Temperature and precipitation are the two most powerful forces shaping Earth’s biomes, but they are far from the only ones. Soil chemistry, mountain ranges, fire, human land use, and even the deep evolutionary history of a continent all help decide whether a given patch of ground becomes tropical forest, grassland, desert, or tundra. The interplay among these factors is what makes biome boundaries fuzzy rather than crisp and what makes the question more interesting than a two-word answer.
Temperature and Precipitation Come First
If you had to pick just two variables to predict which biome covers a particular stretch of land, temperature and precipitation would get you remarkably far. In biogeography, the combination of climatic variables that limit where a species or biome can exist is sometimes called a “bioclimatic envelope,” and temperature and precipitation are its most critical components in terrestrial ecosystems.1ScienceDirect. Relationship between precipitation and species distribution Hot and wet year-round? You get tropical rainforest. Hot with a long dry season? Savanna or tropical dry forest. Cold and dry? Tundra. Cold and moderately wet? Boreal forest. The pattern holds across continents.
What matters is not just the annual averages but the seasonality of temperature and rainfall. Strong seasonal swings in temperature and precipitation across latitude and altitude are what generate the diversity of biomes found in a single region, from lowland rainforest through montane forest to arid savanna and desert.2SpringerLink. Solar Energy, Temperature and Rainfall A place that receives 1,000 millimeters of rain spread evenly across the year supports a very different community of organisms than a place that receives the same total in a concentrated wet season followed by months of drought. Similarly, a region where winter temperatures regularly drop below freezing excludes entire lineages of trees, insects, and reptiles that thrive in the tropics.
The relative cover of different plant types illustrates how tightly vegetation tracks these seasonal patterns. In southern Africa’s Greater Cape Floristic Region, for example, rainfall totals during the wettest and driest quarters were the strongest predictors of shrub cover, while temperatures during the wet and dry seasons best predicted the balance between warm-season and cool-season grasses.3ScienceDirect. The importance of C3 and C4 grasses and CAM shrubs in the Greater Cape Floristic Region under contemporary and Last Glacial Maximum climates Shift those seasonal windows even modestly and the dominant plant types change, which in turn reshapes the whole ecological community built on top of them.
What Mountains Do to Climate
A mountain range can compress several biomes into a few kilometers of elevation change. As air rises over a slope, it cools and dumps moisture as rain or snow. By the time it descends the other side, it is drier and warmer, creating what ecologists call a rain shadow. In the Ecuadorian Andes, this effect is dramatic: one study documented precipitation dropping from about 990 millimeters on the eastern, windward side of the Chimborazo volcano to roughly 145 millimeters on the sheltered western side, turning lush vegetation into rain-shadow desert páramo within a short horizontal distance.4Arctic, Antarctic, and Alpine Research. Rain-Shadow in the High Andes of Ecuador Evidenced by Páramo Vegetation
Topography does not just redistribute rainfall. Altitude itself mimics the effect of moving toward the poles, lowering temperature by about 6°C for every 1,000 meters of elevation gain. That is why you can stand at the base of Mount Kilimanjaro in tropical savanna and hike to glaciers at the summit. Southern Africa offers another window into how steep physical gradients create a patchwork of biomes within a relatively small area, with sharp environmental discontinuities on land and in the adjacent ocean driving high biodiversity.5SpringerLink. Unique Southern African Terrestrial and Oceanic Biomes and Their Relation to Steep Environmental Gradients Coastlines, escarpments, river valleys, and volcanic peaks all generate local climatic conditions that diverge from the regional average, carving out pockets of habitat that might otherwise not exist at that latitude.
Soil as a Hidden Architect
Two neighboring patches of land can share the same temperature and rainfall regime yet support strikingly different vegetation because their soils differ. Soil texture, nutrient content, acidity, and depth all influence which plants can establish and thrive, which in turn shapes everything else in the food web. Across China’s forests, soil parameters like coarse fragment volume and organic carbon stocks were found to contribute meaningfully to the spatial distribution of different forest types, with organic carbon stock alone explaining about 9% of the variation in where conifer-broadleaf forests appeared.6Global Ecology and Conservation. Effects of soil properties on the spatial distribution of forest vegetation across China
The influence of soil is not equal across all biomes, though. In tropical regions, soil variables were far more strongly linked to vegetation structure and diversity in savannas than in forests.7Journal of Plant Ecology. The influence of soil on vegetation structure and plant diversity in different tropical savannic and forest habitats One reason is that savannas often occupy nutrient-poor or seasonally waterlogged soils where only certain grasses and scattered trees can cope, while tropical forests tend to be more constrained by climate and competition for light. Where soil conditions are extreme, such as very sandy substrates, ultramafic bedrock, or peat bogs, they can override climate entirely and create vegetation types that look nothing like the surrounding biome. Think of the white-sand campinas inside Amazonian rainforest or the fynbos shrublands of the Cape region that persist on acidic, nutrient-starved sands despite climatic conditions that could theoretically support denser woodland.
