Placing the Near-Earth Object Impact Probability in Context

Draft version August 5, 2025
Typeset using L
A
TEX default style in AASTeX631
Placing the Near-Earth Object Impact Probability in Context
C. R. Nugent,1
K. P. Andersen,2
James M. Bauer,3
C. T. Jensen,2
L. K. Kristiansen,2
C. P. Hansen,2
M. M. Nielsen,2
and C. F. Vestergård2
1Olin College of Engineering
1000 Olin Way
Needham, MA 02492, USA
2 Department of Materials and Production
Aalborg University Fibigerstræde 16
9220 Aalborg, Denmark
3Department of Astronomy
University of Maryland
College Park, MD 20742, USA
(Received February 12, 2025; Accepted July 11, 2025)
Submitted to The Planetary Science Journal
ABSTRACT
Near-Earth objects (NEOs) have the potential to cause extensive damage and loss of life on Earth.
Advancements in NEO discovery, trajectory prediction, and deflection technology indicate that an
impact could be prevented, with sufficient warning time. We derive an impact frequency of NEOs
140m and larger, using the NEOMOD2 NEO population model and JPL Horizons. We then place
that frequency in context with other preventable causes of death; allowing for comparison between a
planet-wide event and individual events that cause fatalities such as car crashes and carbon monoxide
poisoning. We find that the chance of a > 140 m asteroid hitting the Earth is more likely than the
chance of an individual being struck by lightning.
Keywords: Near-Earth objects (1092) — Asteroids (72) — Asteroid dynamics (2210)
1. INTRODUCTION
An asteroid impact is unique among natural disasters; it is the only one that is, theoretically, technologically
preventable. The Double Asteroid Redirection Test (DART) Mission demonstrated that a spacecraft could change the
along-track velocity of a ∼ 150 m diameter asteroid moon by 2.6 mm s−1
(Naidu et al. 2024), corresponding with an
orbital period change of 33 minutes (Thomas et al. 2023; Daly et al. 2023). With sufficient advance notice of years to
decades, it is plausible that a similar space mission could change the orbit of a potential impactor of roughly the same
size, precluding an impact (Council 2010). To provide advance notice of an impact, asteroid surveys (e.g., Jedicke
et al. 2015) discover new near-Earth objects (NEOs) nightly. Over the last several decades, the number of discovered
NEOs has grown.1
These discoveries have been enabled by legislation such as the American law named “The George
E. Brown, Jr. Near-Earth Object Survey Act” which mandated NASA discover 90% of all NEOs 140 meters or larger
by 2020. NASA has developed a comprehensive planetary defense strategy (Aeronautics & Administration 2023).
There is strong evidence that comet and asteroid impacts can be widely destructive. A ∼ 10 km minor planet
impact remains the most likely cause of the extinction of the dinosaurs 65 million years ago (Alvarez et al. 1980).
Corresponding author: C. R. Nugent
cnugent@olin.edu
1 Regularly updated data is available at https://cneos.jpl.nasa.gov/stats/totals.html.
arXiv:2508.02418v1
[astro-ph.EP]
4
Aug
2025

2
This phenomenon is not one of the distant past, as the 1908 Tunguska (Chyba et al. 1993), 1992 Shoemaker–Levy 9
(Hammel et al. 1995) and 2013 Chelyabinsk (Popova et al. 2013) impacts demonstrate.
When discussing NEO impacts and the destruction they can cause, researchers generally divide the problem into
separate domains. Some researchers have calculated the frequency of NEO impacts for various size ranges of NEOs.
Others have investigated the damaging consequences of impacts. Often, the results of these studies are intended
for specialists and are presented without broader context. Chapman & Morrison (1994, Figure 3) made a notable
contribution by placing the probability of being killed due to an NEO impact in relation to other causes of death, such
as a tornado or fireworks accident.
1.1. Impact Frequencies
NEO impact probabilities have been estimated with some regularity over the last eighty years. Chapman & Morrison
(1994) reference an estimate by Watson (1941), who stated “the Earth probably goes at least a hundred thousand years
between collisions with [minor planets]” (as quoted by Baldwin 1949). Calculation methodology used in the literature
has evolved over time as computational capabilities have increased. For example, Morrison et al. (2002), following
methodology from Harris (1998), examined all objects discovered as of 3 July 2001 with H < 18.0 and perihelion < 1.0
AU (sample size of 244) and calculated close approaches to determine an impact probability. Those results agreed
with those made by Shoemaker (1983) 20 years previously. Recent studies (e.g., Harris & Chodas 2021; Nesvorný
et al. 2024a) use more sophisticated numerical integration techniques with a sample of simulated NEOs derived from
population models.
NEO population models, which de-bias survey observations to infer a full picture of the NEO population, can help
researchers gain an understanding of NEO origins and evolution over time. They also are used to gauge progress
towards discovery mandates. Multiple population models have been created which vary in the methodology specifics
and the reference data used; some focus on particular NEO sub-populations of interest (Rabinowitz 1993; Bottke et al.
2000; Rabinowitz et al. 2000; D’Abramo et al. 2001; Stuart 2001; Bottke et al. 2002; Mainzer et al. 2011; Greenstreet
et al. 2012; Harris & D’Abramo 2015; Granvik et al. 2016; Tricarico 2017; Granvik et al. 2018; Harris & Chodas 2021;
Heinze et al. 2021; Grav et al. 2023). This work employs Nesvorný et al. (2024a), which is a update to Nesvorný
et al. (2023), and is based on Catalina Sky Survey observations. It considers asteroids originating from eleven main-
belt sources, as well as comets originating from an additional source. NEO population modeling is an active area of
investigation. New models were published after the methods described in this paper were completed (Nesvorný et al.
2024b; Deienno et al. 2025), we would not expect our results to be substantially different if these newer models were
used.
1.2. Consequences of an NEO impact
Quantifying the extent of damage or loss of life from an NEO impact is difficult, due to the wide range of variables
at play. This includes (but is not limited to) impactor speed, impactor size, impactor composition (loose regolith, solid
rock, metallic), impactor angle, and where the impact occurs (over land or sea, near or far from a populated center).
One can divide NEO impacts into two categories; ones that produce only local effects, and ones that produce local
and global effects.
Local effects can include damage from the blast overpressure as the NEO explodes, thermal radiation, and possible
tsunamis. To capture the complexity of the possible outcomes, Mathias et al. (2017) used a probabilistic model
which sampled from a distribution of asteroid properties to produce millions of simulated impacts across Earth. They
considered NEOs < 300m; these were considered small enough to not produce global effects. They found that in some
cases, for impactors 200 m and smaller, an impact in a remote location could occur without affecting a population
(Figure 10, Mathias et al. 2017).Rumpf et al. (2016) also considered local effects from NEOs and applied their model
over the Earth to determine country-level risk.
Global effects are possible in the case of a high-energy NEO impact. As described in papers by Chapman & Morrison
(1994) and Toon et al. (1997) these could result in global cooling due to dust, fires, nitrogen oxide generation, the
release of sulfur dioxide, multi-continental fires triggered by raining ejecta from the crater site, acid rain, and poisoning
from heavy metals from the impacting body. The dust lofting alone has the potential, in some cases, to obscure the
sun to the point of stopping photosynthesis, which would then cause a mass extinction. These are in addition to the
local effects.
As there is little data on global effects, it is difficult to validate global effect models, and all studies on high energy
impacts stress the uncertainties inherent in the results. However, this is the rare case where researchers do not wish

