Microclimates: thermal shields against global warming for small herps

22 11 2017

Thermal microhabitats are often uncoupled from above-ground air temperatures. A study focused on small frogs and lizards from the Philippines demonstrates that the structural complexity of tropical forests hosts a diversity of microhabitats that can reduce the exposure of many cold-blooded animals to anthropogenic climate warming.

Luzon forest frogs

Reproductive pair of the Luzon forest frogs Platymantis luzonensis (upper left), a IUCN near-threatened species restricted to < 5000 km2 of habitat. Lower left: the yellow-stripped slender tree lizard Lipinia pulchella, a IUCN least-concerned species. Both species have body lengths < 6 cm, and are native to the tropical forests of the Philippines. Right panels, top to bottom: four microhabitats monitored by Scheffers et al. (2), namely ground vegetation, bird’s nest ferns, phytotelmata, and fallen leaves above ground level. Photos courtesy of Becca Brunner (Platymantis), Gernot Kunz (Lipinia), Stephen Zozaya (ground vegetation) and Brett Scheffers (remaining habitats).

If you have ever entered a cave or an old church, you will be familiar with its coolness even in the dog days of summer. At much finer scales, from centimetres to millimetres, this ‘cooling effect’ occurs in complex ecosystems such as those embodied by tropical forests. The fact is that the life cycle of many plant and animal species depends on the network of microhabitats (e.g., small crevices, burrows, holes) interwoven by vegetation structures, such as the leaves and roots of an orchid epiphyte hanging from a tree branch or the umbrella of leaves and branches of a thick bush.

Much modern biogeographical research addressing the effects of climate change on biodiversity is based on macroclimatic data of temperature and precipitation. Such approaches mostly ignore that microhabitats can warm up or cool down in a fashion different from that of local or regional climates, and so determine how species, particularly ectotherms, thermoregulate (1). To illustrate this phenomenon, Brett Scheffers et al. (2) measured the upper thermal limits (typically known as ‘critical thermal maxima’ or CTmax) of 15 species of frogs and lizards native to the tropical forest of Mount Banahaw, an active volcano on Luzon (The Philippines). The > 7000 islands of this archipelago harbour > 300 species of amphibians and reptiles (see video here), with > 100 occurring in Luzon (3).

CTmax is typically estimated in experimental settings as the body temperature at which an ectotherm, exposed to gradually increasing temperatures, loses motor coordination prior to enzyme degradation and, ultimately, death (4). Scheffers et al. (2) quantified how often the temperature of the microhabitat and surrounding habitat exceeded the CTmax of the study species. They used thermal sensors to record environmental temperatures every 20 minutes above and below ground, within and outside phytotelmata (small hollows in plants that fill with water), in bird’s nest ferns, on ground vegetation, and in tree canopies – from May to September 2011. They found that frog and lizard CTmax varied between 33 and 37 °C, far above the average temperature recorded in their microhabitats (~20-23 °C). All seems nice up to here. But what is ecologically relevant is not the average temperature, but the temperature extremes.


Contrasts of microhabitat temperature and thermal tolerance (CTmax) in 15 species of small frogs and lizards** from Mount Banahaw (Philippines). Empty circles represent single temperature records in different microhabitats***, and red squares are their average. The lethal zone (blue band) comprises the range between the maximum and minimum CTmax of the species set found in any one microhabitat. Individual records in the lethal zone only occurred above the soil level, and in ground vegetation and the tree canopy, and are predicted to occur in all habitats except below soil and inside phytotelmata under projected climate warming of 6 °C (blue bolded line).

Those days and hours of the day when temperatures peaked, Scheffers et al. showed that temperatures above ground, outside phytotelmata, and in tree canopies exceeded CTmax 10 to 30 times more frequently than below ground or within phytotelmata, bird’s nest ferns, or ground vegetation (2). Additionally, for a 1 °C increase in local temperatures, the target microhabitats heated up by only 0.1 (soil), 0.3 (ferns), 0.5 (ground vegetation), and 0.7 °C (phytotelmata).