Fire and Other Disturbances
Fire is not just something that happens to biomes; in many cases it actively maintains them. Grasslands, savannas, and some open woodland types depend on periodic burning to suppress tree encroachment and recycle nutrients. Roughly 80% of fires worldwide occur in grasslands each year, making fire one of the defining processes in grassland dynamics.8PubMed Central. Global fire history of grassland biomes Remove fire from a tropical savanna, and within a few decades many of those landscapes would shift toward closed-canopy woodland. Add fire to a forest edge, and you can push the boundary back.
This means that fire, along with herbivory and wind disturbance, acts as a kind of ecological switch. In parts of Africa, South America, and Australia, climate alone would predict forest, but frequent burning keeps the canopy open and maintains grassland. The charcoal record shows that grassland burning increased through the early to mid-Holocene, peaking around 6,000 to 8,000 years ago, then declined, with human land management increasingly replacing lightning as the primary ignition source.8PubMed Central. Global fire history of grassland biomes Where fire regimes change, biome boundaries shift in response.
Vegetation Feeds Back on Climate
Most people think of biomes as products of climate, but the relationship runs in both directions. Vegetation regulates the energy, water, and carbon cycles across landscapes, consuming water through transpiration, pulling carbon dioxide out of the atmosphere, altering how rough or smooth the surface is to winds, and controlling how much of the sun’s energy goes into heating the air versus evaporating water.9PubMed Central. Vegetation-climate feedbacks across scales A dense tropical forest, for instance, recycles a large fraction of its own rainfall: water pulled from the soil by tree roots evaporates from leaves and falls again as rain downwind. Remove the forest, and regional rainfall drops, which can make it harder for the forest to recover. That self-reinforcing loop is one reason deforestation in the Amazon raises concern about a potential tipping point where the biome shifts irreversibly to a drier state.
In drier systems, the feedback works differently. Light-colored desert surfaces reflect more solar radiation and heat the atmosphere less, which suppresses convective rainfall and keeps the area dry. Plant cover in semi-arid zones darkens the surface, absorbs more energy, and can promote local rainfall. These feedbacks help explain why biome boundaries sometimes sit at stable positions for long stretches of time and then shift abruptly when a disturbance pushes conditions past a threshold.
Human Transformation of the Biosphere
No discussion of what determines biomes today can ignore the fact that humans have rewritten the map. Between 1700 and 2000, the terrestrial biosphere crossed a critical threshold from mostly wild to mostly anthropogenic, passing the 50% mark early in the 20th century. By 2000, less than 20% of the biosphere remained seminatural and only about a quarter was still wild, with the majority converted to agricultural and settled landscapes.10Global Ecology and Biogeography. Anthropogenic transformation of the biomes, 1700 to 2000 In practice, this means that temperate grasslands have become wheat fields, tropical forests have become cattle ranches, and wetlands have become rice paddies. The biome classification on a map still says “temperate broadleaf forest,” but the ground may be covered in corn and suburban housing.
Human influence extends beyond direct land conversion. Introducing invasive species, altering fire regimes, and channeling or damming rivers all reshape the ecological character of a region. Climate change, driven by greenhouse gas emissions, is perhaps the most far-reaching human impact on biome distribution, because it changes the temperature and precipitation patterns that set biome boundaries in the first place.
When Climate Shifts Move Biome Boundaries
Climate change does not just warm things up uniformly. It can eliminate entire temperature regimes that define biomes. Analysis of winter climate thresholds shows potential widespread losses of extreme cold below −20°C in Arctic, boreal, and cool temperate regions, and the possible disappearance of freezing temperatures and large decreases in snowfall in warm temperate and dryland areas. These threshold crossings carry the risk of important and potentially irreversible ecological changes.11Annual Reviews. Threshold Changes in Winter Temperature and Precipitation Drive Threshold Responses Across Nine Global Climate Zones and Associated Biomes For boreal forests, losing the deep cold that kills insect pests and limits competing tree species could fundamentally alter the biome’s character.
We can already see this playing out in mountain ecosystems. In Vermont’s Green Mountains, the boundary between northern hardwood forest and boreal spruce-fir forest shifted upslope by roughly 90 to 120 meters between 1962 and 2005, tracking a 1.1°C increase in regional annual temperature over the same period.12PubMed Central. A rapid upward shift of a forest ecotone during 40 years of warming in the Green Mountains of Vermont The boreal trees at the top of these mountains are running out of room. As lower-elevation hardwood species push upward, high-elevation forests may shrink or vanish from peaks that were once reliably cold enough to support them. The speed of this shift suggests that montane forests have less inertia against climate-driven range changes than previously assumed.
The Deep Past Still Shows
The biome map we see today is not just a reflection of current conditions. It carries the fingerprints of millions of years of continental drift, ice ages, and evolutionary history. Tectonic movements positioned continents at different latitudes, creating or closing ocean gateways that redirected currents and reshaped climate patterns. Evolutionary studies of tropical plant families show biogeographic patterns consistent with a Gondwanan origin, with lineages dispersing across continents as landmasses separated and reconnected, yet with few transitions between biomes even over tens of millions of years.13Global Ecology and Biogeography. Drift in the tropics: Phylogenetics and biogeographical patterns in Combretaceae That conservatism means the species available to fill ecological niches in any given region depend heavily on which lineages happened to be present when the current climate regime developed.
