In the past two hurricane seasons, record-breaking floods have engulfed our coastal zones in the Carolinas and Texas as storms have drawn more water and grown larger from rapidly warming oceans.

As the climate system continues to warm, we will need better prediction systems so we can prepare vulnerable coastal areas for bigger, wetter and faster-strengthening hurricanes. Hurricane season is just six weeks away.

Recent studies confirm that warming of the world’s oceans is taking place faster than previously estimated — as much as 40 percent faster than the United Nations estimated in 2015.

Research confirms that roughly 93 percent of the warming from man-made greenhouse gases is going into the world’s oceans. About two-thirds is absorbed in the ocean’s top 700 meters, noted Zeke Hausfather, a climate scientist at Berkeley Earth. This is the layer from which hurricanes draw much of their energy.

In a warmer, more volatile world, accurate weather predictions will require near-real-time data, not only from the atmosphere, but from regional variations in ocean warming, according to Kevin Trenberth, a climate scientist at the National Center for Atmospheric Research.

“The extra heat from increasing gases from carbon dioxide doesn’t necessarily go into the ocean uniformly,” Trenberth said in a recent interview. “There are increasing indications that the regional manifestations of the ocean heat content — where those spots are warmer — actually matters for subsequent climate over the next six months or so.”

Future improvements in hurricane track and intensity forecasts may rely on increasingly coupled models that combine measurements of the atmosphere with regional differences in surface and upper ocean heat content, in something close to real time.

“This is an important area of research, and I don’t think it’s being adequately capitalized at the current time,” Trenberth said, pointing out Hurricane Harvey tapped record levels of ocean heat content in the Gulf of Mexico before dumping more than 60 inches of rain near Houston.

Similarly, in 2018, Hurricane Florence tapped a regional hot spot of water, where natural variability combined with a warming climate, to fuel a vast swath of historic rains across the Carolinas.

But warm temperatures at the sea surface, by themselves, were insufficient to fuel the storm.

“If there’s not the ocean heat content below the surface to help sustain that, then that high sea-surface temperature will peter out fairly quickly,” Trenberth said. “This is one of the characteristics of Florence and Harvey. Although they took a lot of heat out of the ocean, there was still plenty more heat there, prolonging the lifetime of the storm.”

Like vast atmospheric washing machines, both Harvey and Florence were able to recirculate a continuous supply of deep-ocean warmth, jet fuel for hurricanes, back up to the surface — prolonging intensity and extreme rainfall rates, with catastrophic results.

The longer lifetime and sustained intensity driven by higher ocean content are what energizes what’s known as their eyewall replacement cycles. The eyewall is the zone of towering thunderstorms surrounding the hurricane’s calm center. It’s here that you find the hurricane’s strongest winds.

During an extreme hurricane, spiral arm bands wrap around the center so intensely that the storm forms a new (outer) eyewall, cutting off moisture flow to the old eyewall, which begins to collapse. The result: The storm temporarily weakens in strength (as Katrina did just at landfall in 2005) but expands in size.

Given high ocean heat content, the hurricane can reintensify during eyewall replacement. This happened with Hurricane Irma five times, and Trenberth believes these same dynamics were witnessed with Florence. This is how storms grow bigger, more intense and longer-lived, fueled by a steady supply of ocean heat.

Stating the obvious: For weather forecasters to have any chance of accurately predicting hurricane eyewall replacement cycles, which affects storm size, intensity and severity of the subsequent storm surge, access to reliable temperature data in the upper oceans may be key.

Hurricanes are nature’s quirky, automatic pressure-relief valves, transporting excess heat and moisture away from the tropics toward the poles, leaving behind trails of cooler water. But current weather models can’t capture real-time changes in surface and deep-ocean heating.

Trenberth speculates the latest high-resolution weather models overintensify tropical systems that assume a constant sea-surface temperature, not fully taking into account their cold wake. Accurate prediction of hurricane eyewall replacement cycles and intensification may require near-real-time, high-resolution data from the upper oceans and Gulf of Mexico.

“The European Center for Medium-Range Weather Forecasts is probably leading in terms of doing coupled weather forecasts,” Trenberth admitted, adding that the U.S. National Centers for Environmental Prediction “has been playing with this and making some progress.”

So, the challenge remains: How can we accurately predict the future when we can’t get an accurate snapshot of what’s happening right now, both atmospheric and oceanic?

Data from the Argo network of nearly 4,000 drifting floats in the world’s oceans could be programmed to provide deep-sea temperature data more frequently, close to real time during hurricane scenarios. Right now, Argo data updates roughly every 10 days.

“In places like the gulf and East Coast, they could program those to come up and down more frequently,” Trenberth speculated. “The result would be a more comprehensive understanding of upper-ocean water temperature profiles, giving forecasters the ability to gauge whether a hurricane’s cold wake may be resupplied with additional warming at the surface, thus prolonging the storm.”

Accurate prediction of increasingly powerful (and wet) hurricanes requires increasingly sophisticated and coupled models combining atmospheric measurements with regional hot spots of unusually warm and deep-ocean water. Hurricanes don’t follow identical tracks, because the next hurricane can “feel” where a previous hurricane has been. Trenberth’s take: “Until we properly put those kinds of factors into our models, we’re not going to quite get those tracks right.”

But regional hot spots of unusually warm, deep water may have implications beyond fueling hurricane strengthening and rainfall prediction. Nontropical cyclones, the bread-and-butter low-pressure systems that parade across our daily weather maps, also may be affected, especially the redevelopment of East Coast storms, feeding on tropical warmth and moisture.

“This can lead to some intensification, but it’s also apt to lead to development out ahead of the storm, so it affects the actual track of the storm,” Trenberth said. “It may also lead to some increase in the size of the storm.”

Weather data and high-resolution mapping are more accessible; weather model physics continues to advance and improve, but Hurricanes Harvey, Florence and Michael are reminders that the most accurate forecasts of track, intensity and rainfall require extensive, real-time analyses of ocean heat content, and not just at the surface. With rising carbon dioxide levels in the atmosphere, warming oceans fueling increasingly potent hurricanes makes real-time, deep-ocean data a necessity, not a luxury.

Tropical researchers are making great strides, but Hurricane Michael’s remarkable, last-minute intensification into a catastrophic 155-mph monster, just 2 mph shy of Category 5 strength, before leveling parts of the Florida Panhandle in October, underscores the urgency of removing biases in coupled air-sea models. Such advancements will provide meteorologists the tools necessary to improve tropical cyclone track and intensity forecasts — and save lives.

Paul Douglas is co-author of “Caring for Creation: The Evangelical’s Guide to Climate Change and a Healthy Environment” and co-founder and senior meteorologist at AerisWeather, located in the Twin Cities of Minnesota.