Science and Culture: Journal entries, maps, and photos help ecologists reconstruct ecosystems of the past


SCIENCE AND CULTURE

Journal entries, maps, and photos help ecologists
reconstruct ecosystems of the past
Carolyn Beans, Science Writer

Ilka Feller has a penchant for mangrove hunting. Since
the early 2000s, Feller, an ecologist at the Smithsonian
Environmental Research Center, has periodically gone
road-tripping in search of the northernmost mangrove
tree in Florida. The red mangrove, Rhizophora man-
gle, has draping roots that hold fast in the tide. But all
three of Florida’s mangrove species anchor them-
selves and whole ecosystems—protecting the shore-
line from erosion, forming nurseries for fish, and
sequestering carbon.

Globally, mangroves are threatened by coastal
development and shrimp farming. “More than half of
the world’s mangroves have already been destroyed,”

says Feller. Yet in Florida, Feller kept finding new man-
groves farther and farther north, “solitary plants that
are out there in this forever saltmarsh.”

As part of a 2014 study, Feller’s then postdoc, Kyle
Cavanaugh, analyzed satellite images taken from
1984 to 2011 and saw that mangroves were indeed
advancing northward (1). The spread was associated
with a reduced frequency of extreme cold events. Cli-
mate change, it seemed, was replacing marshland, an
important temperate ecosystem, with mangroves, an
important subtropical and tropical one. But had they
really revealed a trend pointing in one direction, or
was the recent expansion of mangroves part of a

Ecologist Ilka Feller used archival records to investigate whether mangroves in Florida, like these in Indian River Lagoon,
are expanding into new territories or returning to a historical range. Image credit: Ilka Feller.

Published under the PNAS license.

13138–13141 | PNAS | December 26, 2018 | vol. 115 | no. 52 www.pnas.org/cgi/doi/10.1073/pnas.1819526115

S
C
IE

N
C
E

A
N
D

C
U
L
T
U
R
E

D
o
w

n
lo

a
d
e
d
 a

t 
C

a
rn

e
g
ie

 M
e
llo

n
 U

n
iv

e
rs

ity
 o

n
 A

p
ri
l 5

, 
2
0
2
1
 

http://crossmark.crossref.org/dialog/?doi=10.1073/pnas.1819526115&domain=pdf
https://www.pnas.org/site/aboutpnas/licenses.xhtml
https://www.pnas.org/cgi/doi/10.1073/pnas.1819526115


continual cycle of mangrove and saltmarsh replacing
one another? Feller needed more than just conven-
tional scientific data to find out.

So she dug into Florida’s past using tools more
typical of historians. She found mangroves mentioned
in the texts of famed naturalist John Muir, uncovered
data in tourist photos, and parsed the writings of war
surgeons and beekeepers.

Feller and a growing group of ecologists are
piecing together historical records to understand
how ecosystems have changed over timescales that
extend beyond the memories of the current genera-
tion of researchers and residents. These portals to the
past can help researchers better predict and plan for
the future. But as historians already know well, ecol-
ogists first have to figure out how to find these records
and whether to trust them.

Reading Between the Lines
After reading Feller’s article on mangrove range ex-
pansion (1), US Geological Survey researchers Chan-
dra Giri and Jordan Long published a letter to the
editor noting that, because Feller only went back to
1984 when the satellite imaging began, she couldn’t
say definitively whether the northward spread was
really just a return to a historical range (2). Feller
responded that, either way, freezing determined the
range (3). Still, she was curious.

At first, Feller wasn’t sure how to approach histor-
ical records. She consulted textbooks on historical

ecology, a growing field defined in a recent review as
“the study of nature over time” (4). Historical ecolo-
gists use archival sources as well as biological ones,
such as pollen records and tree rings, often with the
ultimate goal of using a landscape’s history to illumi-
nate its potential. “We can draw on our understanding
of the past to creatively reimagine what is possible in
the future,” says review author Erin Beller, an envi-
ronmental researcher and historical ecologist at the
San Francisco Estuary Institute and PhD candidate at
the University of California, Berkeley.