Ice ages are especially important for understanding present-day diversity patterns. During glacial periods, many species retreated to refugia, pockets of relatively stable climate where populations persisted while surrounding areas became uninhabitable. In the Mediterranean Basin, the locations of these refugia shaped plant diversity patterns that persist today, with refugial areas serving as reservoirs of unique genetic diversity.14Journal of Biogeography. Glacial refugia influence plant diversity patterns in the Mediterranean Basin The cumulative effects of these historical events, stretching back well before the last glacial period, mean that two regions with identical current climates can host very different assemblages of species and therefore look and function quite differently as biomes. Historical biogeography has moved beyond asking whether species spread actively or were carried passively by moving continents, evolving into a discipline that addresses how both ecological and historical factors construct biomes.15Evolution: Education and Outreach. Historical Biogeography: Evolution in Time and Space
How Microclimates Buffer the Big Picture
Even within a biome that is responding to climate change, small-scale conditions can slow or complicate the response. Dense forest canopies, for example, create their own internal climate: cooler, more humid, and more stable than open land nearby. Research across European temperate forests found that microclimatic effects from canopy closure can buffer plant communities against warming, attenuating the shift toward warm-adapted species that would otherwise be expected.16PubMed Central. Microclimate moderates plant responses to macroclimate warming In forests where canopy cover has thickened over recent decades, cooler ground-level temperatures have moderated the impact of rising regional temperatures on understory plants.
The flip side is equally telling. When canopy cover is lost through logging, storms, or drought-driven die-off, local temperatures spike and the buffering disappears. Increasing tree canopy cover reduces warming rates inside forests, but loss of that cover exposes the interior to heat that accelerates the mismatch between what the plant community is adapted to and what the climate now delivers.17PubMed. Forest microclimate dynamics drive plant responses to warming This means that forest management decisions, even at a local scale, can influence how quickly a biome’s composition shifts under changing climate. A well-managed forest with an intact canopy may resist biome-level change for decades longer than a fragmented one.
Light and Aquatic Biomes
On land, sunlight is usually abundant enough that temperature and water availability matter more. In aquatic systems, light itself becomes a major structuring force. The depth to which sunlight penetrates ocean water determines where photosynthesis can happen, which in turn controls where phytoplankton grow and how the rest of the marine food web is organized. Colored dissolved organic matter and detrital particles in the surface ocean absorb light and alter these dynamics. Modeling work has shown that accounting for this light absorption changes estimated surface chlorophyll concentrations by large margins, with coastal regions showing a 35% increase in surface chlorophyll under certain parameterizations while total depth-integrated biomass dropped by 18%.18Biogeosciences. Quantifying the biological impact of surface ocean light attenuation by colored detrital matter in an ESM using a new optical parameterization The largest relative shifts in productivity were found in equatorial ocean biomes, while subpolar and polar biomes showed the greatest declines in integrated biomass.
Freshwater biomes face analogous dynamics. Lake depth, water clarity, and nutrient loading all interact with light to determine whether a lake supports dense algal mats, clear-water ecosystems with submerged plants, or dark, oxygen-poor waters. The same lake can shift between these states depending on nutrient runoff from surrounding land use, a reminder that terrestrial and aquatic biomes are tightly coupled. Sediment from deforestation muddies rivers, reducing light and reshaping downstream aquatic communities. Fertilizer runoff from cropland feeds algal blooms that block light to deeper water. These connections mean that changes in one biome often cascade into adjacent aquatic systems.
Why Biome Boundaries Are Fuzzy
Textbooks often present biomes as discrete zones with clear borders, color-coded on a world map as though the boundary between temperate forest and grassland is as sharp as a state line. In reality, biome transitions tend to be gradual zones, called ecotones, where species from both sides overlap. The width of an ecotone depends on how steeply the underlying environmental gradients change. Where a mountain range creates a sudden rain shadow, the transition can be abrupt. Where temperature and rainfall shift gradually across a plain, the ecotone can stretch for hundreds of kilometers, with forest thinning into scattered trees and then grassland over a long distance.
These transitional areas are often the most sensitive to change. Because species there are already near the edge of their climatic tolerance, even small shifts in temperature or rainfall can tip the balance. Fire suppression in a savanna-forest ecotone lets trees advance into grassland. A string of drought years in a forest-steppe ecotone kills back trees and lets grasses take over. The ecotone acts as an early-warning system for the kinds of biome shifts that climate change may eventually bring to core areas.
All of these factors, from the broad physics of solar radiation hitting the planet at different angles to the acidity of a particular soil patch, work together in ways that make real biome boundaries messier and more interesting than any single factor could explain. No two patches of ground experience the identical combination of temperature regime, rainfall seasonality, soil type, fire history, evolutionary legacy, and human impact. That layered complexity is what gives Earth its patchwork of habitats and why predicting biome shifts under future climate scenarios remains one of ecology’s trickiest problems.