Sorry, one of the images was not uploaded properly.

Here's the figure for the validation approach
🧵(6/27)

This is very different from our results with TSM 🤔

In fact, calculating TSM based on averaged temperature profiles hides critical tipping points for thermal stress (c,d)! ⚠️

Therefore, capturing daily temperature variation is key in assessing vulnerability risk!

🧵(17/27)

We did not find uniform latitudinal patterns in the predicted number of overheating events ⚠️

In the S. Hemisphere, tropical species encounter disproportionally more overheating events, while in the N. Hemisphere, non-tropical species are more susceptible (a,b)

🧵(16/27)

Interestingly, our models show that most amphibians will not overheat in aquatic microenvironments 💧. Only 11 species would be vulnerable under extreme climate warming scenarios.

This clearly demonstrates the importance of water bodies for amphibian thermoregulation⚠️

🧵(15/27)

We predicted that 104 species (2%) are already exposed to overheating events, and 4°C of global warming would push 7.5% of species beyond their thermal limits in terrestrial conditions (a)

In arboreal conditions (b), numbers are slightly lower, but we find similar patterns

🧵(14/27)

However, it does not mean that amphibians do not experience overheating events (when body temperatures exceed CTmax)!⚠️

In fact, extreme temperatures occasionally push many amphibians beyond their thermal limits, particularly under extreme warming scenarios😱

🧵(13/27)

If we first look at TSM, we find that it is always positive, but declines towards mid and low latitudes, which is consistent with previous evidence ✅

TSM is also slightly higher in aquatic (b) environments relative to terrestrial (b) and arboreal (c) habitats

🧵(12/27)

We also accounted for plasticity, and CTmax was adjusted daily to the temperatures experienced in the 7 days prior.

We then calculated 2 metrics: thermal safety margins (TSM, mean difference between CTmax and body temperatures), and the number of overheating events

🧵(11/27)

Behavioural thermoregulation is extremely important for amphibians, so we simulated different microhabitats.

We estimated body temperatures in both terrestrial 🪨, aquatic💧, and arboreal 🌱 microenvironments in shaded conditions (thermal refugia)

🧵(8/27)

Now that we know the CTmax of 5203 species of amphibians, we can assess their vulnerability to current and future temperatures 🌡️

To do so, we put @unsw.bsky.social's supercomputer to work and ran thousands of biophysical models using
#NicheMapR

🧵(7/27)

And this imputation approach works incredibly well!

Using a cross-validation approach, we found that we can predict thermal limits with surprisingly high accuracy 🙌

🧵(6/27)

This is where data imputation comes into play!💡

In this study, we developed a new approach to predict the thermal tolerance (CTmax) of data-deficient species🐸

We managed to retrieve the #CTmax of 60% of amphibians (>5000 species) and solve many geographic & taxonomic biases🙌

🧵(5/27)

In amphibians, knowledge on thermal tolerance is available for only 7.5% (⚠️) of species, and is biased towards temperate areas (tinyurl.com/49rcx6p9)

This is problematic because most amphibians inhabit the tropics where they are assumed to be the most vulnerable...

🧵(3/27)

How vulnerable are #amphibians to extreme heat? 🐸🌡️

Our paper in @nature.com shows that many amphibians are already overheating, and many more species will be impacted by climate warming globally.

See the thread below for a digest 🧵

Link to the paper: doi.org/10.1038/s415...

#Nature