Earth’s surface has been photographed from orbit many times over, but satellite images and genuine exploration are very different things. Depending on how you define “explored,” somewhere between a fifth and a quarter of the ocean floor has been mapped to modern standards, large portions of the terrestrial subsurface have never been sampled, and the majority of species sharing the planet with us have not even been named. The question resists a single tidy percentage because land, ocean, underground, and atmosphere each present their own vast frontiers, and the gaps are far larger than most people assume.
The Ocean Floor Is the Biggest Blind Spot
Oceans cover about 71 percent of Earth’s surface, and for decades the commonly cited figure was that only about five percent of the seafloor had been mapped in any meaningful detail. That number has improved thanks to modern multibeam sonar campaigns and the international Seabed 2030 initiative, which aims to map the entire ocean floor by the end of this decade. Even with that progress, roughly three-quarters of the seabed has still not been surveyed at resolutions fine enough to reveal features like small seamounts, hydrothermal vents, or submarine canyons. By comparison, we have higher-resolution topographic maps of Mars and the Moon than of most of our own ocean floor.
The deep ocean is not just unmapped but largely unvisited. Submersible dives have reached extreme depths, including the Challenger Deep in the Mariana Trench at around 10,908 meters, generating video of undisturbed deep-sea communities at hadal and abyssal depths.1Deep Sea Research Part I: Oceanographic Research Papers. Submersible- and lander-observed community patterns in the Mariana and New Britain Trenches But these visits are pinpricks. The total area of seafloor that has been directly observed by human eyes or cameras is vanishingly small relative to the roughly 360 million square kilometers of ocean bottom. Most of what we know about deep-sea biology comes from scattered trawls, lander deployments, and remotely operated vehicle transects that have sampled only tiny fractions of the available habitat.
What Satellites Can and Cannot Tell Us
From space, every square meter of Earth’s land surface has been imaged, many times over and at increasingly fine resolution. This gives the impression that terrestrial exploration is essentially complete. In a cartographic sense it is: we know the shape of every continent, the course of every major river, the outline of every mountain range. But “explored” in any ecological, geological, or biological sense goes far beyond what a camera in orbit can reveal.
Satellite imagery is excellent for tracking large-scale patterns like deforestation, urban expansion, and ice-sheet retreat. It can identify ecological states such as eroded landscapes by analyzing the spatial configuration of bare ground patches.2Open Journal of Ecology. UAV and Satellite-Based Sensing to Map Ecological States at the Landscape Scale But remote sensing cannot catalog the species living in a forest, characterize the microbial communities in a soil sample, or detect cave systems beneath the surface. The distinction matters because it means that “seen from above” and “explored” are measuring fundamentally different things. We have seen all of Earth’s land surface. We have explored a much smaller share of it.
Terrestrial Wilderness and Low-Impact Landscapes
One way to gauge how much of the land surface humans have meaningfully investigated is to look at how much of it we have barely touched. Multiple independent mapping efforts agree that roughly 20 to 34 percent of Earth’s terrestrial area experiences very low human influence, and about half the land surface falls into the broader “low influence” category.3PubMed Central. Global human influence maps reveal clear opportunities in conserving Earth’s remaining intact terrestrial ecosystems That includes vast tracts of boreal forest, tundra, desert interior, and tropical rainforest where scientific surveys have been sparse or nonexistent.
These low-influence zones are not evenly distributed. Much of the remaining wilderness sits in places that are difficult or expensive to reach: the interior of the Amazon, the Congolese rainforest, the Siberian taiga, remote Pacific islands, and the high Arctic. Accessibility drives exploration, so the places we know least about are systematically the ones that are hardest to get to. That creates a bias in our knowledge of biodiversity, soil chemistry, and even basic topography that no amount of satellite coverage can fully correct.
Under the Ice Sheets
Antarctica alone covers about 14 million square kilometers, and nearly all of it sits beneath an ice sheet averaging over two kilometers thick. The landscape hidden beneath that ice remains one of the least understood parts of Earth’s surface. Until recently, maps of subglacial topography relied on interpolation between unevenly spaced radar survey lines, which produced unrealistically smooth terrain models and missed many features entirely.4Journal of Geophysical Research: Earth Surface. Antarctic Topographic Realizations and Geostatistical Modeling Used to Map Subglacial Lakes
New approaches using high-resolution satellite measurements of the ice surface have started to change that picture. Because ice flow over bedrock obstacles creates subtle signatures on the surface above, researchers have used the physics of ice flow to work backward and produce continent-scale maps of subglacial landforms at scales of 2 to 30 kilometers.5PubMed. Complex mesoscale landscapes beneath Antarctica mapped from space These maps reveal a subglacial world of mountain ranges, plains, and valleys that is far more rugged and complex than older models suggested.
One of the more surprising findings beneath the ice is the sheer number of subglacial lakes. A global inventory counts 773 of them, with 675 in Antarctica alone, including 59 that were newly identified at the time of the review. Greenland adds another 64, and a handful exist beneath smaller ice caps in Iceland, Canada, and elsewhere.6Nature Reviews Earth & Environment. Subglacial lakes and their changing role in a warming climate These lakes are not just geological curiosities. They serve as habitats for microbial life, influence how the ice above them flows, and act as repositories of ancient climate information. Only a handful have ever been directly sampled.
Hydrothermal Vents and the Unvisited Seafloor
Hydrothermal vents, where superheated water laden with minerals spews from the ocean floor, support some of Earth’s most unusual ecosystems. Entire food webs thrive on chemical energy rather than sunlight at these sites. Yet after nearly four decades of searching, only about one-third of the global oceanic spreading ridge system has been surveyed for hydrothermal activity.7Deep Sea Research Part II: Topical Studies in Oceanography. Where are the undiscovered hydrothermal vents on oceanic spreading ridges? The remaining two-thirds have not been examined at all.
Of the vent fields that are cataloged, the picture is still incomplete. A global database lists 664 known deep hydrothermal vent fields, but only 270 of those have been confirmed as active ecosystems through direct observation of live organisms and hydrothermal fluids. Another 336 are inferred to be active but have never actually been visited.8npj Ocean Sustainability. Towards a global strategy for the conservation of deep-sea active hydrothermal vents In other words, more than half the vent sites we think exist have never been ground-truthed. Given that two-thirds of the ridge system has not been surveyed, the total number of undiscovered vents could be substantially higher. These are not marginal habitats. Vent ecosystems support unique species assemblages found nowhere else, and each unvisited site could harbor organisms new to science.
The Twilight Zone Below the Surface Waters
Between the sunlit upper ocean and the pitch-dark abyss lies a vast layer called the mesopelagic zone, roughly 200 to 1,000 meters deep. Often called the ocean’s twilight zone, it is thought to contain the largest fish biomass on the planet, yet it remains poorly surveyed. Mesopelagic organisms form dense aggregations known as the deep scattering layer, typically concentrated between about 400 and 600 meters during the day and migrating toward the surface at night, making the daily vertical migration one of the largest movements of animal biomass on Earth.9Regional Studies in Marine Science. Vertical and spatial distributions of the deep scattering layer in the Philippine Sea, West Pacific Ocean
Despite the ecological importance of this zone, basic questions about it remain unanswered. How many fish and invertebrates actually live there? How do their populations fluctuate with climate? Estimates of mesopelagic biomass vary by an order of magnitude because we lack consistent observation methods. New autonomous tools, such as profiling floats equipped with echosounders, are being developed specifically to fill this gap by providing vertically resolved estimates of community abundance.10PubMed Central. Sonophore enables autonomous observation of micronekton communities in the ocean twilight zone For now, the twilight zone remains one of the largest habitats on the planet that we have barely begun to characterize.
Underground and the Deep Biosphere
The subsurface represents another enormous frontier. Below Earth’s surface, in rock fractures, aquifers, and cave systems, there exists a biosphere whose depth limit has not been determined. Microorganisms have been found kilometers below the surface in boreholes and deep mines, but sampling is constrained by the difficulty and expense of drilling. The deep biosphere may contain a significant fraction of all microbial life on the planet, yet access remains extremely limited.11PubMed. Detection of the deep biosphere in metamorphic rocks from the Chinese continental scientific drilling
Even relatively shallow underground environments hold surprises. Cave systems around the world remain incompletely mapped, and many are known only because of chance discoveries during mining or construction. In the Nullarbor Plain of Australia, for example, speleothem dating has revealed that cave formation and mineral growth in the shallow carbonate rock occurred primarily during the Pliocene epoch, spanning several million years, with individual caves supporting growth over similarly long timescales.12Nature. The antiquity of Nullarbor speleothems and implications for karst palaeoclimate archives These caves serve as archives of past climate and hydrology, but only a fraction of the Nullarbor’s cave network has been documented, and the same is true of karst landscapes worldwide.
Soil adds another layer of hidden complexity. The microbial communities in soil are staggeringly diverse, and a large portion of them have never been cultured in a laboratory or genetically characterized. Researchers refer to these uncultivated communities as soil microbial “dark matter,” borrowing the term from astronomy to convey the scale of what remains unknown.13Earth Critical Zone. Exploring and utilizing soil Microbiota: From dark to light The organisms that make up this dark matter are not rare oddities. They include lineages that play functional roles in nutrient cycling, plant health, and carbon storage but have so far resisted laboratory study.
