Biodiversity is measured not by any single technique but through a toolkit of complementary methods, each designed to capture a different slice of life’s variety. Some approaches count species in a forest plot; others sequence DNA filtered from a cup of river water or analyze soundscapes recorded by sensors left in the canopy for months. The choice of method depends on what you want to know, the scale you care about, and the organisms you are tracking. Understanding these methods matters because how we measure biodiversity directly shapes what conservation targets we set and whether we notice declines before they become irreversible.
Species Counts and Diversity Indices
The most intuitive way to measure biodiversity is simply to count how many different species live in an area. Ecologists call this species richness, and it is the backbone of most biodiversity assessments. But a raw species count tells you nothing about how evenly those species are distributed. A forest with 50 tree species where one species makes up 90 percent of all individuals looks and functions very differently from a forest with 50 species in roughly equal numbers. That balance is called evenness, and together richness and evenness feed into mathematical diversity indices.
The most widely used indices include Shannon entropy and Simpson’s index. Both compress richness and evenness into a single number, but they weight rare versus common species differently. A recent analysis found no statistically significant correlation between richness and evenness across a broad sample of abundance distributions, meaning the two components carry largely independent information about a community. Evenness tended to have the strongest influence on diversity indices, though when diversity was expressed in “effective number” form, richness became the more influential component.1PubMed Central. Diversity analysis: Richness versus evenness The practical upshot is that relying on species counts alone can be misleading. Two sites with the same number of species may have very different diversity scores once you factor in how individuals are spread among those species.
Alpha, Beta, and Gamma Diversity
Biodiversity measurements change depending on the spatial scale you choose. Ecologists break this down into three levels. Alpha diversity describes the variety of species at a single local site, like a pond or a forest plot. Gamma diversity is the total diversity across a broader region, like an entire landscape or country. Beta diversity captures how communities differ from site to site within that region. If every pond in a watershed has the same 15 fish species, beta diversity is low. If each pond has a somewhat different mix, beta diversity is high, even if each individual pond is not especially species-rich.
This framework reveals patterns that would be invisible if you only measured biodiversity at one scale. Research mapping plant diversity across Australia at high spatial resolution demonstrated the practical value of estimating all three tiers simultaneously to capture the continent’s highly uneven diversity patterns.2PubMed Central. Estimating Alpha, Beta, and Gamma Diversity Through Deep Learning Work across landscapes in varying degrees of human modification found that converting relatively natural landscapes to a mix of natural and modified land actually increased beta diversity and maintained overall landscape-level diversity, even though average local-level diversity dropped. Only when landscapes were pushed toward complete transformation did both alpha and gamma diversity decline.3Ecography. Alpha, beta and gamma diversity in relatively natural, mixed and transformed landscape scenarios That finding is a useful reminder that human modification does not always reduce every dimension of biodiversity in the same direction.
Environmental DNA
One of the fastest-growing methods in biodiversity science involves collecting water or soil samples and sequencing the DNA that organisms have shed into the environment. Skin cells, mucus, feces, pollen, and decaying tissue all leave genetic traces. By extracting and amplifying that environmental DNA, researchers can identify which species are present without ever seeing them directly.
For aquatic ecosystems, eDNA metabarcoding has proven especially powerful. A study comparing eDNA water sampling against traditional field surveys for amphibians and fish found that eDNA had a detection probability of 0.97 for amphibians, compared with 0.58 for conventional surveys. For fish, the number of taxa detected via eDNA was higher than or equal to traditional methods at 89 percent of studied sites.4Molecular Ecology. Next‐generation monitoring of aquatic biodiversity using environmental DNA metabarcoding In terrestrial settings, soil eDNA is also gaining traction. A pilot study in Kenya’s Laikipia County used soil eDNA metabarcoding to assess both microbial and plant diversity across restoration sites, identifying drought-tolerant plant species and growth-promoting bacteria that would have been difficult to catalog through visual surveys alone.5Environmental DNA. Harnessing eDNA Metabarcoding to Monitor Species Diversity in Restoration Sites: Insights From Laikipia, Kenya
The technology is also beginning to reach back in time. Ancient eDNA preserved in marine sediments has been used to reconstruct fish communities and past marine ecosystems, in some cases revealing biodiversity patterns from millions of years ago.6PubMed Central. Ancient environmental DNA indicates limited human impact on marine biodiversity in pre-industrial Iceland These sediment archives effectively serve as a biological time capsule, giving scientists a baseline against which to compare modern biodiversity losses.
