ArticlePDF Available

Abstract and Figures

Luck has played a big part in many of the world's great fossil discoveries. New models predict where the bones are and put serendipity in the backseat
No caption available
No caption available
Content may be subject to copyright.
Luck has played a big part in many of the world’s great fossil discoveries. New models predict where the bones are and put serendipity in the backseat
By Robert L. Anemone and Charles W. Emerson
sad0514Anem3p.indd 46 3/17/14 4:58 PM
May 2014, 47
Luck has played a big part in many of the world’s great fossil discoveries. New models predict where the bones are and put serendipity in the backseat
By Robert L. Anemone and Charles W. Emerson
fossil hunting in an area as
vast as Wyoming’s Great Divide Basin
(pictured here) has long been akin to
searching for the proverbial needle
in the haystack. But a new technique
improves the odds of nding ancient bones.
sad0514Anem3p.indd 47 3/17/14 4:58 PM
48 Scientifi c American, May 2014
     2009,    -- 
traveled a faint, two-track dirt road in southwestern Wyoming’s
Great Di vide Basin. The expedition was headed for an area known
as Salt Sage Draw in search of buried treasure: fossils dating to
between 55 million and 50 million years ago, at the start of the
Eocene epoch, when the ancestors of many modern orders of mam-
mals were beginning to replace the more archaic mammals that
had existed during the earlier Paleocene epoch. One of us (Anemone) had been leading fi eld
crews of anthropologists, paleontologists and geologists to the basin since 1994, and Salt Sage
Draw had proved a fruitful hunting ground over the years, yielding fossils at several localities.
Yet this time I was having trouble fi nding the site. It dawned on me that the road we were on was
not the one we had used in previous years. My error would turn out to be very fortunate indeed.
As the tracks began to disappear in the sagebrush and tall
grass, I stopped the caravan and walked a ways to see if I could
spot the road ahead. Rounding a small hill, I noticed an exten-
sive bed of sandstone in the near distance and the elusive road
right alongside it. Because sandstone in the Great Divide Basin
and many other sedimentary basins in the American West often
harbors fossils, I decided to spend some time searching these
deposits before we resumed our trip to Salt Sage Draw. After
about an hour of systematically scanning the rock on hands and
knees, my then graduate students Tim Held and Justin Gish
shouted that they had found a couple of nice mammal jaws. I
eagerly joined them. Fossil jaws with teeth are prized because
they contain enough information to identify the kind of animal
they came from, even in the absence of other parts of the skele-
ton, and because they reveal what the animal ate.
What came next can only be described as every paleontolo-
gist’s dream. My students had located a fossil “hotspot.” But
this was no ordinary hotspot with a handful of jaws or a few
dozen teeth and bones eroding out of the sandstone. Rather
they had found an extraordinary trove from which we have
now collected nearly 500 well-preserved jaws and several thou-
sand teeth and bones from more than 20 di erent fossil mam-
mal species that lived here approximately 50 million years
ago. We call the spot “Tim’s Confession,” and today it re mains
not only our best site in the Great Divide Basin but also one
of the richest caches of early Eocene mammals in the entire
American West.
Mine is hardly the fi rst team to make a major fossil discov-
ery more or less by accident. The history of paleontology is lit-
tered with such tales of serendipity. In fact, the ways that verte-
brate paleontologists attempt to locate productive fossil sites
have not changed much since the earliest days of our science.
Like the 19th-century pioneers of our fi eld, we use geologic and
topographic evidence to determine where we might have the
best chance of fi nding fossils eroding out of ancient sediments.
But beyond that, whether we hit pay dirt is still largely a matter
For more than a century paleontologists have used
geologic and topographic information to inform their
search for fossils. Yet the discovery of fossils is still
largely a matter of luck.
New computer models that look for hidden patterns
in satellite images can generate maps of where fossils
are likely to be located, thus helping fossil hunters nar-
row their search.
Ground truthing of such predictive maps in the
American West has shown that they do indeed im-
prove the odds of nding fossil sites. In theory, this ap-
proach could be used anywhere in the world.