We can confidently state that rain forest microhabitats buffer local thermal anomalies. Moreover, modelling of change in local temperatures, given the IPCC’s [Intergovernmental Panel on Climate Change] worst-scenario projections of anthropogenic climate warming by 6 °C by the end of this century (5), further indicated that the native small frogs and lizards of Mount Banahaw might only be able to survive future heat waves by sheltering mainly below ground or within phytotelmata, likely forcing them to change their thermoregulatory behaviour.

Landscape heterogeneity is a composite of variation in vegetation structure and topography (e.g., altitude, shading, exposure), and generates a spatial mosaic of thermal microhabitats. Within the same landscape, it is not rare to find spatial differences in air temperature at the same time of the day of up to 20 °C! (6, 7). In the face of the ongoing and predicted increase in the frequency of heat waves as a result of anthropogenic climate change (8, 9, 10, 11), small animals already rely, and will do so more strongly in the next few decades, on the availability of benign microhabitats and on their capacity to move from one microhabitat to another.

Not only that, air temperature and water availability are two faces of the same coin when it comes to understanding biodiversity responses to climate change. Thus, extreme temperatures come together with rain and air-humidity declines (12), which can also impact microhabitat quality. So a hole in a fallen tree can be thermally cosy for a frog, but inhospitable if waterless or too dry.

Overall, half of the tropical forests of the Philippines have been logged in the 20th Century (13), and South East Asia can claim the fame for the largest planetary rate of human-made forest destruction (14) – recently named ‘Navjot’s nightmare’, giving credit to the legacy of late conservation biologist Navjot Sodhi (15). In particular, large-scale logging and fires* in this region are destroying rainforests (and peatlands) at an unprecedented pace, adding toxic hazes to the problem of ecosystem destruction (16).

Because canopy temperatures in native forests are hotter than at ground levels (17), native forests, including their tallest and shortest trees and bushes, and the associated network of microhabitats, constitute a thermal shield against ongoing climate warming. But logging and fire, and the growing replacement of primary forests by secondary forests and agricultural lands, is wiping out that thermal shield for multiple species; in other words, there is a pronounced synergy in the ecological impacts driven by anthropogenic climate change and habitat loss (18, 19, 20).

One reason (one more!) for a pledge to conserve native forests everywhere.

*See here and here two NASA’s satellite images showing an everyday scene of multiple fires (red dots) and smoke clouds in Southeast Asia.

**Study species: Philippine bent-toed gecko Cyrtodactylus philippinicus (IUCN status = least concern); Green paddy frog Hylarana erythraea (least concern); Smooth-fingered narrow-mouthed frog Kaloula kalingensis (vulnerable); Yellow-stripped slender tree lizard Lipinia pulchella (least concern); Puddle frog Occidozyga laevis (least concern); Common forest tree frog Philautus surdus (least concern); Dumeril’s wrinkled ground frog Platymantis dorsalis (least concern); Rough-backed forest frog Platymantis corrugatus (least concern); Banahao forest frog Platymantis banahao (vulnerable); Luzon forest frog Platymantis luzonensis (near threatened); Luzon swamp frog Limnonectes woodworthi (least concern); Luzon frog Sanguirana luzonensis (near threatened); Black-sided sphenomorphus Sphenomorphus decipiens (least concern); [Common name not yet attributed] Sphenomorphus abdictus (least concern); Jagor’s sphenomorphus Sphenomorphus jagori (least concern)

***Microhabitats (data kindly provided by Brett Scheffers):  soil = 1324 records both below and above ground level; phytolmeta = 20135 records both inside and outside water-filled cavities; bird’s nest ferns = 60305 records; ground vegetation = 49021 records; tree canopy = 49021 records


by Salvador Herrando-Pérez & David R. Vieites

(with support of the British Ecological Society and the Spanish Ministry of Economy, Industry and Competitiveness) 