3
for additional data. Toon writes, “it is to be hoped that no large-scale terrestrial experiments occur to shed light on
our theoretical oversights” (Toon et al. 1997). Researchers have made do with the best available proxies, including
evidence from historical impacts, volcanic eruptions, and nuclear weapons tests. Reinhardt et al. (2016) created a
risk assessment that included local and global effects, though did not include tsunamis. Building on the PAIR model
used by Mathias et al. (2017), Wheeler et al. (2024) used global probabilistic models to evaluate risk. This model is
enhanced by the use of an asteroid property inference model (Dotson et al. 2024), and it can be used to assess the risk
from a specific NEO (such as the hypothetical NEOs in the Planetary Defense Conference tabletop exercises) as well
as global risk.
An enduring challenge is clearly and accurately conveying the risk from NEOs that results from this modeling.
There are many more NEOs of the size that solely cause local effects than NEOs capable of causing global effects upon
impact. Taking the average of the risk does not accurately capture this imbalance. The global effects are so severe that
the possibility of their occurrence, although less probable than an impact that produces solely local effects, should not
be ignored or obscured by averaging. Adding further complexity, as Morrison et al. (2002) explain, estimates often do
not take into account social reactions to an impact that could change fatality rates. An effective large-scale evacuation
of an impact zone, for example, could save lives.
We describe a new calculation of impact probability that employs NEOMOD2 and JPL Horizons. The results are
compared to previous calculations. We then compare this probability to the probabilities of other preventable causes
of death. Although impact probability calculations have been regularly updated as NEO discoveries have increased,
we have not been able to locate a work that contextualizes this frequency in the literature since Chapman & Morrison
(1994).
2. METHODS
The impact frequency calculation was undertaken by a group of six undergraduate students as part of a semester-long
project under the guidance of two co-advisors. A simulated set of semi-major axis, eccentricity, and inclination orbital
elements as well as absolute magnitude H of 5 × 106
NEOs was generated using the NEOMOD2 FORTRAN code (Nesvorný
et al. 2024a). NEOMOD2 was chosen because it was a recently published work, and it provided FORTRAN code to generate
the orbital elements. The simulated set of NEOs was restricted such that 10 < H < 22. This limit is the conventional
cutoff that has been used in the literature, as it corresponds to objects > 140 m assuming an albedo of 0.14 or 0.15.
H of 22 is a threshold set by the George E. Brown, Jr. Near-Earth Object Survey Act. NEOs > 140 m cause regional
or global devastation upon impact.
Using H as a proxy for size, although common in the literature, has drawbacks. The H < 22 limit excludes dark (0.05
albedo) NEOs > 140 m. It could also include a subset of objects that are smaller than 140 m but have albedos higher
than 0.14. However, there is not evidence that this will bias the results of this work. The NEOMOD2 population,
which had associated H values but not sizes, was compared to the NEOMOD3 population, which had H values and
diameters (Nesvorný et al. 2024b, Figure 15). For an albedo of 0.14 and sizes greater than 0.1 km, there is close
agreement in Earth impact flux between the two population models, ruling out a significant excess of large, dark NEOs
and small, bright NEOs biasing the results.
The elements were plotted and compared to the figures in Nesvorný et al. (2024a) to ensure accuracy. To complete
the set of orbital elements, for each simulated asteroid mean anomaly, longitude of ascending node, and argument of
perihelion values were randomly chosen from the valid range of values for these elements.
The trajectories of these simulated asteroids were then calculated, to determine if any would impact Earth over a 150
year interval. This was accomplished by querying the JPL Horizons ephemeris calculation service (Giorgini 2015)2
for a close approach table via API, providing it with the orbital elements. JPL Horizons automatically determines
the appropriate integration step size and precisely computes trajectories, including close approaches. Its accuracy has
been extensively verified by astronomers and spacecraft mission planners. Submission of API requests and ingesting of
responses was handled by code written in Python. A print interval value STEP SIZE of 150 years was specified in the API
request. To accommodate the large number of requests, the requests were parallelized using the concurrent.futures
Python module, and the queries were divided over the six student computers. Total time for the queries was roughly
780 hours. The results of the API queries were collected in a MySQL database to simplify analysis and enable data
verification. An impact was identified when the close approach distance to Earth was less than the radius of the Earth.
2 https://ssd.jpl.nasa.gov/horizons/