Researchers have drawn on historical records to
explore past landscapes since as early as the 18th
century (5). The field took off with North American
ecologists in the 1990s (5), as researchers used archi-
val records to explore historical changes in a range of
ecosystems—from shifts in vegetation since the es-
tablishment of Yellowstone National Park (6) to de-
clines in green turtle abundance in Caribbean coastal
ecosystems since Columbus (7).

Feller began her historical ecology exploration by
searching the used books section of Amazon and
online holdings of historical documents. She soon
stumbled on an 1823 book by Charles Vignoles, a
surveyor for the city of St. Augustine, FL (8). In one
passage, he says nearby Ponce Inlet “affords water for
vessels drawing ten feet: the anchorage is good inside
and on the south shore a vessel may lay alongside and
make fast to the mangroves.” A book on historical
ships revealed that a vessel “drawing ten feet” meant

Photographs of Fort Matanzas National Monument, located south of St. Augustine, FL, offer visual evidence that the
mangrove’s range has expanded and contracted multiple times. Mangroves appear to varying degrees in photos taken in
the early 1900s (Top Left), 1934 (Top Right), 1981 (Bottom Left), and September 2018 (Bottom Right). Archival images
credit: State Library and Archives of Florida. 2018 image credit: Matthew Hayes (Villanova University, Villanova, PA).

Beans PNAS | December 26, 2018 | vol. 115 | no. 52 | 13139

D
o
w

n
lo

a
d
e
d
 a

t 
C

a
rn

e
g
ie

 M
e
llo

n
 U

n
iv

e
rs

ity
 o

n
 A

p
ri
l 5

, 
2
0
2
1
 



a boat 100–150 feet long. And that meant tying up to
a pretty large, robust tree.

In the journal of US Army surgeon Jacob Rhett
Motte, Feller found evidence that freeze events could
shrink the range. Motte described traveling by boat in
1836 during the Second Seminole War in the same
region that Vignoles mentioned more than a decade
earlier—except that in Motte’s account, the man-
groves are dead (9). Feller notes that 1835 is still on
record as including the most severe cold event in
Florida’s history.

Looking to photographic evidence, Feller and her
team searched Wikimedia and the online archives of
the Florida Division of Library and Information Services
for images of Fort Matanzas National Monument, a
fort built by the Spanish in the mid-18th century that
sits on a barrier island south of St. Augustine. Tourists
have taken photos of this monument since the early
1900s. In some photographs, mangroves dominate. In
others, it’s saltmarsh. “This was months of struggling
to find all of these little bits and pieces,” says Feller.
“The connection is not made for you. It’s like reading
between the lines.”

Advance and Retreat
Cavanaugh, now an ecologist and geographer at the
University of California, Los Angeles, placed these
findings on a map using GIS Software. Looking at the
points, they now know that the current northward

expansion of mangroves has not, in fact, reached its
historical limit. Their proof comes from the writings of
John Muir. In 1867, Muir caught his first glimpse of
Florida by boat, “a flat, watery, reedy coast, with
clumps of mangrove” (10). Just south of Georgia, Muir
was about 20 kilometers north of Feller’s current
northernmost record.

The current hypothesis of Feller and Cavanaugh’s
team is that hurricanes sweep the mangrove propa-
gules and, hence, the trees’ range northward, whereas
freeze events drive it back southward (11). Gleaning
from all the records they’ve sifted through, the team
now estimates that the mangrove range has con-
tracted and then expanded three or four times since
the early 1800s. The shifts occurred quickly, says
Cavanaugh. “A single severe freeze can be enough to
shift the system back to saltmarsh dominance.”