Remote Sensing and Satellite Imagery
You cannot send field ecologists into every square kilometer of rainforest or savanna, but satellites and airborne sensors can cover vast areas continuously. Remote sensing contributes to biodiversity measurement primarily by mapping habitat types, vegetation structure, and canopy chemistry, which serve as proxies for species diversity below.
The most informative results come from combining different types of sensor data. Hyperspectral imaging, which measures reflected light across hundreds of narrow wavelengths, captures subtle differences in leaf chemistry and plant species composition. LiDAR, which bounces laser pulses off surfaces, maps the three-dimensional structure of vegetation, revealing canopy height, layering, and gaps. Research in temperate forests found that models combining canopy structural diversity with spectral reflectance predicted taxonomic, functional, and phylogenetic tree diversity better than either data type alone, and that canopy structure was critical to those predictions.7Global Ecology and Biogeography. Towards mapping biodiversity from above: Can fusing lidar and hyperspectral remote sensing predict taxonomic, functional, and phylogenetic tree diversity in temperate forests? A study of Mediterranean Natura 2000 sites similarly concluded that merging fine-spectral-resolution hyperspectral data with LiDAR vegetation height information produced habitat maps with much higher accuracy than those from standard multispectral satellite images like Sentinel-2.8Ecological Indicators. Monitoring habitat diversity with PRISMA hyperspectral and lidar-derived data in Natura 2000 sites: Case study from a Mediterranean forest
Drones for Fine-Scale Wildlife Surveys
Satellites see the big picture but cannot spot individual animals. Drones fill that gap. Equipped with thermal cameras, they can detect warm-bodied animals hiding in tree canopies at night, a task that ground-based spotlighting struggles with in dense forest. A comparison of thermal drone surveys and ground-based spotlight surveys for nocturnal arboreal mammals in southeastern Australia found that both methods achieved similar detection probabilities, but drones consistently recorded more species and more individuals. Interestingly, drone-derived density estimates over larger survey areas were significantly lower than density figures extrapolated from smaller ground-survey plots, suggesting that ground surveys may overestimate animal densities when their results are scaled up.9Ecological Applications. Thermal drone surveys to detect arboreal fauna: Improving population estimates and threatened species monitoring Drones are particularly useful for threatened species monitoring where repeat visits on foot would disturb the animals being studied.
Camera Traps and Acoustic Monitoring
For ground-dwelling mammals, camera traps have become the workhorse field method. Motion-activated cameras placed along trails and at water holes generate millions of images, which are then analyzed to determine which species are present and how abundant they are. The real analytical power comes from pairing camera-trap data with occupancy models, which use patterns of detection and non-detection across sites to estimate the true probability that a species occupies an area, accounting for the fact that you sometimes miss species even when they are there.
Multi-species occupancy modeling has been applied in tropical forests in Laos, reserve forests in Bangladesh, and savanna landscapes in Botswana to estimate species richness and community responses to human disturbance.10Wildlife Biology. Multi‐species occupancy modeling of ground‐dwelling mammals in central Laos: a case study for monitoring in tropical forests11Journal of Applied Ecology. Using camera trapping and hierarchical occupancy modelling to evaluate the spatial ecology of an African mammal community These approaches are valuable because they produce defensible baselines that can be resurveyed years later to track population trends.
Passive acoustic monitoring works on a parallel principle but targets sound-producing organisms. Automated recorders capture bird songs, bat echolocation, frog calls, whale vocalizations, and insect choruses continuously. This method provides fine temporal resolution across multiple spatial scales and captures species that are difficult to observe visually. Researchers have highlighted its potential not only for biodiversity estimation and population trend analysis but also for tracking climate-driven shifts in the timing of breeding calls and changes in species distributions.12Functional Ecology. Passive acoustic monitoring provides a fresh perspective on fundamental ecological questions
Beyond Species Names
Counting species tells you what lives somewhere but says little about what those species do or how they are related to one another. That is why ecologists increasingly measure functional diversity and phylogenetic diversity alongside taxonomic lists.
Functional diversity focuses on the roles organisms play in an ecosystem, described by measurable traits like body size, feeding strategy, root depth, leaf thickness, or dispersal distance. Trait-based approaches have gained traction because they preserve more information about community structure than species names alone and can directly link community composition to ecosystem functioning.13PubMed Central. Exploring the impact of trait number and type on functional diversity metrics in real-world ecosystems Several indices exist to capture different facets of functional diversity. Functional richness measures how much of the total trait space a community fills. Functional dispersion measures how spread out species are around the community’s average trait values.14Ecology. A distance‐based framework for measuring functional diversity from multiple traits A newer combined index integrates four independent facets, including functional richness, biomass evenness, trait evenness, and dispersion, into a single geometric mean, aiming to give a more complete picture of how ecosystem function might respond to environmental change.15Methods in Ecology and Evolution. Measuring overall functional diversity by aggregating its multiple facets
Phylogenetic diversity takes a different angle, measuring how much evolutionary history is represented in a community. A set of species spanning many distantly related lineages captures more evolutionary heritage than the same number of closely related species. The foundational measure, often abbreviated PD, sums the total branch length on a phylogenetic tree connecting the species in a community, including the root.16PubMed Central. The role of the phylogenetic diversity measure, PD, in bio-informatics: getting the definition right The practical appeal is that phylogenetically diverse communities are thought to harbor a wider range of traits and adaptations, making them more resilient to disturbance.
Genetic Diversity Within Species
Most of the methods described so far operate at the species level or above. But biodiversity also includes variation within a single species, specifically genetic diversity. A population with high genetic diversity has more raw material for adaptation and is less vulnerable to inbreeding and disease. Two widely used metrics are expected heterozygosity, which estimates the probability that two randomly chosen gene copies at a given locus differ, and allelic richness, the number of distinct gene variants present after correcting for sample size.17Genetic Variation. The Sensitiveness of Expected Heterozygosity and Allelic Richness Estimates for Analyzing Population Genetic Diversity
These two metrics do not always tell the same story. A classic example comes from European beech trees, studied across 389 populations spanning the entire species range. Allelic richness was highest in southeastern Europe, where beech persisted during the last ice age, as you might expect. But heterozygosity was actually lower in those refugia and higher in recently colonized northern regions, producing a negative correlation between the two measures. The explanation involved bottlenecks during long-distance colonization events that eliminated rare alleles but shuffled common ones in ways that boosted heterozygosity.18PubMed Central. Diverging trends between heterozygosity and allelic richness during postglacial colonization in the European beech The lesson is that even within genetics, different measures of diversity can point in opposite directions, and relying on a single metric can mislead conservation decisions.
Citizen Science Platforms
A massive share of global biodiversity data now comes from non-professionals uploading observations to platforms like iNaturalist, eBird, and Observation.org. These records feed into the Global Biodiversity Information Facility (GBIF), which aggregates hundreds of millions of occurrence records used in research worldwide. The contribution is enormous, but the data come with systematic biases.
An analysis of major citizen science biodiversity platforms found that while GBIF, iNaturalist, and Observation.org nominally cover all life forms, birds are by far the most observed group, while fish, marine organisms, arthropods, and other invertebrates are dramatically underrepresented. None of the major platforms directly analyzed expertise variability among observers, and apart from eBird, none accommodated data on species absence or sampling effort.19Journal of Biogeography. Improving Online Citizen Science Platforms for Biodiversity Monitoring That last gap matters a great deal. Knowing that a species was not detected at a site, combined with information about how hard someone looked, is essential for distinguishing genuine absence from a missed detection. Without it, scientists struggle to use citizen science data for the same occupancy modeling approaches that make camera-trap studies so powerful.
Bioindicators and Shortcut Species
Measuring everything everywhere is not feasible, so ecologists sometimes focus on indicator species whose presence, absence, or abundance signals broader environmental conditions. Indicator species are used to monitor environmental changes, assess whether management actions are working, and provide early warning signals for ecological shifts.20Ecological Indicators. How do ecologists select and use indicator species to monitor ecological change? Insights from 14 years of publication in Ecological Indicators Certain freshwater invertebrates, for instance, are highly sensitive to pollution and their disappearance from a stream indicates degraded water quality long before chemical tests might flag a problem.