4
To contextualize the NEO impact frequency, literature searches were conducted to locate studies on other causes of
death that could be considered preventable. We considered events that affect individuals for this contextualization,
instead of other planetary-scale events such as volcanic super-eruptions or large-magnitude earthquakes. Although a
comparison to planetary-scale events may be interesting to the small field of experts who are familiar with the Earth’s
geologic history, it would not be broadly comprehensible to non-specialists. By comparing the NEO impact frequency
with familiar events such as influenza illness, we aimed to increase overall understanding of the likelihood of an impact.
Chapman & Morrison (1994) previously placed an asteroid impact in context with other causes of death such as
murder, fireworks accidents, and botulism. In that work, they considered the chance of death due to an impact
alongside the chance of death due to other factors. This work addresses a slightly different question; we place the
chance of an impact occurring anywhere on Earth relative to the chance of other events of concern happening to
an individual. This work is therefore intended to provide context to those who wish to know the probability that
a > 140 m impact will occur, anywhere on Earth, in their lifetime. A medium to large Earth-NEO impact would
be a remarkable, historic event. It would likely attract media attention, and footage of the impact would likely be
recorded and shared worldwide. It would be witnessed by, and would likely emotionally affect, a significant fraction
of the human population, even if only a very small fraction of that population was directly affected via loss of life or
property. Because of the enormity of an impact event it is worth contextualizing the likelihood that it would happen
anywhere on the planet within an average human lifetime.
In the literature search, recent data (2020-2024) was prioritized for higher-frequency events. Studies of lower fre-
quency events often needed a several-decade baseline to accumulate sufficient data to report. Data from peer-reviewed
journals and from governmental and intergovernmental organizations was considered. Many studies focused on spe-
cific countries. This data was normalized by year and by population; country-wide studies were normalized by the
population of the country at the midpoint of the study.
3. RESULTS
3.1. Impact Frequency Calculation
In the sample of 5×106
simulated NEOs, three objects were determined to impact the Earth, producing a per-object
impact probability of 4 × 10−9
yr−1
. Despite the differences in method detail and sample size, this is the same order
of magnitude as Morrison et al. (2002)’s per-object probability of 1.68 × 10−9
yr−1
.
The current number of discovered asteroids thought to be 140 m in diameter or larger (mostly based on absolute
magnitude values) is 11,1863
. However, to compare our results with Nesvorný et al. (2024a), we use the number of
discoveries as of mid-2023, which was 10,502. Nesvorný et al. (2024a) places the survey completeness for NEOs brighter
than H = 22 to be roughly 47% (Nesvorný et al. 2024a, Figure 16). Harris & Chodas (2021) places it a bit higher than
55% (Nesvorný et al. 2024a, also Figure 16), whereas Grav et al. (2023) find ∼ 38%4
. The average of these findings is
46%; which we use to infer a total existing population of 22,800 NEOs in the size range of interest.
This allows us to estimate an impact frequency of 9.1 × 10−5
from our results. Our method is limited by small
number statistics. This constraint was imposed by the queries being part of a semester-long project. We consider the
case of our results being changed by ±1 impactor to gain a rough sense of uncertainties. Two impactors would have
produced an impact frequency of 6.0×10−5
, four an impact frequency of 1.2×10−4
. This range is in line with previous
studies. We summarize a subset of results in Table 1.
One could argue that some fraction of the impact probability is retired, as no currently discovered NEO has a
meaningful chance of impact.5
Using the average of the reported discovery fractions, this would reduce impact
probabilities by 38% to 55% (Nesvorný et al. 2024a; Grav et al. 2023).
Table 1 shows that, broadly, recent impact frequency estimates are similar to early estimates such as Opik (1958),
despite the increase of NEO discoveries over time6
. This trend has been noted before (e.g., Chapman & Morrison
1994; Morrison et al. 2002). Table 1 provides further confirmation with the addition of studies from the last 25 years.
3 Data from https://cneos.jpl.nasa.gov/stats/totals.html; accessed Jan 18 2025.
4 This work notes that their results are consistent with Harris & Chodas (2021), citing personal communications and “recent recalculations
in the absolute magnitudes of NEOs.” This highlights the complexity inherent in these estimates. We include the original Harris & Chodas
(2021) as it is a current peer-reviewed published value.
5 Data from the Center for NEO Studies (CNEOS). https://cneos.jpl.nasa.gov/stats/totals.html; accessed Jan 8 2025.
6 The estimates of impact frequencies of > 1km NEOs has shown more historical variation, this may be partly due to the smaller total
numbers of NEOs of this size range. The inferred population of these large objects are <∼ 5% of the population > 140m.

5
Table 1. Comparison of > 140m NEO Impact Frequency Calculations over Time
Study Magnitude Range Size Range Impact Frequency, per year
Watson (1941), quoted in Baldwin (1949) − - 10−5
Opik (1958) − > 130 m 4.6 × 10−5
Rabinowitz (1993) H < 22 > 140 m ∼ 10−4
Toon et al. (1997) − > 140 m 5 × 10−4
Granvik et al. (2018) H < 22 > 140 m ∼ 10−5
Harris & Chodas (2021) H < 22 > 140 m 4 × 10−5
Nesvorný et al. (2024a) H < 22 > 140 m 4 × 10−5
This work H < 22 > 140 m 9 × 10−5
Note—Size ranges are generally inferred, assuming 0.14 or 0.15 albedo. Albedo assumptions by Opik (1958)
are not known.
This implies that observational sampling of the Earth-approaching NEO population has, in a broad sense, accurately
reflected the total population. In other words, unless there is a survey discovery bias that has totally escaped detection
and persisted for the history of asteroid discovery, undiscovered NEOs are likely to be “known unknowns” which have
orbital elements in line with known populations, not “unknown unknowns,” or a population of NEOs on orbits that
are substantially different than discovered NEOs.
A caveat to this conclusion is the population of interstellar objects, or asteroids or comets that originate in an
extrasolar system and arrive in or pass through our solar system. Only two are known. The first, ‘Oumuamua, was
discovered in 2017 and the second, Borisov, in 2019. There have been no discoveries since. Although it is likely that
these objects are quite rare, we cannot entirely rule out the possibility that they represent a “unknown unknown”, or
a class of object that is systemically missed by surveys.
3.2. Impact Probability in Context
We compare the NEO impact probability to other technologically-preventable causes of death in Figure 1. This
places an event that would affect the planet (NEO impact) in context with potentially deadly events that can befall
individuals. This serves the purpose of allowing experts and non-experts to place the probability of an NEO impact
within a mental framework of events they may have some familiarity with, such as car crashes and animal attacks.
Each individual has a unique risk profile, which can vary based on factors such as age, local region, and hobbies.
Figure 1 is intended to provide some familiar deadly events for most audiences; we do not expect all events to be
relevant to all readers due to the geographically varied nature of the comparison event studies. An individual could
gain context about the NEO impact probability by selectively choosing a subset of events most relevant to them; a
different individual, living in a different country, may select a different subset. Individuals might also adjust their own
interpretations of Figure 1 based on their behavior. For example, a building inspector who regularly tests and maintains
their home carbon monoxide detector may conclude that they have a lower chance than average of experiencing carbon
monoxide poisoning. Details on each comparison study, including caveats, are discussed below.
This figure presents results in terms of frequency over a human lifetime, following research into effective technical
communication (Peters 2017). The average human lifetime is taken to be 71 years, the current global average.7
The
data used to generate Figure 1 is shown in Table 2.
The wide range of outcomes from an NEO impact are represented by large error bars. While a 140 − 200m NEO
impacting over the ocean could produce no fatalities (Mathias et al. 2017; Wheeler et al. 2024), Mathias et al. (2017)
(Figure 9) shows that 180 − 200m NEOs have small chance of affecting 106
people if an impact was to occur in a
highly populated region. The largest NEO impacts have the capability to affect the entire world population (Toon
7 World Health Organization. accessed Jan 8 2025. https://www.who.int/data/gho/data/indicators/indicator-details/GHO/
life-expectancy-at-birth-(years)

6
Figure 1. The chance of a > 140m NEO impacting Earth placed in context with the chance that other preventable events may
happen to an individual. Individual event frequency and fatality chances are taken from peer reviewed studies on particular
regions over specific time intervals; these are broad averages, individual risks vary widely. We expect most readers to be able
to identify some events that are relevant to their lives, and some that are not. Readers interested in their own chances are
encouraged to consider the events most relevant to their demographics, geographic region, and hobbies. Section 3.2 has a further
discussion of details and caveats. This figure is intended to contextualize the > 140m NEO impact frequency with events people
may be familiar with, such as flu sickness and lightning strikes. The x-axis is the chance of the rare event happening to the
planet (NEO impact) or an individual (all other events) over an average human lifetime. The y-axis shows the chance of fatality
to an individual if the event occurs. The chance of fatality from a > 140m impact is dependent on a large range of variables,
such as impactor size (140 m to > 10 km) and impact location (populated city to middle of a large ocean). An impact could
result in no fatalities (less likely) some fatalities (more likely) or mass extinction (very unlikely, but possible; see Section 1.2),
this is represented by large error bars (blue). The placement of the impact symbol on the y-axis is in the visual center to avoid
the pitfalls of averaging discussed in Section 1.2
.

7
Table 2. Comparison of Selected Preventable Events
Event
Number Affected
Yearly
Number Killed
Yearly
Population
considered
References
Dry sand hole collapse (USA) 5.2 3.1 2.80 × 108
Maron et al. (2007)
Coyote attack (USA) 7.8 3.00 × 10−2
2.70 × 108
Baker & Timm (2017)
Elephant attack (Nepal) 2.70 × 101
1.80 × 101
2.70 × 107
Acharya et al. (2016)
Lightning strike (USA) 2.77 × 102
2.77 × 101
3.20 × 108
CDC
Skydiving accidents (the Netherlands) 1.08 × 102
1.0 1.60 × 107
Damhuis et al. (2024)
Carbon Monoxide poisoning (Denmark) 1.15 × 103
1.05 × 102
5.40 × 106
Simonsen et al. (2019)
Injury-causing car crash (Mass, USA) 3.06 × 104
3.66 × 102
7.10 × 106
Massachusetts DOT
Rabies (USA) 8.00 × 105
5.0 3.40 × 108
Ma et al. (2023), CDC
Influenza illness (World) 1.00 × 109
4.70 × 105
8.10 × 109
WHO
Note—Summary of data from papers on preventable fatal events. We calculate the number of people affected by the event
each year from the given reference. The number of people killed per year is the number of fatalities due to the event. None of
these events are 100% (or 1:1 chance) fatal. The population considered is the population of the region at the midpoint of time
considered by the study; for example, the study on carbon monoxide poisoning examined those events in Denmark between 1995
and 2015. Therefore the population considered is the population of Denmark in 2005. Country and world population estimates
are from World Population Prospects, United Nations Population Divisiona
. Massachusetts population value is from the United
States Census.b
a https://population.un.org/wpp/
b https://www.census.gov/quickfacts/fact/table/MA/PST045223
et al. 1997; Wheeler et al. 2024). To avoid the pitfalls of averaging (see Section 1.2) we place the NEO icon in the
visual center of the y-axis.
This comparison contextualizes the NEO impact frequency, providing reference points for the scientific community
and enhancing communication with the non-expert public. However, just as NEO impacts are complex and dependent
on several variables, so do each of these rare events. The frequency of each of these events is normalized by the relevant
country or worldwide population, as appropriate, but this generalization averages over sub-populations that are more
vulnerable, such as children. Each of these events, as well as the rationale for why they are considered preventable,
are discussed below.
Dry sand hole collapse refers to when a hole is dug, generally at a beach, “for recreational purposes” but collapses,
trapping someone inside. Maron et al. (2007) writes of their American study, “The risk of this event is enormously
deceptive because of its association with relaxed recreational settings not generally regarded as hazardous.” It is
preventable by not digging large holes in dry sand. The mean age of victims was 12. Older people tend not to dig
large holes in the sand.
Deaths from animal attacks and related infections (elephant, coyote, rabies) can be prevented, broadly, by avoid-
ing wild animals, making adjustments to the human-created environment, and medical treatment (such as the rabies
vaccine). The study on coyote attacks referenced in this work (Baker & Timm 2017) required a 39 year baseline to
establish statistics on this vanishingly rare occurrence in the USA and Canada8
. They recommend nine programs
to prevent coyote-human habituation, including public education. They note that children are much more likely to
be the targets of a coyote attack. The study of elephant attacks in Nepal (Acharya et al. 2016) has four “Man-
agement recommendations” to reduce human-elephant encounters, such as “Restore corridors in critical areas along
elephant migratory routes” so that the elephants can complete their migrations without traveling through populated
regions. And although 800,000 Americans seek treatment for rabies following an animal bite yearly, in 2021 only five
8 The USA-only results were used in this work.