Robert L. Anemone is a professor and head of the department of anthropology
at the University of North Carolina at Greensboro. A paleontologist interested
in human and primate evolution, he has conducted eldwork in Wyoming,
Montana, Kenya and South Africa.
Charles W. Emerson is an associate professor of geography at Western Michigan
University. In addition to his work on developing predictive maps to nd fossils, he is
collaborating on a project aimed at using satellite imagery to evaluate how economic
factors and environmental protection policies aff ect grazing lands in rural China.
sad0514Anem3p.indd 48 3/17/14 4:58 PM
May 2014, 49
of luck, and more often than not the hard work of looking for
fossils goes unrewarded.
Our experience at Tims Confession got me thinking about
whether there might be a better way to determine where my
field crew should spend its eorts searching for new fossil sites.
We knew that the fossils we were interested in occur in sand-
stone dating to between 55 million and 50 million years ago,
and we knew where in the basin some of these sedimentary lay-
ers were exposed and thus suitable for exploration. But although
that information helped to narrow our search somewhat, it still
left thousands of square kilometers of ground to cover and plen-
ty of opportunities to come up empty-handed.
Then one night in camp, an idea began to germinate. Out in
the field, kilometers away from the nearest source of light pollu-
tion, we often noticed satellites passing overhead. I wondered
whether we could somehow combine our expert knowledge of
the local geology, topography and paleontology
of the Great Divide Basin with a satellites view of
the entire 10,000-square-kilometer area to, in es -
sence, map its probable fossil hotspots. Perhaps
satellites could “see” features of the land invisi-
ble to the naked eye that could help us find more
sandstone outcrops and distinguish those that
contain accessible fossils from those that do not.
EyEs in thE sky
 , of course, have speculated
about whether satellite imagery might improve
our ability to find fossils in the field. As a specialist
in the fossil record of primate and human evolu-
tion, I knew that in the 1990s, Berhane Asfaw of
the Rift Valley Research Service and his colleagues
had used such images to identify rock exposures in
Ethiopia that might yield fossils of human ances-
tors. At around the same time, Richard Stucky of
the Denver Museum of Nature & Science demonstrated that dif-
ferent rock units in the fossil-rich Wind River Basin in central Wy-
oming could be distinguished and mapped based on analysis of
satellite imagery of the region. Both these projects involved collab-
orations between paleontologists and remote-sensing specialists
 and proved the value of such cross-disciplinary eorts.
But I wondered if there was a way to tease more information out
of the satellite images and thus better focus our search.
I turned to a geographer, the other author of this article
(Emerson), and the two of us soon sketched out a plan. We would
obtain freely available images of the basin from the Landsat 7
satellite and its so-called Enhanced Thematic Mapper Plus sen-
sor, which detects radiation reflected or emitted from the earths
surface in wavelengths spanning the electromagnetic spec-
trum—from the blue to the infrared—and represents it in eight
discrete spectral bands. The bands can be used to distinguish
paleontologists accidentallY found a trove of 50-million-
year-old fossils at a site dubbed Tims Confession in Wyoming’s Great
Divide Basin (right) in 2009. Among the nds were hundreds of well-
preserved mammal jaws (above). Computer models have since enabled the
team to focus its eorts in those areas most likely to yield fossils, including
this spot south of an extinct volcano known as the Boar’s Tusk (below).
sad0514Anem3p.indd 49 3/17/14 4:58 PM
soil from vegetation, for example, or to map mineral deposits.
Then we would develop a method that would allow us to charac-
terize the radiation profi les of known productive fossil localities
in the Great Divide Basin based on satellite imagery and see if
they shared a telltale spectral signature. If so, we could search
the entire Great Divide Basin from our computers to locate new
sites that share this spectral signature and thus have a high
probability of bearing fossils. We could then visit those places (as
well as places with di erent spectral signatures) in person and
ex haustively search them for fossils to test the model.
