PERRY, Fla. — On a crisp, cloudless morning, Andy Jackson stood adjacent to a row of orange trees, his mind on the disease that has wiped out much of Florida’s citrus industry. He gazed intently at the weapon he had hired to defend his 25-acre grove against that scourge: a springer spaniel named Bello.

At first glance, Bello might look more cute than formidable. But watch her work — running lickety-split alongside rows of trees, occasionally sitting under one to signal she has sniffed out the disease — and you might arrive at the opposite conclusion.

Bello and her handler, Tyler Meck, work for F1 K9, a Florida firm that trains and provides dogs for an array of purposes, including the detection of explosives, drugs and agricultural disease. She and her keen-nosed canine colleagues represent an innovative, government-backed effort to counter the devastation caused by Huanglongbing, or HLB, a bacterium that prevents fruit from ripening.

Agricultural experts and analysts estimate that since it surfaced in about 2005, HLB has caused a 75 percent decline in the state’s $9 billion citrus industry and forced nearly 5,000 growers to abandon the business.

Like something out of a bad science-fiction flick, the culprit annihilating Florida’s citrus crops is a small flying insect, the psyllid, which acts as a highly effective HLB distributor. The insect feeds on the plant, drawing out its juices and the bacteria, and when it feeds again, it deposits the bacteria into the next plant.

“That psyllid is really a little flying hypodermic needle,” said Tim R. Gottwald, a plant epidemiologist with the U.S. Department of Agriculture.

As HLB continues to spread with deadly efficiency, some hope dogs will be the heroes to vanquish this crop-destroying foe. After all, canines have already demonstrated the ability to sniff out plant diseases, including citrus canker and plum pox virus, which infects peach, plum and cherry trees.

The USDA has spent the better part of 15 years studying dogs’ capability to sniff out HLB, and it has found that the animals do so with 99 percent accuracy, according to Gottwald. Nineteen dogs, all owned by F1 K9, now boast this expertise and have been deployed in Florida and California. Experts say the dogs aren’t likely to stem the tide of the citrus industry’s destruction — it may be too little, too late — but they do offer the fastest and most accurate diagnosis available.

Using a computer model to track the trajectory of the disease with and without canine assistance, Gottwald said he found that early detection by dogs, followed by removal of infected trees, can keep orchards profitable over a 10-year period. Relying on other detection methods — visual inspection and a laboratory technique — would result in orchards operating at a loss within one to two years, the model found.

Using dogs, Gottwald said, is “far, far superior.”

The citrus industry in south and central Florida, home to the bulk of the business and ground zero for the HLB crisis, has mostly been wiped out. “Out of Business” signs proliferate throughout the region.

Against this bleak backdrop, north Florida, where Jackson operates, seemed to offer hope of a promised land for a time. HLB was scarce there.

“We were told the psyllids weren’t going to come this far north,” Jackson said, gesturing toward a row of his satsuma orange trees. “Well, the psyllids never got the memo.”

Jackson is a relative newcomer to the citrus business. He started out growing pine trees, as do many in Perry, which sits in a county that calls itself the “Pine Tree Capital of the South.” He decided in 2015 to convert 25 acres of his family’s 260-acre farm to citrus, and he planted the following year.

By November 2018, Jackson had amassed a grove of 4,000 full-grown trees — and he was aware that HLB had invaded the northern section of the state. So he turned to F1 K9.

Bad news: The dogs detected HLB in 79 trees. “I went home,” Jackson recalled, “had a stiff drink, and pulled those trees out.”

In December, he booked F1 K9 for a return engagement. Huddled at a picnic table in a shady part of Jackson’s farm, he and Jerry Bishop, F1 K9’s director of training, examined a color-coded map of the crops.

The hues represented different citrus varieties: navel, hamlin, satsuma and so on. The map also indicated areas of possibly imperiled trees, the spots where Jackson wanted Bishop and Meck to focus the dogs’ efforts.

For about three hours, Bishop and Meck took turns guiding Bello and one of three other dogs — a German shepherd, a Belgian Malinois and a shepherd-Malinois mix — along those rows. No lollygagging was allowed: Pooch and handler zoomed from one end to the other. It takes two to three seconds for a dog to assess a tree, according to Bishop.

When the dogs sat under a tree, signaling that it smelled of disease, their handlers rewarded them with praise and a moment with a chew toy or tennis ball.

Bishop tied a ribbon on a branch of that tree to flag it as HLB-positive so Jackson could later remove it. Then the high-speed ritual resumed.

Bello and her canine comrades have been deployed on professional gigs for a little more than a year. And the speed with which they detect HLB is another way the dogs set themselves apart.

Gottwald, who pioneered the use of dogs for plant disease detection, has scrutinized and developed techniques for training the dogs to recognize HLB — and fast. “The dogs are detecting it months to years earlier than the two prevalent methodologies,” he said.

One of those methods is visual inspection, though once you observe signs of HLB, it’s typically too late to intervene. The second is a laboratory procedure known as a polymerase chain reaction test, or PCR. But it analyzes only pieces of tissue taken from a given tree, which may not be infected, even if other parts of the tree are.

“Whereas, when the dog sniffs the tree,” Gottwald said, “it’s analyzing it holistically, so it doesn’t make any difference where the infection is.”

The lab test is also slower and less accurate. “PCR never picked up more than 25 percent of the infected trees at any one time, while the dogs were hitting at over 99 percent,” Gottwald said.

How do the dogs do it?

Bishop said the dogs are initially trained just like those that detect other substances, such as drugs or explosives, and the best have particular traits. “We’re looking for high drive, toy drive and hunt drive,” he said. “As long as they have those things, you could almost train them for anything.”

But when it comes to specifics of HLB detection training, Bishop is circumspect. “That’s proprietary,” he said.

For Jackson, the how of it doesn’t matter. The quartet of canines that recently worked his farm made him even more of a believer.

The dogs flagged about 25 diseased trees, he said, making him likely to hire F1 K9 again next year: “It’s another tool in the drawer.”

Duncan Strauss is a longtime journalist and host of WMNF Tampa’s weekly radio program “Talking Animals.”

Read more: