaged or destroyed – $2.8 Billion • 30% telephone poles down • 3 years later unemployment 12% • FY2005 Estimated • Social fabric altered-10% move away Impact on General Fund Tax Revenues – $3.6 Billion • REMEMBER: This coincided with the bursting of the Japan Bubble Economy of the late 1980’s.

7 8 Hurricane Impact Saffir/Simpson Hurricane Scale

Category 1 74 - 95 mph Storm surge 4 - 5 ft Category 2 96 - 110 mph Storm surge 6 - 8 ft Category 3 111 - 130 mph Storm surge 9 - 12 ft Category 4 131 - 155 mph Storm surge 13 - 18 ft Category 5 > 155 mph Storm surge > 18 ft

• Energy of winds = wind velocity squared: E = V2 • A doubling of the wind speed (e.g., from 70 to 140 mph) 2 • Energy of winds = wind velocity squared: E = V results in four times the destructive energy. • A doubling of the wind speed (e.g., from 70 to 140 mph)

results in four times the destructive energy.

9 10 TRIPLE THREAT FROM Triple Threat TROPICAL CYCLONES Injuries and Loss of Life High winds Structural Damage Flash Floods TRIPLE THREAT flying debris Destruction of vegetation/crops Storm Surge Flooding of coastal areas (winds & low pressure) Erosion of Beaches Storm Surge Saline Intrusion–loss of fertility Large Surf Loss of Power/Communications High Winds Fires - Urban and Brush Contamination of Water Supply Heavy Rains Land Subsidence Flooding of Inland Areas Mud and Landslides

11 12 Hurricane Iniki Hurricane Hazards in Hawaii

The storm surge and storm waves are greatest just to the The storm surge and storm right of the storm track. waves are greatest just to the right of the storm track. Why? Wind hazard is greatest on exposed ridges and mountain slopes. Flash flood hazard is greatest near steam beds.

13 14

Past Hurricane Impact: Death vs Damage Hurricane Impact

Deaths Damage Population change and hurricane strikes for Miami Dade County.

1900 2005 The record is dominated by a few extreme events: Galveston hurricane of 1900 and Katrina in 2005 (2005 Dollars)

15 16 Storm Surge Storm Surge Current (N.H.) The strength of a storm surge depends on: – Wind strength Coriolis force Length and duration of the fetch deflects ocean currents – wind – Angle between storm path and coastline (90˚ to the right of the wind optimal) in the Northern Hemisphere – Shapes of coastline and continental shelf. Convergence of wind-driven currents, bays and estuaries tend to amplify the surge. Cyclonic surface wind stress causes cyclonic – Gently sloping continental shelf enhances surge. gyre in ocean with a dome of water, – The stage of the astronomical tide is important. upwelling, and surface water divergence. Low surface air pressure also contributes.

17 18

Storm Surge - Elevated Sea-Level Storm Surge - Elevated Sea-Level

Flooding of coastal areas Storm surges are historically responsible for the greatest Erosion of Beaches numbers of deaths and they cause some of the worst damage. Saline Intrusion–loss of fertility

19 20 Storm Surge Storm Surge

Debris line from Iniki storm surge

±8000 people died in the storm surge

21 22

Storm Surge Ivan Aftermath Storm Surge

 52 deaths in US Storm Surge  70 deaths in Caribbean

Storm Surge

Storm surge, before and after hurricane Camille. Storm Surge

23 24 Katrina ±1800 deaths in US Katrina Aftermath

25 26

Katrina Aftermath Hurricane Wind Damage

Oil production takes a hit from Rita and Katrina.

Wind Speed in hurricane Celia as it passed Gregory, TX.

27 28 Hurricane Wind Damage Iniki Wind Damage

29 30

Tangential Wind Profile Hurricane Wind Damage

Wind damage to Darwin, Australia from cyclone Tracy. Similar wind damage occurred in Andrew.

31 32 Wind Distribution Port Allen

180

Pressure Trace 160

140

120

Assuming cyclostrophic 100 balance, dp/dt ~ dv/dy 80

60

40

20

0

0.00 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00 2.25 2.50 3.50 4.50 6.00 7.00 8.00 9.00 miles 10.00 11.00 12.00 13.00 15.00 17.00 19.00 21.00 23.00 25.00 27.00 29.00 31.00 33.00 35.00 37.00 39.00 41.00 43.00 45.00 47.00 49.00 51.00 53.00 55.00 57.00 59.00 61.00 Cyclostrophic tangential wind distribution calculated for Port Allen.

33 34

Wind Distribution Tornados in Hurricanes

Observed wind gusts distribution at Lihue during Strongest tornadoes are found in right front quadrant. Hurricane moving toward 360˚, range rings at 100 km. Iniki. It is the wind gusts that do the damage.

35 36 Hurricane Rains and FLOODING River Flooding

Precipitation Precipitation Hydrologic Duration Intensity Aspects

37 38

Precipitation Duration TC Stagnation

• Evolution of Tropical-Storm Scale Circulations • TC in streamline col region. No steering current. !• Evolution of Mesoscale Features – convective ! • TC surrounded by highs. Good environment for TC. bands and elements • TC approaching another low, may cause TC to • Moisture availability in surroundings stagnate or loop. • Orographic Anchoring and Enhancement • High to NE may block progress of northeastward • Instability Replenishment for New Storms moving storm. • Non-Interfering Up and Down Drafts • Steering currents from opposite directions.