Determining whether our known fossil sites shared a dis-
tinctive spectral signature was no small task, because for each
site we had to assess the combination of values in six bands of
the electromagnetic spectrum provided by the Landsat data.
Our problem was essentially one of pattern recognition in mul-
tiple dimensions, something that humans do not do particularly
well but that computers excel at. So we enlisted a so-called arti-
cial neural network—a computational model ca pable of learn-
ing complex patterns.
Our artifi cial neural network revealed that the basins known
fossil sites do indeed share a spectral signature, and it was able
to easily tell these sandstone localities apart from other types of
ground cover, such as wetlands and sand dunes. But the model
had its limitations. Neural networks, by their very nature, are
analytical “black boxes,” meaning they can distinguish patterns,
but they do not reveal the actual factors that allow di erent pat-
terns to be distinguished. So whereas our neural network could
easily and accurately distinguish fossil localities from wetlands
or sand dunes, it could not tell us how the spectral signatures of
di erent land covers actually di ered in the six bands of the
Landsat data—information that could conceivably help us con-
duct a more targeted search. Another limitation of the neural
network approach is that it is based entirely on the analysis of
individual pixels. The problem is that the area of an individual
Landsat pixel, which measures 225 square meters, does not nec-
essarily correspond to the size of a fossil locality: some localities
are larger than an individual pixel; some are smaller. Thus, the
neural network’s predictions about the location and extent of
potential fossil sites (or a certain type of ground cover, for that
matter) do not always match up with reality.
To overcome these constraints, we needed to be able to ana-
lyze multiple adjacent and spectrally similar pixels and to sta-
tistically describe the distinctive spectral signature of the entire
area, whether it was a fossil site or a forest. We turned to a tech-
nique known as geographical object-based image analysis and
to commercially available, high-resolution satellite imagery in
which individual pixels were less than one meter in diameter.
Unlike an artifi cial neural network, this approach allows satel-
Treasure Map
Computer models can analyze satellite images of an area’s known
fossil sites to identify their shared radiation pro le. The models can
then assess the broader region to nd other spots that share that
profi le and thus may harbor fossils of interest. This technique
enabled the two of us to generate a predictive map of fossil lo calities
( red ) in the Great Divide Basin that helped to guide our search
for fossils there ( yellow ). Restricting our surveying to focus
on some of these areas ( blue ) greatly increased our success rate
in nding fossil sites ( green ), compared with that of previous
ex peditions conducted without such a map. R.L.A. and C.W.E.
For more photographs from expeditions to the Great Divide Basin, visit Scientifi
fossil sites
Survey path
Survey points
Success spots
sad0514Anem3p.indd 50 3/17/14 4:58 PM
May 2014, 51
lite images to be segmented into image objects—that is, groups
of spectrally homogeneous pixels—that can then be character-
ized by statistical parameters such as mean or median bright-
ness or texture. These image objects more closely match points
of interest on the ground, such as fossil sites or stands of forest.
Using this image-analysis technique, we were able to develop
an independent set of predictions about where to find fossils.
    yielded maps of the Great Divide Ba-
sin that pinpointed unexplored areas whose spectral signatures
most closely resembled those of the known localities. Although
the models exhibited a good degree of overlap in their predictions,
they also diverged in some cases. We chose to focus on those places
that both models identified as high-priority potential sites. Maps
in hand, we headed out to Wyoming during the summers of 2012
and 2013 to see if our models would lead us to new fossil caches
in the Great Divide Basin. Gratifyingly, they did exactly that.
The artificial neural network model turns out to be extremely
ecient at identifying sandstone deposits, which are almost
always worth exploring because so many of the ones in this
basin contain fossil vertebrates. One of the first sandstones it led
us to in July 2012 yielded a dozen fossils of characteristic Eocene
mammals, including the five-toed horse Hyracotherium,
the ear
ly primate Cantius
and se
veral other creatures belonging to an
extinct group of hoofed mammals known as the Condylarthra.
The neural network also guided us to several spots that yielded
aquatic fossil vertebrates, including fish, crocodiles and turtles.
