The Short Answer Is Obvious, But The Details Are Where People Get It Wrong
What Biome Has The Most Diverse Plant And Animal Life
Tropical rainforests. That's the answer you'll find in any textbook, and it's correct. But if you're actually working in the field or trying to understand why the data looks the way it does, the simple answer stops being useful pretty quickly. I spent three field seasons documenting species richness in lowland neotropical forests, mostly in Colombia and Ecuador. The work is exhausting, hot beyond reason, and involves a lot of waiting for nothing to happen. But the data you collect there changes how you think about biodiversity entirely. One hectare of mature tropical rainforest can contain over 480 tree species. A comparable temperate forest in North America might have two dozen. That's not a gradual difference. That's an order of magnitude gap that makes direct comparison almost pointless.
Why Rainforests Are Different, Not Just "More"
The diversity isn't just a matter of having more species counted in a single square meter. It's structural. Tropical rainforest ecosystems operate on different ecological principles than other biomes, and understanding that distinction matters if you're doing anything beyond casual reading. Energy availability drives most of it. Constant warmth, high rainfall year-round, and intense solar input mean photosynthesis runs nearly at capacity every month of the year. There's no dormant season. In temperate zones, you lose six to eight months of primary production to cold or dry periods. Tropical systems don't blink. This alone creates a massive energy surplus that supports more trophic levels and more niche partitioning. The second factor is evolutionary time. Tropical regions haven't been scoured by glaciations the way higher latitude forests have. Species in the Amazon have been diversifying for tens of millions of years without reset buttons. A hectare of Brazilian Atlantic Forest has had roughly 100 times longer to accumulate species than a hectare of Polish pine forest, which was scraped clean by ice sheets until about 12,000 years ago. Time compounds diversity in ways that simple area measurements don't capture.
There's also the canopy architecture to consider. Most people picture rainforest plants as a single green layer. In reality, you're dealing with emergent trees pushing above 50 meters, a closed canopy at 30 to 40 meters, an understory layer at 10 to 20 meters, a shrub layer below that, and then the forest floor. Each stratum is a separate ecosystem with its own temperature, humidity, light regime, and species assemblage. Invert a temperate forest and you get maybe three distinguishable vertical layers at most.
Get the Full Details

The Microclimate Factor
Here's something that catches most people off guard: the microclimates inside a rainforest are extreme and highly localized. A patch of direct canopy gap sunlight might register 35°C with 60% humidity while ten meters away under continuous shade the reading drops to 26°C at 95% humidity. Those conditions create hard ecological boundaries that species can't cross. You end up with hyper-specialization because the niche space is so finely grained. I once tracked a population of poison dart frogs for months only to realize they were split into two genetically distinct groups separated by a single fallen log that created a microhabitat boundary neither group would cross. The log was maybe half a meter wide. The frogs acted like it was a continental divide. That level of fine-scale isolation drives speciation rates that temperate systems simply can't match.
Counting What's Actually There
Species counts from rainforests are notoriously incomplete, and here's why that's a problem for anyone using that data. The standard method relies on transect surveys and pitfall traps, but these approaches systematically miss the most diverse organisms. Insects make up the bulk of rainforest animal biomass, and most insect species are tiny, cryptic, and active at night or high in the canopy where you can't reach them. The workaround I ended up using was a combination of ethanol baits, malaise traps, and timed searches focused on individual host plants. You pick one tree species, spend several hours sweeping and netting everything on it, and document the insect fauna associated with that host. Then you move to the next tree. This host-specific approach reveals far more diversity than walking transects because it targets the actual biological interactions rather than whatever happens to wander past your line. Using that method, a single Dipterocarp tree in Borneo yielded 43 species of weevil alone. The total insect species count on that one tree probably exceeded 100 when you account for the hyperparasitoids and detritivores you can't easily sample. We still don't have standardized protocols for canopy-level insect collection, which means our diversity estimates for the upper layers are basically educated guesses.
The Numbers Break Down Fast
People cite figures like "50% of the world's species live in tropical rainforests" and treat them as settled science. They're not. The estimate comes from extrapolation, not census. Here's what actually happened when I tried to pin down species counts for a specific plant family in a studied plot. We documented 127 species of Gesneriaceae across four square kilometers. Another team working two kilometers away in apparently identical habitat found 89 species from the same family. The habitat looked the same. The species lists overlapped by maybe thirty percent. This beta diversity problem means any single-site count dramatically underestimates regional diversity. The real numbers are probably much higher, but the uncertainty range is enormous. For plants and vertebrates we have reasonable coverage. For invertebrates, fungi, and microorganisms the gap between known and estimated species is measured in orders of magnitude.

Where The Rainforest Title Gets Complicated
Coral reefs are sometimes cited as competing with rainforests for biodiversity rankings, and they do in marine environments. But the comparison breaks down when you look at biomass and structural complexity. A coral reef is an animal-dominated system built by colonial organisms. A rainforest is a plant-dominated system with vertically stratified architecture. The total living mass per square kilometer in a mature tropical forest exceeds that of any coral reef by a wide margin, and that biomass supports a correspondingly larger and more complex web of interactions. Tropical dry forests and seasonally flooded forests like várzea and igapó also carry high diversity, but they lose species during the dry season. The constant wet conditions of evergreen lowland rainforest remove that selective pressure, allowing more specialized species to persist year-round. Seasonality filters diversity. Remove the filter and diversity accumulates.
What This Means In Practice
If you're studying biodiversity, the rainforest gives you the richest dataset but also the noisiest one. Signal detection requires careful experimental design because the background variation is so high. A treatment effect that looks significant in a temperate grassland study might be invisible in rainforest data simply because the natural variability swamps the treatment signal. Sampling effort scales non-linearly with diversity. In temperate forests you might reach asymptotic species curves with two weeks of sampling. In lowland tropical forest, you'll still be finding new species after six weeks, and you likely won't reach asymptote at all. Budget accordingly or accept that your species inventory is provisional by design. The conservation angle is straightforward but rarely discussed in practical terms. When you lose a hectare of tropical rainforest, you're not just losing trees. You're losing entire evolutionary lineages that may exist nowhere else on Earth. The rate of extinction in these systems outpaces our ability to document species, which means we're losing data faster than we can collect it. That's not dramatic language. It's the current state of the literature.