I don't remember whose idea it was to follow the tropicbird. Perhaps it was mine — a vain hope to chase the memories of my coastal childhood under the pretence of observing an odd visitor to the bay. Perhaps it was Hector's idea: Hector, who was afraid of nothing that the increasingly violent and unpredictable Gulf could produce, in the early years of that storm-wracked decade. Hector, whose catamaran, the Danaus, was rarely tethered in its slip at the university harbour for more than two days at a stretch. In the iterated dream, he turns his handsome face to smile at me, and releases the sail of the Danaus as the tropicbird burns white against the bruised horizon.

Credit: JACEY

The destruction of Gulf Coast communities, a consequence of anthropogenic climate change, began with Hurricane Katrina in 2005 and Hurricane Ike in 2008. By 2015, even the sceptics could not deny the dramatic increases in the annual number and strength of hurricanes in the Gulf of Mexico. Despite extensive improvements achieved during President Obama's third term, the New Orleans levees failed in 2023, and the resulting flood made the city uninhabitable. Five years later, only the tops of the high-rise buildings were visible above the waves, and Lake Pontchartrain, along with the crescent bend of the Mississippi River, had been swallowed by the Gulf. Galveston, island site of a bioterrorism research facility, was destroyed by Hurricane Omicron in 2032, the year I was born. By that time, the hurricanes inevitably killed so many people each year, that no one had the heart to give them familiar human names.

We met as undergraduates in Houston, both of us seeking degrees in environmental science. Hector, consumed with youthful wanderlust, was interested in how climate change and expansion of the Gulf affected the patterns of animal migrations, particularly for the monarch butterfly. One had to be energetic to engage in ecological research during those oil-impoverished years. Our older professors regaled us with tales of gas-powered vehicles, used routinely in field work during the early decades of the twenty-first century, and our eyes widened in disbelief. We knew only the immediate world, in which remote research stations were reached on horseback, and coastal sites accessed by kayak or sailboat. Our youthful restlessness and romantic notions of science drove us out into the field, to conduct studies of a coastal environment that changed rapidly with the climate. It would not be an exaggeration to label our university years as halcyon ones for environmental research: funding was available from federal agencies, and a number of solar-powered instruments allowed us to continue imaging and molecular investigations in the laboratory. On that October day, with our senior projects focused on climate influences on seabirds and butterflies, Hector and I were eager to begin mapping the impact of a Category 1 hurricane on our research populations. It was predicted to make landfall east of Houston, and its path lay within reach of our boats.

No traces of the Danaus, or of Hector, were ever discovered by the Coast Guard. Hurricane Laguz blew ashore east of Oak Island and Double Bayou, dissipating as it swirled northward, over the abandoned interstate highway. I tried to explain to the authorities who questioned me that Hector was an experienced sailor, familiar with the mood swings and sudden squalls of the Gulf, aware of the submerged buildings. He had been monitoring the migratory streams of monarchs for several years, and routinely collected data about wind speed and direction, currents, temperatures, tides and man-made obstacles. The hurricane coincided with the southward autumn journey of the monarchs, to the great trees in the mountains of Michoacán. I had followed Hector's catamaran in my kayak until the waves had threatened to swamp the craft. I mentioned neither the tropicbird nor my own routine excursions to collect data on the behaviour of this species and its strange association with hurricanes. Now, I am certain that the association is no accident, but it would have been foolhardy to mention my embryonic suspicions at the time.

A few months after Hector disappeared, I received notice that I had been accepted to a medical school in Houston. I graduated with honours in 2058, and entered a dual residency programme in neurosurgery and oncology. My professional expertise, and the work that consumes my waking hours, is the resection of brain tumours. Don't ask me to speculate about reasons for the increased incidence of nervous-system cancers in Gulf Coast communities. I leave it to the researchers to argue about elevated ultraviolet exposures, pollution from submerged refineries and oncogene vectors released from inundated laboratories. My job, technical rather than scientific, is to remove the tumours surgically, and to try to leave the patient with memories and cognitive functions intact.

Which brings this story, ironically, to my current situation: greying hair partially shaved, scalp scrubbed and prepped with anaesthetics, sedatives flowing through my veins. It occurred to me that if memories can be localized and preserved during surgery, they can also be destroyed. Quite recently, I managed to convince two of my colleagues to attempt to obliterate a few of my memories, using a probe that I designed. Memories are the subunits of dreams: they remain locked, like political prisoners of a repressive cortical regime, in the hippocampus by day, released only when the rational brain lets down its guard during sleep. I no longer desire to understand the evolutionary origins of tropicbirds, or to discover why they suddenly appear in the Gulf, dozens of them, in advance of the frequent hurricanes. The longevity, unique DNA sequences and unfathomably odd neuroanatomy of these seabirds no longer interest me.

I just want the dreams to stop.