Our geographical object-based image analysis model took us
to new sites, too. After a slow start in which the first three or
four places the model pointed us to gave up no fossils, we moved
to the northern part of the Great Divide Basin, near a place
called Freighter Gap, for a week of intensive “ground truthing
of our new technique. Graduate student Bryan Bommersbach,
who a week before had led us on a long hike to a place that was
entirely barren of fossils (we dubbed it “Bryan’s Folly”), took the
lead in choosing which areas to survey based on the models
predictions. Almost immediately, we began to find bones at
many of these locations. We searched for remains at 31 separate
places on the landscape that our model indicated were spectral-
ly similar to known localities and found vertebrate fossils at 25
of these places, which is a much higher success rate than is typ-
ical when surveying without the help of a predictive map. Mam-
mal fossils emerged from 10 of these localities, one of which
dates to the latest part of the Paleocene—an extremely rare find.
We have every reason to believe that predictive models akin to
the ones we developed will work in regions other than the Great
Divide Basin. In fact, they should work virtually anywhere in the
world. In theory, as long as one has satellite images of the region
in question and a handful of known fossil localities with which to
train the model, one can generate a custom map showing those
spots in the region that are likely to contain fossils of interest.
In a conservative test of this approach, we used the neural
network we developed for the Great Divide Basin to predict the
locations of fossil-bearing sedimentary deposits in the nearby
Bison Basin, which is known to harbor Paleocene mammal fos-
sils. (We did not train the model with fossil sites specifically from
Bison Basin, because it contains the same kinds of fossil depos-
its as the Great Divide Basin.) Encouragingly, our neural net-
work predicted the three most productive fossil
localities known in the Bison Basin. Thus, a field
crew exploring this vast area for the first time
using our predictive model would have had a far
better chance of discovering these sites than a
crew using traditional survey methods.
Our trial runs in 2012 and 2013 in Wyoming
showed that the use of satellite imagery in combi-
nation with geospatial predictive models greatly
increased the eectiveness of our fieldwork, help-
ing us to find more fossils in less time. But we still
have more to do. We are now focused on refining
our models to better characterize and dierenti-
ate the spectral signature of productive localities.
And we are working on ways to apply more con-
straints to our predictive models to limit the number of false
positive results in the maps we generate and thus improve our
ability to determine the highest-priority areas to survey.
We are convinced that with these tools we can put the future
of paleontological exploration on a more secure and scientific
footing and reduce the role of serendipity in finding important
fossils. Achieving that goal will be well worth the eort re -
quired. Piecing together the origin and evolution of life on earth
is far too interesting and important an endeavor to leave to
chance. And we cant aord to wait another 15 years to find the
next Tim’s Confession.
We searched for fossils at
31 separate places on the land-
scape that our model indicated
were spectrally similar to known
localities and found vertebrate
fossils at 25 of these places.
More to explore
GIS and Paleoanthropology: Incorporating New Approaches from the
Geospatial Sciences in the Analysis of Primate and Human Evolution.
R. L. Anemone, G. C. Conroy and C. W. Emerson in American Journal of Physical
Anthropology, Vol. 54, Supplement No. 53, pages 19–46; 2011.
Finding Fossils in New Ways: An Articial Neural Network Approach to
Predicting the Location of Productive Fossil Localities. Robert Anemone,
Charles Emerson and Glenn Conroy in Evolutionary Anthropology, Vol. 20, No. 5,
pages 169–180; September/October 2011.
An Articial Neural Network–Based Approach to Identifying Mammalian
Fossil Localities in the Great Divide Basin, Wyoming. Charles W. Emerson and
Robert L. Anemone in Remote Sensing Letters, Vol. 3, No. 5, pages 453–460; 2012.
For more on this research, including its funding source, visit Scienti
FroM our Archives
First of Our Kind. Kate Wong; April 2012.
sad0514Anem3p.indd 51 3/17/14 4:58 PM
ResearchGate has not been able to resolve any citations for this publication.
ResearchGate has not been able to resolve any references for this publication.