Cavanaugh recently built a model that predicts
past changes in mangrove abundance back to 1850
for an area near Fort Matanzas National Monument.
The model employs temperature records, experi-
mental data on cold tolerance, satellite images taken
since the 1980s, and aerial images reaching back to
the 1940s. When he compared the results with

historical records Feller had collected, Cavanaugh
found close agreement. With the historical records
lending confidence in the model, Cavanaugh then
modeled future mangrove abundance. “One out-
standing question is: Is this more recent shift toward
mangrove dominance going to be made more per-
manent because of climate change?” says Cavanaugh.
His results, which he plans to publish soon, suggest
that it will.

A Trove of Data
Ecologists might delve into historical archives for
myriad reasons. Beller says history offers a “trove”
of data that she and others at the San Francisco
Estuary Institute use to help nonprofits, private
companies, and government agencies see restora-
tion opportunities. In Silicon Valley, her team dis-
covered that, before around 1900, willow groves
covered more than 2,000 acres. The group found
written records and maps of willows, including some
from before the Gold Rush, and a drawing of pic-
nickers in a willow grove in what is now downtown
San Jose, CA. Google, inspired by these findings,
recently removed more than 100 parking spaces to
expand a rare existing willow grove on its campus in
Mountain View, CA.

To gauge changes in seasonal events, such as
flowering and migration, plant ecologist Richard Pri-
mack of Boston University put out a call for local his-
torical datasets that might capture changes in
phenology (12). He and his team learned that Henry
David Thoreau had made handwritten tables of the
date of first flowering for more than 500 species in
Concord, MA from 1851 to 1858. By combining
Thoreau’s and botanist Alfred Hosmer’s notes with
their own observations, the team discovered that
plants in Concord now flower 10 days earlier, on av-
erage, than during Thoreau’s time (12).

Looking to reveal long-term changes in marine eco-
systems, marine historical ecologist Loren McClenachan
of Colby College uses archival records to estimate
species’ population sizes before they experienced
any appreciable human influence. Identifying these
“historical baselines” can help managers make re-
covery targets closer to species’ actual potentials (13). In
a 2017 study, her team used 18th-century British nautical
charts to estimate changes in the coral reef area in the
Florida Keys (14). Typically, coral decline is measured
by the percentage of loss of coral in existing reefs. But
McClenachan’s team discovered that about half of the
coral observations marked on these charts are in pla-
ces where coral reefs no longer exist today.

Outside the Lens
For ecologists, historical records are both a boon and
a challenge. The first hurdle is simply finding the re-
cords. Students of McClenachan’s recently asked his-
torical societies in Maine for images of a seaweed
called rockweed but came back empty-handed. When
going to archivists, ecologists need be careful how
they frame the question, explains McClenachan. His-
torical societies don’t typically organize their photo

“This was months of struggling to find all of these little
bits and pieces. The connection is not made for you. It’s
like reading between the lines.”

—Ilka Feller

13140 | www.pnas.org/cgi/doi/10.1073/pnas.1819526115 Beans

D
o
w

n
lo

a
d
e
d
 a

t 
C

a
rn

e
g
ie

 M
e
llo

n
 U

n
iv

e
rs

ity
 o

n
 A

p
ri
l 5

, 
2
0
2
1
 

https://www.pnas.org/cgi/doi/10.1073/pnas.1819526115


archives according to marine species. When the stu-
dents instead asked for tourist pictures taken along
the coast, the rockweed appeared alongside smiling
vacationers.

But in some cases, the data may not exist. In ad-
dition to flowering time notes, Primack has also found
Thoreau’s records on bird and tree phenology (12).
Insect data, however, are proving more elusive. He’s
uncovered records on butterfly flight times from a lo-
cal butterfly club; other biologists have found bee
records using museum specimens. “But for a lot of
other insect groups, including flies and dragonflies,”
he says, “there is surprisingly little information.”