8
died; four who did not seek rabies post-exposure prophylaxis treatment and one who did receive treatment but was
immunocompromised (Ma et al. 2023)
The data on lightning strikes was obtained from the American Center for Disease Control (CDC).9
Lightning
strikes can be prevented by being inside during electrical storms, and avoiding close proximity to plumbing, electrical
equipment, windows, and concrete walls and floors reinforced with rebar.10
Sub-populations are more vulnerable, for
example, lightning is more likely to strike people whose jobs require significant outdoors work. Skydiving data was
taken from a study on the sport in the Netherlands (Damhuis et al. 2024); it is an optional recreational activity.
Carbon monoxide poisoning statistics are from Denmark (Simonsen et al. 2019), those fatalities can be prevented via
the use of working carbon monoxide detectors.
To the right of the x-axis of the graph are events that are fairly likely to happen to an individual over a lifetime;
namely car accidents11
and influenza infections12
. There exist technologies that can reduce the likelihood of death
from these events or prevent it,, such as personal protective equipment and advanced safety features. Individuals and
societies make trade-offs between the monetary and other costs of this safety equipment, and generally decide to adopt
some measures while also assuming some level of risk.
It is impossible to monetarily quantify the full moral and social benefits of saving a single human life, much less
quantifying the benefits of preventing the regional to global destruction of an NEO impact. For each of the rare events
presented here (with the exception of dry sand hole collapse, which is extremely uncommon), there is often community
investment in precautions to reduce risk. For example, some regional laws require the purchasing of carbon monoxide
detectors for each floor of a residence. This results in a cost of roughly 20 to 40 USD over 5-10 years, the lifetime of
the detector. Health departments and hospitals stock rabies vaccines. Comparatively, the cost of the NASA-funded
DART mission was 324.5 million, or, on average, 1 USD per American. The National Academies report “Defending
Planet Earth” states, “The committee considers work on this problem as insurance, with the premiums devoted wholly
toward preventing the tragedy.” (Council 2010).
4. CONCLUSIONS
We present a calculation of the impact frequency from NEOs based on an updated model of the NEO population
and precision orbital integrations using JPL Horizons. We show that despite varying methodologies, NEO impact
frequency estimates have remained roughly constant over the last 80 years. We place the result in context with other
rare, preventable events.
The discovery of a significant fraction of the > 140 m NEO population has enabled detailed studies into the risk
from those objects. However, there remains significant work to be done. Asteroid surveys are continuously discovering
new NEOs, growing our understanding of the sub-140 m population. This work will eventually improve statistics on
the impact frequency of smaller asteroids.
However, there are sub-populations of Earth-approaching objects that are poorly understood. We have very limited
information on interstellar objects from the two known objects in this population. The Vera Rubin Observatory
Legacy Survey of Space and Time may shed light on this issue by discovering more interstellar objects (Cook et al.
2016; Schwamb et al. 2023), which may allow for the risk from these objects to be evaluated.
Despite advances in understanding (Oort 1950; Francis 2005; Bauer et al. 2017), the population of long-period comets
is less well known than the near-Earth asteroid population. Whereas the impact velocity for asteroids is, on average,
20 km/s, long period comets can have velocities three times as large (Chapman & Morrison 1994). As kinetic energy
E = 1
2 mv2
, where m is mass and v is velocity, comet impact energies can be nine times as large as asteroids of
equivalent mass. To enable a precise calculation of the risk from near-Earth comets, there is value in prioritizing the
study and discovery of long-period comets, towards creating a comprehensive cometary population model.
9 Accessed 27 Jan 2025, https://www.cdc.gov/lightning/data-research/index.html
10 US Weather Service Lightning Tips, https://www.weather.gov/safety/lightning-tips
11 Massachusetts Department of Transportation Crash Data Portal, accessed 21 Jan 2025, https://apps.impact.dot.state.ma.us/cdp/report
12 World Health Organization, accessed 21 Jan 2025, https://www.who.int/news-room/fact-sheets/detail/influenza-(seasonal)

9
5. ACKNOWLEDGMENTS
This work was supported by a Fulbright Denmark US Scholar Grant. We thank everyone at the Institut for Materialer
og Produktion at Aalborg University for their hospitality. We are grateful to Aalborg University Vice Dean Olav Geil
for writing the letter of support that enabled this collaboration and Professor Thomas Tauris for his support, advice,
and kindness. We thank T. Spahr for his advice and feedback on this draft. The Olin College library staff obtained
access to many of the studies referenced in this work, particular support was provided by M. Anderson. R. Stevenson
provided valuable feedback as well. This work was enabled by the JPL Horizons on-line system. We are grateful to
the anonymous referees whose thoughtful comments improved this manuscript.
Figure 1 uses graphical icons (CC BY 3.0) from The Noun Project, including dig by Adrien Coquet, Fox by pic-
tohaven, Lightning by Dong Gyu Yang asteroid by Icongeek26, rabies virus by Shocho, flu by Arjuna, Elephant by
Harianto, Skydiving by Adrien Coquet, car crash by Tritan Pitaloka, and depression by Luis Prado.
Software: JPL Horizons (Giorgini 2015), NEOMOD2 (Nesvorný et al. 2024a)

