Imagine waking up early, eager to take in the dazzling carpets of bright orange flowers in Antelope Valley's California Poppy Reserve. The Instagram posts promised a spectacle.
You drive to the reservation north of Los Angeles, but the hills are not full of color.
Journal. Flowering is over.
Thanks to AI and a local scientist, that disappointment will soon be a thing of the past.
This year, Steve Klosterman, a biologist working on natural climate solutions, released a “wildflower forecast,” powered by a deep learning model, satellite imagery, and weather data.
In a sense, Klosterman, of Santa Monica, developed the tool to meet his own needs.
Last spring, the Midwest transplant was longing to see some wildflowers. He assumed there was some online resource that offered predictions or leveraged satellite imagery.
“Surely there must be something,” he recalled thinking. “But there was nothing.”
There are tools. The state reserve operates a live camera focused on a strip of land. Theodore Payne, a California native plant nursery and education center, runs a wildflower hotline, where people can call and listen to weekly recorded reports on hot spots.
“These are all essential resources,” Klosterman said. “At the same time, they are limited.”
Klosterman is not green when it comes to plants. His PhD, at Harvard, focused on the timing of new leaves on trees in the spring and the color change in the fall.
For a class project, a team I was part of created a website that predicted those leaf changes in the Boston area. It was a success.
California poppies bloom in Lancaster, near the state nature reserve, in mid-March.
(Kayla Bartkowski / Los Angeles Times)
To create the poppy bloom predictor, Klosterman turned to AI initially developed for medical imaging. He has used it to analyze satellite images of the Antelope Valley.
The model scans 10 by 10 meter squares of land to determine if there are poppies by their telltale orange color. (Also identifies small yellow flowers called fields of gold.)
The model is based on satellite images, dating back nine years, along with past weather data.
Then use the current forecast and recent flower status to look into the future.
If the mercury is going to hit 100 degrees and the wind is picking up (and in previous years that caused the flowers to wilt), that will guide the prediction.
Right now, the model can forecast five days and is, as Klosterman says, “very much a work in progress.” It would be better, more powerful, if I had 100 years to learn.
As more data is collected, one day you will be able to forecast a week or two.
Right now, poppies are appearing on the preserve in the western Mojave Desert.
It rained all fall and into winter, and poppies need at least seven inches of rain to do well, said Lori Wear, preserve interpreter.
January snowfalls seem to take them to another level, but that didn't happen this season. “It's a good flowering, but not extraordinary,” he said.
Still, poppies, California's state flower, cover swaths of protected land.
“It almost looks like Cheeto dust,” he said, “like someone had Cheetos on their fingers and just spread them on the landscape.”
Poppies here typically peak in mid-April, but the variable weather in recent years has made it difficult to predict, he said. Klosterman believes this moment is likely the zenith.
Also blooming now: goldfields, purple grape soda lupins, and owl's clover. Wear described the latter, also purple, as a “short owl with little eyes looking at you and a small beak.”
An SUV drives among flowers near the reserve. “It almost looks like…someone had Cheetos on their fingers and just smeared them on the landscape,” said Lori Wear, a reservation interpreter.
(Kayla Bartkowski / Los Angeles Times)
On Sunday, Klosterman experienced the flowers himself, using his technology as a guide.
It offers predictions in two ways. The first is the amount of the valley (shown on a satellite image) covered in poppies and gold deposits, expressed as a percentage. The other is an overlay of orange and yellow spots on the ground.
The map showed a fairly high concentration of poppies near a stretch of Highway 138. He went there and, lo and behold, vibrant blooms were waiting for him. She sent proof: a smiling selfie in front of a sea of flowers.
Klosterman's tool can help answer possibly more complex questions than poppy or not, such as a more precise understanding of the conditions flowers need to thrive.
Experts know that rain is key, but it's more complicated than that.
Steve Klosterman takes a selfie in a California poppy field.
(Steve Klosterman)
Heavy rains can overwhelm invasive grasses, displacing flowers. In reality, natives tend to thrive after several years of drought, once invaders not adapted to the arid climate disappear. That's what led to an epic superbloom in 2017, Joan Dudney, assistant professor of forest ecology at UC Santa Barbara, he told The Times in 2024.
Klosterman wondered if the recent heat wave would dry them out. But his model did not prove it, nor did his trip. Therefore, it is possible that other factors play an important role in its persistence, such as day length.
The model could also shed light on what might happen to flowers as the climate warms. Will they migrate north? Will there be fewer flowers?
To solve it, Klosterman said a weather forecast could be invented and connected to higher temperatures.
For now, Klosterman's forecast is limited to the Antelope Valley. But if it expands to other areas and other types of flowers, it could help people like Karina Silva.
Silva woke up at 5 a.m. last Wednesday to travel from his home in Las Vegas to Death Valley National Park, hoping to beat the heat and crowds at the super bloom.
But several hours later, she and her husband, David, were still trying to find him.
The hillside behind her was dotted with desert golds, but the display fell short of the riotous eruption of blooms posted on social media. The superbloom ended in early March, according to park officials.
“I was thinking it was going to be an explosion of different colors,” Silva said on the side of the road overlooking Badwater Basin.
