Tropical forests are approaching critical temperature thresholds

In late 2023, Hydrosat’s Science Lead, Josh Fisher published a paper in Nature that made headlines around the world. “Tropical forests are approaching critical temperature thresholds” made the front page of Nature, and was covered in a diversity of media outlets ranging from CNN and the Los Angeles Times to Popular Mechanics and Breitbart.

The study was led by his long-time colleague, Chris Doughty, who deftly corralled hard-to-get data from across the Tropics. Chris has been looking at tropical forest temperatures for a while, though the origin of this paper came from a spontaneous conversation between Chris, Josh, and Greg Goldsmith at a NASA ECOSTRESS team meeting.

At JPL, Josh launched ECOSTRESS to the International Space Station as Science Lead, and soon thereafter a solicitation for proposals to use the data came out. Josh joined forces on Chris’s ECOSTRESS proposal. It was selected, they met up at the ECOSTRESS team meeting, and this is where the genesis of the conversation began to develop.

More than 150 years ago, Julius Sachs reported that leaves from different plant species could withstand temperatures of up to 50 °C (122 °F), but then would die if it got any hotter. For tropical plants, it’s even earlier — the critical temperature, or Tcrit, is more like 47 °C. Chris had been painstakingly measuring tropical canopy temperatures by precariously climbing trees and placing sensors on individual leaves, powered by car batteries carried on his back in the face of a canopy swaying in the wind and visits from the local monkeys. Chris had observed leaf temperatures exceeding Tcrit on individual leaves, but didn’t know if the occurrence was widespread.

There are many problems with remotely sensing tropical canopy temperatures, especially Tcrit. First, tropical forests are extremely biodiverse; if one part of the forest canopy is heating up, it doesn’t mean the whole forest is heating up because of different water access and use strategies, and structural differences. Consequently, we need high spatial resolution, high accuracy and high precision to detect those differences. Second, tropical forests are extremely cloudy, necessitating frequent imaging to have any chance of getting cloud-free images. The problem is that there is an engineering/cost tradeoff in the remote sensing world between high spatial and temporal resolution — historically you can’t have both, although Hydrosat is now smashing through that paradigm (Figure 4).

So, while Landsat could, in theory, give scientists that high spatial resolution, they rarely see anything when they want to in the Tropics because of its 16-day repeat. MODIS, on the other hand, gives great measurement cadence, but the km-scale resolution ends up blurring parts of the landscape with hot temperatures and cold temperatures together into the same pixel. ECOSTRESS moved the dial forward by getting high resolution (70 m) surface temperature at somewhat frequent cadence (~5 days) by being on the ISS and not going over the poles. The ISS orbit (aka precessing) also allows ECOSTRESS to pick up different times of day, which increases the probability of capturing peak temperatures. Crucially, ECOSTRESS can see the long-tail of the temperature distribution that MODIS smears, and significantly increases the probability of clear-sky acquisitions relative to Landsat. And, that’s how the team found that tropical forests all over the world were starting to exceed Tcrit, matching and expanding what they were seeing with the in situ sensors.

Unfortunately, the implications of their findings were even more concerning than the observations. They also evaluated some warming experiments, which showed that increases in air temperature (i.e., global warming) lead to non-linear increases in canopy temperature. These increases are within scenarios that could lead to the Tcrit tipping point within decades. However, tropical rainforests are extremely diverse and complex, the models and experiments are very simple, so this had to be taken into account.

While the results aren’t pretty reading, it is incredible that science can now detect such changes in the Tropics. This is something of a medical marvel for the health of the planet — early detection and action can help us avoid a larger disaster. As Josh noted in CNN, “It’s a little bit of a canary in the coal mine that we’re starting to see. You want to be able to detect something happening before it’s widespread.” In the paper, they highlight that “a combination of ambitious climate change mitigation goals and reduced deforestation can ensure that these important realms of carbon, water, and biodiversity stay below thermally critical thresholds.”

Temperature detection capabilities continue to improve in leaps and bounds. As important as ECOSTRESS was for this analysis, Hydrosat advances science significantly further. Hydrosat’s constellation of 16 TIR/VNIR satellites enables 50 m surface temperature with 10 m VNIR daily and globally (Figure 7). This is a game-changer in planetary health monitoring. Hydrosat’s IrriWatch platform enables management from monitoring: action from observation. Ecosystem managers, agriculturalists, and policymakers are now equipped with the most cutting-edge data to tackle climate challenges of feeding 9B people by 2050, managing forests for increasing wildfires, and tracking changes in our planet’s temperature as linked to climate mitigation policies. Hydrosat’s IrriWatch is used in over 60 countries and growing. The future is exciting, and the constellation couldn’t come soon enough.

This article, originally written by Hydrosat’s Science Lead, Josh Fisher,  can be found at the link here.