Now Blooming: Digital Models

Their approach was far different from that of DeFeo, whom they consulted and who admits he understands little of what they report. "I don't have a clue what they're saying," he said yesterday.

Neither would most people.

Dass and Brennan said they focused on computational intelligence and essentially tried to mimic the working of the human brain. This involved considering such things as "multiple-layered feed forward neural networks," they wrote in their paper, as well as "delta rule," "topology" and "Stochastic gradient method."

A neural network model, by the way, "is like the brain," Dass said. "You know how our human brain absorbs complex relationships? It's something very similar to that. You would train a neural network . . . like the brain, and then after a while, it would be able to . . . predict the actual phenomenon."

Fuzzy logic is another malleable brainlike data processing system that adjusts itself as it gets feedback, Dass said. And the famous equation of Swedish chemist Svante August Arrhenius calculates the speed of a chemical reaction based on temperature.

Complex as all this sounds, Dass and Brennan pointed out that computer modeling is widely used in Japan to predict the cherry blossom blooming period and in the United States to predict soybean flowering, corn yield, and aspects of tomato and lettuce growing.

The students started their project in January 2007 and observed the start of the annual blossom bloom. "They were quite striking, very beautiful," said Brennan, who had never seen the flowers before.

The two did not hazard a forecast but plugged in historical data about past blooms and associated weather conditions. Because they used previously recorded data and outcomes, they were able to see which models worked best. They found several models that were accurate to within a few days of past peak dates.

The students say some models, according to their calculations, came three days closer to the peak bloom date than DeFeo's predictions.

But DeFeo focuses more on a bloom range, and, anyway, "it's a crapshoot," he said, adding: "The trees will be in full bloom when the blossoms are fully open."

DeFeo said yesterday that his track record is good, though his prediction is subject to the whims of the weather. He fretted, for example, about a forecast for some stormy weather next week, which could strip the delicate blossoms.

"I missed it three years," he said. "All three years, they bloomed early on me. Two of those years, I missed it by five or six days; the other year, I missed it by one day."

You must have "intimate daily contact with your tree population, or any living thing, in order to understand it," he said. Computers "certainly have their use, but when they forecast the blossoms, I would never want to substitute a computer for going out and looking at the buds and seeing where they're at."

Dass and Brennan, meanwhile, are onto other subjects this year. But they did well in their class last year. Lu, the professor, said yesterday that both got an "A" in the course.