  1. Huey, R. B. et al. (2012) Predicting organismal vulnerability to climate warming: roles of behaviour, physiology and adaptation. Phil Trans R Soc B 367:1665-1679
  2. Scheffers, B. R. et al. (2014). Microhabitats reduce animal’s exposure to climate extremes. Glob Change Biol 20:495-503
  3. Brown, R. M. et al. (2013) The amphibians and reptiles of Luzon Island, Philippines, VIII: the herpetofauna of Cagayan and Isabela Provinces, northern Sierra Madre Mountain Range. ZooKeys 266
  4. Lutterschmidt, W. I. & Hutchison, V. H. (1997) The critical thermal maximum: history and critique. Can J Zool 75:1561-1574
  5. IPCC (2014). Climate Change 2014: Synthesis Report (IPCC, Geneva, Switzerland) Available at www.ipcc.ch/report/ar5
  6. Sears, M. W. et al. (2011) The world is not flat: defining relevant thermal landscapes in the context of climate change. Integr Comp Biol 51:666-675
  7. Suggitt, A. J. et al. (2011) Habitat microclimates drive fine-scale variation in extreme temperatures. Oikos 120:1-8
  8. Meehl, G. A. & Tebaldi, C. (2004) More intense, more frequent, and longer lasting heat waves in the 21st Century. Science 305:994-997
  9. Perkins, S. E. et al. (2012) Increasing frequency, intensity and duration of observed global heatwaves and warm spells. Geophys Res Lett 39:L20714
  10. Prein, A. F. et al. (2017). The future intensification of hourly precipitation extremes. Nat Clim Change 7:48-52
  11. Fischer, E. M. et al. (2013) Robust spatially aggregated projections of climate extremes. Nat Clim Change 3:1033-1038
  12. Fischer, E. M. & Knutti, R. (2013) Robust projections of combined humidity and temperature extremes. Nat Clim Change 3:126-130
  13. Lasco, R. D. et al. (2001) Secondary forests in the Philippines: formation and transformation in the 20th century. J Trop For Sci 13:652-670
  14. Sodhi, N. S. et al. (2004) Southeast Asian biodiversity: an impending disaster. Trends Ecol Evol 19:654-660
  15. Wilcove, D. S. et al. (2013) Navjot’s nightmare revisited: logging, agriculture, and biodiversity in Southeast Asia. Trends Ecol Evol 28:531-540
  16. Tacconi, L. (2016) Preventing fires and haze in Southeast Asia. Nature Climate Change 6:640
  17. Frey, S. J. K. et al. (2016) Spatial models reveal the microclimatic buffering capacity of old-growth forests. Sci Adv 2:e1501392
  18. Brook, B. W. et al. (2008). Synergies among extinction drivers under global change. Trends Ecol Evol 23:453-460
  19. Mantyka-pringle, C. S. et al. (2012) Interactions between climate and habitat loss effects on biodiversity: a systematic review and meta-analysis. Glob Change Biol 18:1239-1252
  20. Oliver, T. H. & Morecroft, M. D. (2014) Interactions between climate change and land use change on biodiversity: attribution problems, risks, and opportunities. Wiley Interdisc Rev: Clim Change 5:317-335



3 responses

31 07 2019
Nothing like a good forest | ConservationBytes.com

[…] blanketed by good shade reinforces the scientific evidence here. We have described in a different blog that arboreal microclimates (fallen trees, bark holes, epiphytes, leaf litter) act as thermal […]


12 11 2018
Why a (young) scientist should blog | ConservationBytes.com

[…] Microclimates: thermal shields against global warming for small herps (22/11/17,CC) / Microrefugia […]


22 11 2017
Nicolas L.

Thanks for this post. I just wanted to bring to your attention a paper that was published about a month ago in GCB according to which “Tropical forests are thermally buffered despite intensive selective logging” (http://onlinelibrary.wiley.com/doi/10.1111/gcb.13914/full). That could add to the discussion regarding this topic. Best.

Liked by 1 person

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: