Spinosaurus was not built for speed in either environment. The best biomechanical estimates put its maximum swimming velocity at roughly 1 to 1.4 meters per second, or about 3 to 5 kilometers per hour, depending on whether it was at the surface or fully submerged. On land, its unusual body proportions, including shortened hindlimbs and a forward-shifted center of mass, almost certainly made it slower than other large theropods. The full picture involves a decades-long scientific debate about just how aquatic this animal really was, and the answer shapes every speed estimate researchers can produce.
How Fast Could Spinosaurus Swim?
The most detailed attempt to calculate Spinosaurus swimming speed came from a 2022 study that modeled the animal’s drag, tail thrust, and overall hydrodynamics. Under turbulent water conditions, the researchers estimated that Spinosaurus could manage about 0.8 meters per second while swimming at the surface and roughly 1.4 meters per second when fully submerged. The maximum velocity from tail thrust alone came out to around 1.2 meters per second. To put that in perspective, that is slower than a brisk human walk and roughly an order of magnitude slower than large modern aquatic predators that chase down prey.
1eLife. Spinosaurus is not an aquatic dinosaurThose numbers come with important caveats. The model had to account for several sources of drag that are unique to Spinosaurus: the undulation of its tail, wave formation near the surface, and the enormous sail on its back. The sail alone increased body drag by about a third. Maximum tail thrust was estimated at around 820 watts, which sounds like a lot until you consider that only a small fraction of the caudal muscle power actually translates into forward thrust during undulation. The mismatch between muscle mass and propulsive output is a major reason the speed estimates come out so low.
1eLife. Spinosaurus is not an aquatic dinosaurThese figures have been contested. A separate group of researchers argued that Spinosaurus likely had two distinct swimming modes. At low speeds, it could propel itself using just the flattened, fin-like portion of its distal tail, with its feet possibly helping to steer. During bursts of acceleration, deformation of the entire body would have allowed the dorsal sail itself to contribute thrust rather than purely acting as drag. Under that framework, the sail shifts from a handicap to a propulsive structure during fast swimming, though no one has produced a competing speed estimate that would place it dramatically higher.
2PubMed Central. Contributions to a Discussion of Spinosaurus aegyptiacus as a Capable Swimmer and Deep-Water PredatorThe Tail That Changed Everything
Before 2020, the idea of Spinosaurus as an active swimmer was mostly speculative. That changed when researchers described a remarkably preserved tail with extremely tall neural spines and elongated chevrons, forming a large, flexible, fin-like structure capable of sweeping side to side through water. Using a robotic apparatus that flapped physical models of different dinosaur tail shapes in a water tank, they demonstrated that the Spinosaurus tail generated significantly more thrust and operated more efficiently than tails of terrestrial theropods. Its performance was closer to that of living aquatic animals that use vertically expanded tails for propulsion.
3PubMed. Tail-propelled aquatic locomotion in a theropod dinosaurThat finding was a milestone, but “more efficient than a land dinosaur’s tail” is a low bar. Crocodile tails also generate thrust through lateral undulation, and crocodiles are themselves relatively slow swimmers compared to fully aquatic animals. The analogy matters because it sets a ceiling on what tail-driven propulsion can achieve in a body that is not hydrodynamically streamlined. A study of saltwater crocodile swimming kinematics found that while their kinematic efficiency in water is comparable to fully aquatic mammals, crocodiles are still slow in absolute terms, constrained by muscle performance and the forces that oppose forward movement during each tail beat.
4Australian Journal of Zoology. Ontogenetic changes of swimming kinematics in a semi-aquatic reptile (Crocodylus porosus)Spinosaurus faced those same physical constraints but with additional complications. Its body was much larger and less streamlined, the sail added considerable frontal area, and the limbs were not flipper-like. It had a paddle tail bolted onto a body that was not shaped for cutting through water. That combination is why most biomechanical analyses converge on relatively modest swimming speeds, even if researchers disagree about the exact number.
How Fast on Land?
Estimating land speed for Spinosaurus is harder than estimating swimming speed, partly because researchers cannot agree on how it walked. When a major subadult skeleton was described in 2014, the proportions were striking: the hindlimbs were noticeably short relative to the body, and the center of mass sat much further forward than in other large theropods. That forward shift meant Spinosaurus would have tipped forward while walking on two legs, leading the describing team to suggest it may have adopted a quadrupedal stance on land, at least some of the time.
If Spinosaurus walked on all fours, even occasionally, its top land speed would have been severely limited. Quadrupedal dinosaurs were generally slower than bipeds of comparable size, and Spinosaurus was not built like a typical quadruped either. Its forelimbs were strong but not designed for sustained locomotion the way a ceratopsian’s were. Conversely, if it did walk bipedally, those short hindlimbs still put it at a disadvantage relative to longer-legged theropods. No published study has modeled Spinosaurus land speed with the same rigor that exists for other theropods, but the anatomy strongly suggests it was among the slowest large predatory dinosaurs on land.
For comparison, musculoskeletal modeling of bipedal dinosaur running speeds has produced estimates for several theropods that agree reasonably well with what we’d expect from their anatomy.
5PubMed Central. Estimating dinosaur maximum running speeds using evolutionary roboticsSpinosaurus has not been put through that same modeling pipeline, in part because its locomotor mode is still debated. Whether bipedal or quadrupedal, though, the consensus among researchers is clear: this was not an animal that chased prey across open ground.
Why Being Slow in Both Environments Is Actually Expected
Animals that split their time between water and land tend to be mediocre at moving through both, because the physical requirements of aquatic and terrestrial locomotion pull body design in opposite directions. Limbs that are good for walking are poor paddles. Streamlined bodies that cut through water are less stable on uneven ground. A rigid torso helps with running but limits the lateral undulation needed for swimming.
This trade-off has been documented experimentally in living animals. A study of semi-aquatic salamanders found a strong negative correlation between an individual’s sprint speed on land and its burst velocity in water: the faster a salamander was on the ground, the slower it was in the water, and vice versa. The relationship held across genetically distinct populations, suggesting it reflects a fundamental biomechanical constraint rather than a quirk of any one species.
6PubMed Central. Trade-offs between aquatic and terrestrial locomotion and functional parallelism in Desmognathus salamandersSpinosaurus appears to sit squarely in this trade-off zone. Its anatomy reflects a partial commitment to water: dense, ballast-like bones; a paddle tail; retracted nostrils; and conical teeth suited to catching fish. But it retained enough terrestrial capability to walk on land, nest, and possibly forage along shorelines. That dual functionality came at a cost to performance in both environments. Expecting Spinosaurus to be fast in water or on land misunderstands how semi-aquatic body plans actually work.
Dense Bones as Built-In Ballast
One of the clearest indicators that Spinosaurus spent serious time in water comes from its bones. The tibial bones show an osteosclerotic condition, meaning they are unusually dense and compact rather than hollow like most theropod bones. This kind of bone density is strongly associated with semi-aquatic lifestyles across living animals: it acts as ballast, helping the animal sink below the surface instead of bobbing like a cork.
7Cretaceous Research. Semi-aquatic adaptations in a spinosaur from the Lower Cretaceous of BrazilA partial tibia from a related Brazilian spinosaur, dating to millions of years before the Moroccan Spinosaurus specimens, already shows this same high bone compactness. That suggests the adaptation evolved early in the spinosaurid lineage and was not unique to one species or one point in time. The practical effect is significant for swimming speed: denser bones make the animal less buoyant, which means less energy is wasted fighting to stay submerged, but the extra weight also makes every stroke require more power. It is another piece of the puzzle that points toward slow, energy-efficient swimming rather than bursts of speed.
7Cretaceous Research. Semi-aquatic adaptations in a spinosaur from the Lower Cretaceous of BrazilThat said, researchers have cautioned against reading bone compactness as a simple lifestyle indicator. A 2024 study pointed out that disparate lifestyles have been proposed for Spinosaurus based on similar skeletal evidence, ranging from a shoreline wader to a surface swimmer to a fully aquatic underwater pursuit predator. Bone density alone cannot distinguish between these scenarios, which is part of why the speed debate remains so active.
8PLoS One. Diving dinosaurs? Caveats on the use of bone compactness and pFDA for inferring lifestyleCould It Even Float Comfortably?
Before worrying about how fast Spinosaurus could swim, some researchers asked a more basic question: could it stay upright in the water at all? A buoyancy analysis modeled what would happen if you dropped a Spinosaurus into water and let it reach equilibrium. The results were reassuring for the semi-aquatic hypothesis. With fully inflated lungs, the models floated with their heads and nostrils above the surface, their center of mass and center of buoyancy roughly aligned. The sail stayed visible above the waterline.
9PubMed Central. A buoyancy, balance and stability challenge to the hypothesis of a semi-aquatic Spinosaurus Stromer, 1915 (Dinosauria: Theropoda)That stability matters for speed estimates because an animal that is fighting to keep its head above water cannot devote all its energy to forward propulsion. Spinosaurus appears to have been stable enough at the surface that breathing and seeing were not a problem, freeing its tail to focus on generating thrust. Floating came naturally to it, which fits the profile of an animal that spent extended time in water without needing to be especially fast. Compare that to a terrestrial animal that falls into deep water: much of its energy goes to panicked paddling just to keep its nose above the surface, leaving nothing for efficient swimming.
If It Was So Slow, How Did It Eat?
A top swimming speed of roughly 1 to 1.4 meters per second sounds dismal for a predator, but the assumption that Spinosaurus needed to be fast to catch food may be wrong. Its skull and teeth are specialized for fish: the jaws are long and narrow, the teeth are conical and interlocking, and pressure-sensing structures in the snout may have helped detect prey in murky water. None of that points to high-speed pursuit. It points to ambush.
Modern crocodilians are an instructive parallel. They are not fast swimmers in sustained terms, yet they are spectacularly effective predators. They use stealth, patience, and explosive short-range strikes rather than chasing prey over distance. Spinosaurus may have operated similarly: wading or floating quietly, then lunging at fish that came within range. The two-mode swimming hypothesis supports this interpretation, with slow cruising for positioning and brief whole-body acceleration for the strike.
2PubMed Central. Contributions to a Discussion of Spinosaurus aegyptiacus as a Capable Swimmer and Deep-Water PredatorThe ecosystem also mattered. Spinosaurus lived alongside other enormous theropods, and researchers believe niche partitioning kept them from competing directly for the same food. Spinosaurus occupied the aquatic niche, targeting fish and other water-dwelling prey, while other large predators hunted terrestrial animals.
10Geology Today. The lifestyle of SpinosaurusIn that context, raw speed was less important than access. Being the only 15-meter predator willing to wade into a river gave Spinosaurus a food source that no other large theropod could exploit, regardless of how slowly it moved through the water.
What the Sail Was Actually For
The giant dorsal sail is probably the most visually striking feature of Spinosaurus and one of the most debated. Whatever its primary function, it clearly had consequences for locomotion. In water, it increased drag by about a third based on the biomechanical modeling discussed earlier. On land, a tall sail supported by neural spines up to 1.5 meters long would have raised the center of gravity, potentially making the animal less agile during turns or rapid movements.
Proposed functions for the sail include thermoregulation, sexual display, species recognition, fat storage, and even hydrodynamic stabilization while swimming. One analysis described Spinosaurus as a semi-aquatic hunter and explored the sail’s possible role in that context.
11Geological Magazine. The riddle of Spinosaurus aegyptiacus’ dorsal sailResearchers who favor the aquatic interpretation have suggested that the sail provided roll stability during surface swimming, acting like a keel in reverse. Others counter that a structure adding 33 percent more drag is a steep price for stability and is better explained by a function that had nothing to do with swimming at all, like display to potential mates.
From a speed standpoint, the sail is almost certainly a net negative in both environments. It adds drag in water and raises the center of gravity on land. Whichever explanation for its existence turns out to be correct, the sail is one more reason Spinosaurus was not an animal optimized for velocity.
How Crocodile Research Informs the Debate
Crocodilians are the closest living analogue to semi-aquatic archosaurs like Spinosaurus, and their locomotion has been studied in considerable detail. In saltwater crocodiles, aquatic propulsion comes primarily from tail undulations, with limbs contributing mainly at low speeds or in very small individuals. Tail-beat frequency increases with swimming speed, but amplitude stays roughly constant for a given body size, and overall kinematic efficiency actually drops as the animal swims faster.
4Australian Journal of Zoology. Ontogenetic changes of swimming kinematics in a semi-aquatic reptile (Crocodylus porosus)That declining efficiency at higher speeds is a key insight. It means that for a tail-driven swimmer, pushing beyond a moderate cruising speed costs disproportionately more energy per meter traveled. Spinosaurus, with its even less streamlined body and massive sail, would have hit that efficiency wall at an even lower speed than a crocodile. Researchers who modeled Spinosaurus at crocodile-level Reynolds numbers used a hypothetical speed of about 15 kilometers per hour for comparison purposes, but that figure reflects what a crocodile can do, not what Spinosaurus could sustain.
12PubMed Central. Contributions to a Discussion of Spinosaurus aegyptiacus as a Capable Swimmer and Deep-Water Predator – Section: Hydrodynamic AspectsDespite those speed limitations, crocodile swimming efficiency is comparable to fully aquatic mammals on a per-stroke basis. The bottleneck is not wasted motion but raw power and the physics of pushing a non-streamlined body through water. Spinosaurus faced the same bottleneck amplified by its size. A 10-meter-long animal operating at the Reynolds numbers of a crocodile but shaped less favorably would need enormous power output to reach even moderate speeds. The biomechanical models bear this out: the roughly 820 watts of maximum tail thrust Spinosaurus could produce simply was not enough to overcome the drag its body generated at higher velocities.