The Species We Have Not Named Yet
Perhaps the most concrete measure of how much remains unexplored is the species gap. After more than 250 years of systematic taxonomy and over 1.2 million species cataloged in central databases, an estimated 86 percent of existing species on Earth and 91 percent of species in the ocean still await formal scientific description.14PLoS Biology. How Many Species Are There on Earth and in the Ocean? That means for every species we have identified and named, roughly six more are out there unrecognized.
The gap is not distributed evenly across the tree of life. Large mammals and birds are relatively well cataloged, which gives a misleading impression of completeness. The vast majority of undescribed species are invertebrates, fungi, and microorganisms. Insects alone likely account for millions of unnamed species. The marine realm is especially undersampled: the deep sea, coral reef crevices, and open ocean habitats all harbor communities that have rarely or never been surveyed. Even in well-studied ecosystems, new species regularly turn up when researchers apply modern genetic tools to specimens that looked identical under a microscope.
The ocean’s viral diversity illustrates the point at a microscopic scale. Environmental sampling reveals enormous numbers of viral sequences that cannot be matched to anything in existing databases, a phenomenon researchers call “viral dark matter.” Metaproteomic techniques applied to ocean samples have identified thousands of virion-associated proteins, over half of which were newly functionally annotated.15PubMed Central. Illuminating structural proteins in viral “dark matter” with metaproteomics Mining of publicly available microbial genome data has uncovered over 12,000 high-confidence viral genomes linked to their hosts, augmenting existing datasets roughly tenfold and providing the first viral sequences for 13 bacterial phyla.16eLife. Viral dark matter and virus–host interactions resolved from publicly available microbial genomes Every new sampling effort seems to reveal that the unseen portion of life is larger than previously estimated.
Forest Canopies and the Air Above Us
Some of the least explored habitats on Earth are not remote or deep but simply hard to reach in the right way. Tropical forest canopies form vertically layered ecosystems with distinct microclimates and microhabitats at different heights. They are among the most species-rich environments on the planet, yet many biodiversity patterns and food webs within canopies remain poorly documented.17PubMed. Forests and Their Canopies: Achievements and Horizons in Canopy Science The difficulty is logistical: collecting standardized data across a sufficient diversity of vertical structures, replicating those observations across seasons and between sites, and doing it all while suspended 30 to 50 meters above the ground. Vertical microclimatic gradients structure arboreal biodiversity across many groups of organisms, particularly in tropical forests, but this relationship remains understudied globally.18Trends in Ecology & Evolution. Forests and Their Canopies: Achievements and Horizons in Ecology and Conservation
Even the atmosphere itself is an underexplored biosphere. Microorganisms are carried aloft by wind and have been detected in the troposphere and lower stratosphere, but quantitative observations at altitudes more than a few kilometers above the surface are extremely rare.19PubMed Central. Abundance and survival of microbial aerosols in the troposphere and stratosphere Part of the problem is that reliable sample collection systems for high-altitude bioaerosols have been lacking.20PubMed Central. Airborne Bacteria in Earth’s Lower Stratosphere Resemble Taxa Detected in the Troposphere Airborne microbes are not just passive passengers. They may influence cloud formation, precipitation, and the long-distance spread of plant and human pathogens. Until the atmospheric biosphere is better sampled, its ecological role will remain largely guesswork.
Why the Percentage Keeps Shifting
One reason it is so difficult to pin down a single number for “what percent of Earth has been explored” is that the meaning of exploration keeps changing. A century ago, filling in the blank spaces on a map counted. Today, exploration means high-resolution bathymetric surveys of the seafloor, genetic barcoding of microbial communities, ice-penetrating radar transects across glaciers, and autonomous instruments profiling the twilight zone. Each new tool reveals that what we thought was mapped or cataloged was only a rough sketch.
Consider how the subglacial lake count has grown. Early inventories relied on airborne radar and identified a few dozen lakes. Improved geostatistical modeling and satellite observations have pushed the count to 773 and still climbing.6Nature Reviews Earth & Environment. Subglacial lakes and their changing role in a warming climate The same pattern plays out across domains. The number of known hydrothermal vent sites increases with each research cruise. The species count grows every time molecular tools are applied to a new sample. Even the percentage of mapped seafloor creeps upward as new sonar data stream in. Exploration is not a task that finishes; it is a process whose resolution keeps improving, and at every new level of detail, the unmapped fraction reasserts itself.
If you had to summarize the state of play in rough terms: we have imaged all of Earth’s surface from space, meaningfully mapped perhaps a quarter of the ocean floor, directly surveyed a small fraction of the deep sea, sampled only a sliver of the subsurface biosphere, and formally described well under a fifth of the species that exist. The planet we live on is, in many practical ways, still a stranger to us.