10
REFERENCES
Acharya, K., Paudel, P., Neupane, P., & Köhl, M. 2016,
Human-Wildlife Conflicts in Nepal: Patterns of Human
Fatalities and Injuries Caused by Large Mammals, PloS
one, 11, e0161717, doi: 10.1371/journal.pone.0161717
Aeronautics, N., & Administration, S. 2023, NASA
Planetary Defense Strategy and Action Plan: In Support
of the National Preparedness Strategy and Action Plan
for Near-Earth Object Hazards and Planetary Defense, in
Support of the National Preparedness Strategy and
Action Plan for Near-Earth Object Hazards and
Planetary Defense (National Aeronautics and Space
Administration). https://purl.fdlp.gov/GPO/gpo215632
Alvarez, L. W., Alvarez, W., Asaro, F., & Michel, H. V.
1980, Extraterrestrial Cause for the Cretaceous-Tertiary
Extinction, Science, 208, 1095,
doi: 10.1126/science.208.4448.1095
Baker, R. O., & Timm, R. M. 2017, Coyote attacks on
humans, 1970-2015: implications for reducing the risks,
Human-Wildlife Interactions, 11, 120
Baldwin, R. B. 1949, The face of the moon. (The
University of Chicago Press)
Bauer, J. M., Grav, T., Fernández, Y. R., et al. 2017,
Debiasing the NEOWISE Cryogenic Mission Comet
Populations, AJ, 154, 53, doi: 10.3847/1538-3881/aa72df
Bottke, W. F., Jedicke, R., Morbidelli, A., Petit, J.-M., &
Gladman, B. 2000, Understanding the Distribution of
Near-Earth Asteroids, Science, 288, 2190,
doi: 10.1126/science.288.5474.2190
Bottke, W. F., Morbidelli, A., Jedicke, R., et al. 2002,
Debiased Orbital and Absolute Magnitude Distribution
of the Near-Earth Objects, Icarus, 156, 399,
doi: 10.1006/icar.2001.6788
Chapman, C. R., & Morrison, D. 1994, Impacts on the
Earth by asteroids and comets: assessing the hazard,
Nature, 367, 33, doi: 10.1038/367033a0
Chyba, C. F., Thomas, P. J., & Zahnle, K. J. 1993, The
1908 Tunguska explosion: atmospheric disruption of a
stony asteroid, Nature, 361, 40, doi: 10.1038/361040a0
Cook, N. V., Ragozzine, D., Granvik, M., & Stephens,
D. C. 2016, Realistic Detectability of Close Interstellar
Comets, ApJ, 825, 51, doi: 10.3847/0004-637X/825/1/51
Council, N. R. 2010, Defending Planet Earth:
Near-Earth-Object Surveys and Hazard Mitigation
Strategies (Washington, DC: The National Academies
Press), doi: 10.17226/12842
D’Abramo, G., Harris, A. W., Boattini, A., et al. 2001, A
Simple Probabilistic Model to Estimate the Population of
near-Earth Asteroids, Icarus, 153, 214,
doi: https://doi.org/10.1006/icar.2001.6650
Daly, R. T., Ernst, C. M., Barnouin, O. S., et al. 2023,
Successful kinetic impact into an asteroid for planetary
defence, Nature, 616, 443,
doi: 10.1038/s41586-023-05810-5
Damhuis, M., van der Wal, R., Frielink, H., et al. 2024, The
ground is the limit: epidemiology of skydiving accidents
over 25 years and in 2.1 million jumps in the Netherlands
with sub-analysis of injuries reported by medical
professionals in the past five years, World Journal of
Emergency Surgery, 19, doi: 10.1186/s13017-024-00535-w
Deienno, R., Denneau, L., Nesvorný, D., et al. 2025, The
debiased Near-Earth object population from ATLAS
telescopes, Icarus, 425, 116316,
doi: 10.1016/j.icarus.2024.116316
Dotson, J. L., Wheeler, L., & Mathias, D. 2024,
Consequences of asteroid characterization on the state of
knowledge about inferred physical parameters and
impact risk, Acta Astronautica, 222, 550,
doi: 10.1016/j.actaastro.2024.04.020
Francis, P. J. 2005, The Demographics of Long-Period
Comets, ApJ, 635, 1348, doi: 10.1086/497684
Giorgini, J. D. 2015, in IAU General Assembly, Vol. 29,
2256293
Granvik, M., Morbidelli, A., Jedicke, R., et al. 2016,
Super-catastrophic disruption of asteroids at small
perihelion distances, Nature, 530, 303,
doi: 10.1038/nature16934
—. 2018, Debiased orbit and absolute-magnitude
distributions for near-Earth objects, Icarus, 312, 181,
doi: 10.1016/j.icarus.2018.04.018
Grav, T., Mainzer, A. K., Masiero, J. R., et al. 2023, The
NEO Surveyor Near-Earth Asteroid Known Object
Model, PSJ, 4, 228, doi: 10.3847/PSJ/ad072e
Greenstreet, S., Ngo, H., & Gladman, B. 2012, The orbital
distribution of Near-Earth Objects inside Earth’s orbit,
Icarus, 217, 355, doi: 10.1016/j.icarus.2011.11.010
Hammel, H. B., Beebe, R. F., Ingersoll, A. P., et al. 1995,
HST Imaging of Atmospheric Phenomena Created by the
Impact of Comet Shoemaker-Levy 9, Science, 267, 1288,
doi: 10.1126/science.7871425
Harris, A. W. 1998, Evaluation of ground-based optical
surveys for near-Earth asteroids, Planet. Space Sci., 46,
283, doi: 10.1016/S0032-0633(97)00036-6
Harris, A. W., & Chodas, P. W. 2021, The population of
near-earth asteroids revisited and updated, Icarus, 365,
114452, doi: 10.1016/j.icarus.2021.114452
Harris, A. W., & D’Abramo, G. 2015, The population of
near-Earth asteroids, Icarus, 257, 302,
doi: 10.1016/j.icarus.2015.05.004