Then there’s the issue of veracity. “There is no
document that gives us a transparent view into the
past,” says Joseph Taylor III of Simon Fraser Univer-
sity, a Western North America and environmental
historian. “Even a photograph is literally framed by
the lens, and what is outside the lens is outside the
record.” Historians, he notes, are trained to place

evidence in context, thinking critically about where a
document was found, who produced it, and for what
audience. Some researchers follow this same practice,
says Taylor. But others, in an attempt to arrive at a
definitive data point, may run “roughshod over the
limitations of their resources,” pitfalls Taylor described
in a 2013 commentary (15). Biases abound, notes
McClenachan—in the case of fish-catch records, for
example, a fisherman who’s taxed on his catch might
be tempted to underreport.

Feller, though, isn’t finished. After retiring at the
end of this year, she’ll visit historical societies that
haven’t yet digitized their records. New discoveries
keep her motivated. Only recently, for example, she
found that beekeepers once placed hives in black
mangrove forests. When freezes hit, the beekeepers
wrote about them, hence offering another window
into how past freezes affected mangrove health.
“There’s a lot to be learned,” says Feller, “in reading
these obscure beekeeping journals.”

1 Cavanaugh KC, et al. (2014) Poleward expansion of mangroves is a threshold response to decreased frequency of extreme cold
events. Proc Natl Acad Sci USA 111:723–727.

2 Giri CP, Long J (2014) Mangrove reemergence in the northernmost range limit of eastern Florida. Proc Natl Acad Sci USA
111:E1447–E1448.

3 Cavanaugh KC, et al. (2014) Reply to Giri and Long: Freeze-mediated expansion of mangroves does not depend on whether
expansion is emergence or reemergence. Proc Natl Acad Sci USA 111:E1449.

4 Beller E, et al. (2017) Toward principles of historical ecology. Am J Bot 104:645–648.
5 Szabó P (2015) Historical ecology: Past, present and future. Biol Rev Camb Philos Soc 90:997–1014.
6 Kay CE (1995) Ecosystems then and now: A historical-ecological approach to ecosystem management. Proceedings of the Fourth
Prairie Conservation and Endangered Species Workshop, eds Willms WD, Dormaar JF (Provincial Museum of Alberta, Edmonton,
Alberta, Canada), pp 79–87.

7 Jackson JBC (1997) Reefs since Columbus. Coral Reefs 16(Suppl 1):S23–S32.
8 Vignoles CB (2010) Observations Upon the Floridas (1823) (Kessinger Publishing, Whitefish, MO).
9 Motte JR (2017) Journey Into Wilderness: An Army Surgeon’s Account of Life in Camp and Field During the Creek and Seminole Wars,
1836-1838, ed Sunderman JF (University Press of Florida, Gainesville).

10 Muir J (1916) A Thousand-Mile Walk to the Gulf, ed Badè WF (Houghton Mifflin Company, Boston).
11 Rodriguez W, Feller IC, Cavanaugh KC (2016) Spatio-temporal changes of a mangrove-saltmarsh ecotone in the northeastern coast of

Florida, USA. Glob Ecol Conserv 7:245–261.
12 Primack RB, Miller-Rushing AJ (2012) Uncovering, collecting, and analyzing records to investigate the ecological impacts of climate

change: A template from Thoreau’s Concord. Bioscience 62:170–181.
13 McClenachan L, Ferretti F, Baum JK (2012) From archives to conservation: Why historical data are needed to set baselines for marine

animals and ecosystems. Conserv Lett 5:349–359.
14 McClenachan L, O’Connor G, Neal BP, Pandolfi JM, Jackson JBC (2017) Ghost reefs: Nautical charts document large spatial scale of

coral reef loss over 240 years. Sci Adv 3:e1603155.
15 Taylor JE, III (2013) Knowing the black box: Methodological challenges in marine environmental history. Environ Hist 18:60–75.

Beans PNAS | December 26, 2018 | vol. 115 | no. 52 | 13141

D
o
w

n
lo

a
d
e
d
 a

t 
C

a
rn

e
g
ie

 M
e
llo

n
 U

n
iv

e
rs

ity
 o

n
 A

p
ri
l 5

, 
2
0
2
1