39 40 North Carolina Flooding Ivan Aftermath Heavy rains & Flooding in NC

Slow moving tropical cyclones result in very large rainfalls.

41 42 Hurricane Forecasting NWS Hurricane Advisories

Hurricane Mobile Homes • Hurricane Watch - Hurricane may threaten the coast. Issued ~ 36 hours in advance

• Hurricane Warning - Hurricane conditions are expected within 24 hours. Issued ~12-24 hours in advance. Hurricane Gridlock

Hurricane forecasting involves predicting the future track

and intensity of a hurricane.

43 44 Hurricane Forecasting Hurricane Forecasting

Hurricane forecasting involves predicting the track and intensity of a hurricane. More progress has been made Hurricane tracks are erratic and in hurricane track forecasting difficult to predict. than in forecasting hurricane A combination of observations, intensity change. numerical model output, and climate data are used to forecast hurricane tracks.

45 46

Cloud Drift Winds Satellite Rainfall Measurement

IR Imagery from Bonnie 8/25/98 1145 UTC (Blue Rings at 100 and 350 km Radius) • Hurricane Katrina

47 48 Hurricane Forecasting Hurricane Risk Analysis

• Hurricane Rita Simulation using a high resolution model with data assimilation

49 50

The Insurance Crisis of 1993 Band Aid Solutions

• Pre-Iniki “Hurricane” wind peril coverage was an add- on to fire and theft. • Florida created a catastrophe fund which is a re- • Iniki wiped out all profits of the homeowners insurance insurer. industry in Hawaii since its founding. • Florida also created a wind pool (insurer of last resort) • Hawaii is a small revenue source for big firms and for high-risk properties (think east of I-95 in Miami). provided a disproportionately large loss. • Hawaii created the Hawaii Hurricane Relief Fund • Iniki occurred within 18 days of Andrew (Miami) and (sounds like a Red Cross operation), a primary insurer. Omar (Guam). • The HHRF filled in until the industry recovered. • The global re-insurance industry was reeling. • Now in repose, still on the books • In both Florida and Hawaii companies went bankrupt or attempted to withdraw.

• Serious problems for the real estate markets.

51 52 Risk Modeling Risk Modeling (cont)

A model consists of • Issue : events like Hurricanes are very rare, Iniki is at 1.A peril. Here a hurricane least a centennial time scale event 2.The model simulates a massive number of hurricanes by • Normal actuarial techniques are not applicable projecting forward perhaps 100,000 years using statistical approaches. • In 1985 Karen Clark founded Applied Insurance Research and presented the first Hurricane risk model 3.The model then assesses impacts upon a portfolio of properties (vulnerability). 4.The model then applies actuarial functions to produce insured losses. 5.Based upon these projected losses, rates are determined.

53 54

Risk Modeling (cont) Risk Modeling (cont)

These went over like the proverbial lead • Florida created a Commission on Hurricane Loss balloon. Projection Methodology 1. The methodology is strange to those familiar with • Commission reviews models for use in the State of traditional insurance practice (think life insurance). Florida 2. Everybody’s rates were going up. • Due to Sunshine Laws, modeling companies resist 3. Insurance companies are evil anyway. public presentation of trade secrets. • Florida Commission created a professional team to conduct on site audits.

55 56 Risk Modeling (cont) Risk Modeling (cont)

The Florida Professional Team consists of a lead • The Pro Team signs non-disclosure forms and back-up in the areas of : • We audit against standards developed by the 1. statistics Florida Commission via a public process 2. meteorology (I have acted as reviewer for this section) • We file reports on our findings 3. structural engineering • We attest to our findings in public hearings 4. actuarial science • We work with the Commission to develop new, 5. computer science improved standards (raising the bar)

57 58

Risk Modeling - Hawaii Hurricane Risk Analysis

• Hawaii uses a simpler process • Three consultants are retained 1. A meteorologist 2. A structural engineer 3. An actuary 1 and 3 are both members of the Florida Pro Team • The consultants review applications of the models to the Hawaii situation • The first question asked is whether or not the model has been accepted in Florida

59 60 Hurricane Risk Analysis Wind Model Validation

The characteristics of the hurricane that are needed to produce an estimate of wind speed at a site are as follows: (i) Central Pressure Difference (ii) Radial Pressure Profile Parameter (iii) Radius to Maximum Winds (iv) Storm Translation Speed (v) Storm Track (vi) Latitude and Longitude of Site (vii) Surface Roughness at the Site (viii) Distance of the Site from the Coast.

61 62

Vulnerability Functions Questions?

• Damage is characterized by four parameters: occupancy, number of stories, structural system, and exterior cladding strength. • For residential structures, the 21 structure types (e.g., frame, Heavy Timber Joisted Masonry, Timber, Weak- Cladding, etc.) • The vulnerability functions in the model were developed largely on the basis of claims data from 1954- 1994.

Hurricane Luis

63 64