11
Heinze, A. N., Denneau, L., Tonry, J. L., et al. 2021, NEO
Population, Velocity Bias, and Impact Risk from an
ATLAS Analysis, PSJ, 2, 12, doi: 10.3847/PSJ/abd325
Jedicke, R., Granvik, M., Micheli, M., et al. 2015, in
Asteroids IV, ed. P. Michel, F. E. DeMeo, & W. F.
Bottke (University of Arizona Press), 795–813,
doi: 10.2458/azu uapress 9780816532131-ch040
Ma, X., Bonaparte, S., Corbett, P., et al. 2023, Rabies
surveillance in the United States during 2021, Journal of
the American Veterinary Medical Association, 261, 1045,
doi: 10.2460/javma.23.02.0081
Mainzer, A., Grav, T., Bauer, J., et al. 2011, NEOWISE
Observations of Near-Earth Objects: Preliminary Results,
ApJ, 743, 156, doi: 10.1088/0004-637X/743/2/156
Maron, B. A., Haas, T. S., & Maron, B. J. 2007, Sudden
Death from Collapsing Sand Holes, New England Journal
of Medicine, 356, 2655, doi: 10.1056/NEJMc070913
Mathias, D. L., Wheeler, L. F., & Dotson, J. L. 2017, A
probabilistic asteroid impact risk model: assessment of
sub-300 m impacts, Icarus, 289, 106,
doi: 10.1016/j.icarus.2017.02.009
Morrison, D., Harris, A. W., Sommer, G., Chapman, C. R.,
& Carusi, A. 2002, in Asteroids III, ed. J. Bottke, W. F.,
A. Cellino, P. Paolicchi, & R. P. Binzel (University of
Arizona Press), 739–754
Naidu, S. P., Chesley, S. R., Moskovitz, N., et al. 2024,
Orbital and Physical Characterization of Asteroid
Dimorphos Following the DART Impact, PSJ, 5, 74,
doi: 10.3847/PSJ/ad26e7
Nesvorný, D., Deienno, R., Bottke, W. F., et al. 2023,
NEOMOD: A New Orbital Distribution Model for
Near-Earth Objects, AJ, 166, 55,
doi: 10.3847/1538-3881/ace040
Nesvorný, D., Vokrouhlický, D., Shelly, F., et al. 2024a,
NEOMOD 2: An updated model of Near-Earth Objects
from a decade of Catalina Sky Survey observations,
Icarus, 411, 115922, doi: 10.1016/j.icarus.2023.115922
—. 2024b, NEOMOD 3: The debiased size distribution of
Near Earth Objects, Icarus, 417, 116110,
doi: 10.1016/j.icarus.2024.116110
Oort, J. H. 1950, The structure of the cloud of comets
surrounding the Solar System and a hypothesis
concerning its origin, BAN, 11, 91
Opik, E. J. 1958, On the Catastrophic Effect of Collisions
with Celestial Bodies, Irish Astronomical Journal, 5, 36
Peters, E. 2017, in The Oxford Handbook of the Science of
Science Communication (Oxford University Press),
388–398, doi: 10.1093/oxfordhb/9780190497620.013.42
Popova, O. P., Jenniskens, P., Emel’yanenko, V., et al.
2013, Chelyabinsk Airburst, Damage Assessment,
Meteorite Recovery, and Characterization, Science, 342,
1069, doi: 10.1126/science.1242642
Rabinowitz, D., Helin, E., Lawrence, K., & Pravdo, S. 2000,
A reduced estimate of the number of kilometre-sized
near-Earth asteroids, Nature, 403, 165,
doi: 10.1038/35003128
Rabinowitz, D. L. 1993, The Size Distribution of the
Earth-approaching Asteroids, ApJ, 407, 412,
doi: 10.1086/172524
Reinhardt, J. C., Xi, C., Wenhao, L., Manchev, P., &
Pate-Cornell, M. E. 2016, Asteroid Risk Assessment: A
Probabilistic Approach, Risk Analysis, 36, 244,
doi: 10.1111/risa.12453
Rumpf, C., Lewis, H. G., & Atkinson, P. M. 2016, On the
influence of impact effect modelling for global asteroid
impact risk distribution, Acta Astronautica, 123, 165,
doi: 10.1016/j.actaastro.2016.03.015
Schwamb, M. E., Jones, R. L., Yoachim, P., et al. 2023,
Tuning the Legacy Survey of Space and Time (LSST)
Observing Strategy for Solar System Science, ApJS, 266,
22, doi: 10.3847/1538-4365/acc173
Shoemaker, E. M. 1983, Asteroid and Comet Bombardment
of the Earth, Annual Review of Earth and Planetary
Sciences, 11, 461,
doi: 10.1146/annurev.ea.11.050183.002333
Simonsen, C., Thorsteinsson, K., Mortensen, R. N., et al.
2019, Carbon monoxide poisoning in Denmark with focus
on mortality and factors contributing to mortality, PloS
one, 14, doi: doi:10.1371/journal.pone.0210767
Stuart, J. S. 2001, A Near-Earth Asteroid Population
Estimate from the LINEAR Survey, Science, 294, 1691,
doi: 10.1126/science.1065318
Thomas, C. A., Naidu, S. P., Scheirich, P., et al. 2023,
Orbital period change of Dimorphos due to the DART
kinetic impact, Nature, 616, 448,
doi: 10.1038/s41586-023-05805-2
Toon, O. B., Zahnle, K., Morrison, D., Turco, R. P., &
Covey, C. 1997, Environmental perturbations caused by
the impacts of asteroids and comets, Reviews of
Geophysics, 35, 41, doi: 10.1029/96RG03038
Tricarico, P. 2017, The near-Earth asteroid population from
two decades of observations, Icarus, 284, 416,
doi: 10.1016/j.icarus.2016.12.008
Watson, F. G. 1941, Between the planets
Wheeler, L., Dotson, J., Aftosmis, M., et al. 2024, Risk
assessment for asteroid impact threat scenarios, Acta
Astronautica, 216, 468,
doi: 10.1016/j.actaastro.2023.12.049

Placing the Near-Earth Object Impact Probability in Context

More Related Content

Similar to Placing the Near-Earth Object Impact Probability in Context (20)

More from Sérgio Sacani (20)

Recently uploaded (20)

Placing the Near-Earth Object Impact Probability in Context