For much of the country, summer is synonymous with good timesand having fun. But on the Gulf Coast and along the EasternSeaboard—and particularly at property and casualty insurancecompanies—it's also a time for trepidation. As we approachSeptember, and hurricane season is fully underway, people fromMaine to Florida, Alabama to Texas, brace for the potentiallydevastating effects of hurricanes developing in the Gulf and theAtlantic.

Hurricanes are massive, complicated weather systems, due totheir complexity and the difficulty of gathering data in some ofthe most extreme conditions on Earth. But advanced technology likeLiquid Robotics' (LRI) Wave Glider®, an unmanned ocean robotcapable of operating through even the most severe Category 5storms, can continuously collect and transmit hurricane intensitydata in real time. This is game changing.

Wave Gliders connect the undersea world to air, space and shorethrough unprecedented monitoring, assessment and protection of theworld's oceans. They have survived 17 hurricanes, cyclones andtyphoons, and most recently tweeted their way through the 2014hurricane season. They collect real-time data to aid meteorologistsin better modeling and predicting hurricane intensity; a key toimproving evacuation preparedness, reducing false alarms and savinglives.

Certainly weather services and emergency preparednessprofessionals stand to gain a great deal from these advancedsystems. Catastrophe (CAT) modeling in the insurance industry alsois realizing the benefits of having superior and more accuratedata.

It's all about the data

Crucially important to the work of insurance providers, CATmodels provide a reasonable estimation of the potential damage,strength, and trajectory of an incoming storm system. Historically,CAT models were used only by reinsurers and typically applied onlyto aggregate portfolios. Over the past 10 years, however, CATmodels have been increasingly applied through risk-based assessmentand analysis. But despite this rise in popularity, CAT models stillsuffer from a lack of accurate data.

A model is only a representation of reality. Depending on thequestions being asked a model could be highly complex or extremelysimple, and it is in understanding the limits of a model that itsvalue can be properly achieved. If you don't have real data toinsert into the CAT model—for example, the true intensity of thestorm—then the model will be flawed.

This was particularly true in 2005, when Hurricane Katrina madeland fall in Louisiana. CAT models failed to predict Katrina'shistoric intensity and direction, leading to an underpreparedemergency response and catastrophic destruction.

On average, it costs $1 million to evacuate and prepare one mileof coastline for an incoming hurricane. Though it is a hefty pricetag, this cost is but a fraction of the potential life and monetarycost incurred by lack of preparedness for a serious storm. WhenHurricane Katrina hit an underprepared Louisiana, more than 1,500people were killed and total property damage was estimated to be inexcess of $100 billion.

During Hurricane Andrew, which occurred in August 1992, morethan $10 billion in insured residential damage and $16 billion intotal insured damages were reported.

CAT modeling failed to predict these storms and the intensedamage they would cause—a reflection of the poor quality of dataupon which they are built. Many models are inaccurate simplybecause they're constrained by a lack of data and scientificknowledge. This is certainly the case with the catastrophe modelsused extensively by the insurance industry. No matter how manyPh.D.s work on a catastrophe model, the fundamental uncertaintiesaround the frequencies and intensities of large magnitude eventscan't be removed.

Sources of data gathering

Historically, data on hurricanes has been gathered by threemethods: P-3 "hurricane hunters," planes flown directly abovemassive storms to collect data from 10,000 feet above the ocean'ssurface; moored weather buoys that often become inoperable duringthe violent storms; or satellites collecting data 250 miles abovethe surface of the ocean where collection of the intensity dataoccurs.

These sources of data play a significant part in understandingthe development and trajectory of a hurricane, and data collectedfrom such a significant distance paints an inadequate picture ofthe storm's intensity and movement potential. CAT models used forKatrina and Andrew found this to be true, and the resultingunder-preparation was devastating. But the Wave Glider will improvethe accuracy of the CAT modeling because it is providing real-timedata from the surface of the ocean during the entire storm.

As insurers continue to demand higher quality data, thetechnology industry will rise to meet the challenge. Currently,LRI's Wave Glider is the only unmanned ocean robot proven tocollect and transmit real-time data from the sea's surface duringactive hurricanes. Far more accurate than a plane 10,000 feet abovethe surface, Wave Gliders provide a critical stream of storm datato enable insurers and forecasters to more accurately predict theintensity of an incoming storm. With the potential to savethousands of lives and billions of dollars, utilization oftechnology like the Wave Glider is an easy way to drasticallyincrease the efficacy of existing CAT modeling.

I firmly believe that improved and accurate data from LRI's WaveGlider technology could encourage insurers who have left thevoluntary market in coastal states—where homeowners must obtaincoverage from government-mandated plans—to return, ultimatelybringing down the cost of coverage with more competition and ashared risk pool.

What's more, a fleet of Wave Gliders pre-positioned off thecoast of the Atlantic and in the Gulf has the potential to savethousands of lives and billions of dollars by providing earlywarning data on storm intensity. As CEO of LRI, I know that theWave Glider is a true game-changer for the actuarial models used byproperty and casualty insurers. Our technology should be part ofevery major Property & Casualty insurer's portfolio of tools toassist in making the underwriting process for catastrophes morereliable with real-time, accurate data. LRI is poised not only tohelp save lives and minimize property damage, we have the abilityto also revolutionize and transform the economics of the insuranceand reinsurance marketplace.