Microbial communities are emerging as another class of bioindicator. Many microbial taxa grow fast and respond quickly to environmental change, making them potentially efficient proxies for ecosystem health. If globally relevant microbial indicators can be developed, they could serve both as standalone monitoring tools and as complements to existing physical and chemical measurements.21PubMed. Microbial assemblages and bioindicators as proxies for ecosystem health status: potential and limitations The challenge is that microbial ecology is still building the baseline knowledge needed to distinguish normal variability from meaningful shifts, especially across different climate zones and habitat types.
Tracking Biodiversity at the Global Scale
Individual field surveys and genetic analyses measure biodiversity at the site or population level. Scaling up to continents and the entire planet requires composite indicators. The Biodiversity Intactness Index (BII) estimates how much of a region’s original species abundance remains relative to a pre-industrial baseline. An analysis of BII trends from 2000 to 2020 estimated the global mean intactness at about 76 percent, with a slow downward trend of roughly 0.3 percent per decade. Africa experienced the steepest decline at about 1.4 percent per decade, while Europe showed a modest increase of around 1.0 percent per decade, likely driven by reforestation and conservation policy in some countries.22PubMed Central. Decline in global biodiversity intactness over the past two decades
These global indices inevitably smooth over enormous regional variation, and they rely on models and assumptions about what a “pristine” baseline looks like. Still, they are the best tool available for tracking whether international targets, such as the 30-by-30 goal of protecting 30 percent of land and sea by 2030, are on track. Some emerging frameworks are even trying to translate biodiversity measurements into economic units through biodiversity credits, defined as standardized, quantifiable units of measurable conservation outcomes, to channel restoration funding toward the most effective interventions.23Earth Critical Zone. Towards 30×30 biodiversity targets: Potential contributions of ecological restoration based on biodiversity credit
The Life Underfoot That Rarely Gets Counted
Most biodiversity surveys focus on what lives above ground and in open water. Soil is routinely overlooked, despite harboring an extraordinary density of organisms. A single handful of forest soil can contain thousands of species of bacteria, fungi, mites, springtails, and nematodes. Measuring this belowground diversity is technically demanding. Extracting microarthropods, for example, requires specialized devices that use heat gradients or chemical flotation to drive tiny animals out of soil cores. A comprehensive bibliography of methods for microarthropod extraction cataloged 378 publications spanning more than a century of methodological development, reflecting both the difficulty and the ongoing innovation in this field.24Soil Organisms. Methods for the extraction of microarthropods from soil: A bibliography and guide to the literature Soil eDNA, as mentioned in the Kenyan restoration study, is beginning to open this hidden world more efficiently, but belowground biodiversity remains one of the largest blind spots in global monitoring.
The Taxonomic Bottleneck
Every measurement method described here eventually runs into the same problem: someone has to know what the species are. Taxonomy, the science of identifying, naming, and classifying organisms, underpins the entire enterprise of biodiversity measurement. Yet the global workforce of trained professional taxonomists has been shrinking for decades. Many species, particularly invertebrates, fungi, and microorganisms, have never been formally described, meaning they literally have no name. Estimates of the total number of species on Earth range from roughly 8 million to over 10 million, but only about 1.5 to 2 million have been scientifically described. The only way to increase the pace of a well-documented inventory of the world’s biodiversity is to increase the number of professional taxonomists, and that requires institutional investment that has been declining rather than growing.25Europe PMC. The unappreciated, fundamentally analytical nature of taxonomy and the implications for the inventory of biodiversity
DNA-based methods have partially compensated for this gap by allowing researchers to detect and distinguish species without having to identify them morphologically in the field. But sequences without a name still need to be matched to reference databases, and those databases are only as complete as the taxonomic work that built them. Cryptic species, organisms that look identical to the naked eye but are genetically distinct, further complicate matters. In some invertebrate groups, molecular studies have revealed two or three cryptic species hiding under a single name, instantly revising local diversity estimates upward. Closing this bottleneck is not glamorous work, but it is the foundation on which all other biodiversity measurements rest.