Review in Global Change Biology: near-ground temperature trends differ within a single landscape

A team led by Caroline Greiser of Stockholm University and its Bolin Centre for Climate Research, with 20 colleagues from Europe, South Africa and the United States, published a review in Global Change Biology. It examines how macroclimate change translates into the microclimate that organisms experience, focusing on near-surface air and soil temperature. wiley.comGlobal Change Biologywiley.comOpen source ↗ The review reports no new field measurements. It summarizes published work from atmospheric science and ecology and adds three simulations of real landscapes, run with models built by Ilya Maclean of the University of Exeter, which track air 5 cm above the ground.

The authors show that microclimate trends do not necessarily track regional ones. In a coastal patch in southwest England, 2 km on a side, a computer reconstruction of the hottest days in 1983–2015 gave regional warming of about 0.3°C. Within the patch, one spot warmed by 2.9°C and another cooled by 1.3°C.

Cloud cover, wind and snow

Changes in weather shift how sunlight reaching the ground is split between heating the surface and air, evaporating water, and storage in soil and plants. A 2022 study cited in the review found that daytime cloud cover over Europe fell by roughly 1.5–2% between 1980–1995 and 2000–2015. In the simulation of a landscape in the Spanish Pyrenees, doubled cloud cover, compared with halved, lowered near-ground temperatures on the hottest days by about 12.4°C on average, most strongly on open ground.

Snow mattered most on the coldest nights. A snow blanket raised minimum near-ground temperatures by about 3°C on average, roughly ten times the effect of wind or air humidity. wiley.comGlobal Change Biologywiley.comOpen source ↗

Vegetation and open questions

In the third simulation, a 1 km² patch of the Anatolian steppe in Turkey was filled with shrubs and trees from a nearby site. On the hottest days, near-ground temperatures fell by up to 16.6°C under tall, dense shrubs and trees, while open patches with shorter grass warmed by up to 6°C. In a mixed forest in New Hampshire, spring leaves emerged earlier in 1982–2019. The second simulation found that the extra April leaf cover changed the forest floor's minimum and maximum temperatures by only about 0.2°C each.

The review also notes that a single disturbance, such as drought, insect outbreak, fire or storm, can expose a forest floor to conditions equal to several decades of regional warming. In a tree-planting experiment in China, plots with 24 tree species had denser canopies than single-species plots, and the ground beneath was up to 4.4°C cooler.

Evidence on whether cool pockets warm more slowly is mixed. Forty-five years of records from a forested mountain watershed showed little difference in long-term warming between sites. In Austria, winter temperature inversions became 11% less frequent in 1961–2017, and low-lying sites warmed faster. wiley.comGlobal Change Biologywiley.comOpen source ↗ The authors argue that predictions of species ranges, built mostly on regional data, need to account for ground-level conditions, and point to growing long-term near-ground records for checking such simulations. The simulations changed one factor at a time while holding the others steady.