Pleistocene Tornadoes and Windthrow Ecology

Unstable weather conditions spawn outbreaks of tornadoes.  Cold fronts collide with warm air causing the chilled air from the upper layer of the atmosphere to plummet, creating swirling winds of great destructive force.  Tornado intensity is classified according to the Fujita scale or F scale for short.  Tornado wind speeds range from less than 73 mph (an F0 tornado) to estimated wind speeds of 261 mph-318 mph (an F5 tornado).  One of the largest outbreaks of tornadoes in recorded history occurred in early April, 1936.  At least 12 tornadoes struck the south from Tupelo, Mississippi to Anderson, South Carolina.  A tornado from this system that hit Tupelo left a path of destruction 15 miles long.  Another tornado from this storm traveled 50 miles from Alabama to Tennessee.  Two tornadoes merged in Gainesville, Georgia, killing 200 people in a factory and a department store.  Overall, this storm system wiped out 454 human lives.

F5 tornado in Oklahoma.  I hypothesize storms and tornadoes were much more frequent and severe during some Pleistocene climate phases than they are today, but they may have been less severe during others.

I hypothesize tornado frequency and intensity was greater during some climatic phases of the Pleistocene than it is today.  As far as I can determine, no scientist has ever published a study of paleotornado frequency, probably because there just isn’t any method to collect data about past transient phenomena. Incidentally, I invented the term, paleotornado, in case a scientist figures how to study them.  My hypothesis is conjecture, but I am confident it is correct.  I base it on 3 lines of indirect evidence.

a) Data from ice cores in Greenland shows average annual temperatures fluctuated dramatically during Ice Ages.  There was an alternating cycle of sudden warm spikes in temperature that melted ice dams which in turn released glacial meltwater and icebergs into the ocean, shutting down the gulf stream.  This caused an equally sudden reversal in temperatures.   By comparison today’s climate is relatively stable, yet even with a stable climate, tornadoes form with regularity. When climate changed more rapidly in the past, it seems logical to assume there was an increased frequency of colliding warm and cold weather fronts.  I believe the middle south was an Ice Age tornado alley.  Temperatures in south Florida and the Gulf Coast were warmer than they are today because oceanic circulation ceased and warm water stayed in the Caribbean, but the upper south was only a few hundred miles from the Laurentide Ice Sheet that covered Canada and New England.  Cold fronts blowing off the Ice Sheet met warm fronts originating from the Gulf of Mexico in what must have been an exceptionally stormy transition zone.

b) An unusually cold phase of climate, known as The Little Ice Age, occurred between 1310-1850.  Anecdotal historical references suggest storms were more frequent and intense during this time period.  In Europe several storms killed hundreds of thousands of people.  The Little Ice Age is a tiny blip compared to the climate fluctuations of the Wisconsinian Ice Age as recorded from Greenland ice core data.

c) Geological evidence suggests river flooding in southeastern North America was much more severe during the early Holocene (11,000 BP-6,000 BP).  These massive floods caused supermeandering river patterns.  An increase in river flooding indicates an increase in storm activity and hence tornadoes.

Windthrows open up the forest canopy and dramatically change the local ecology.

Tornadoes, thunderstorm downbursts, and hurricanes have a profound impact on forest ecosystems and may be a primary driver of evolutionary relationships.  Areas of forest felled by wind are known as windthrows among ecologists.  Tornadoes can travel for many miles, and they leave long scars of fallen and splintered trees that can be seen in satellite and aerial photographs.  These long windthrows create gaps in the canopy where shade intolerant species can thrive.  In southeastern North America canopy gap formation is beneficial for oak, pine, persimmon, sumac, grapevine, blackberry, composites, and grasses.  Windthrows can become tangles of luxuriant vegetation that provide forage and cover for forest edge species such as whitetail deer, cottontail rabbits, and ruffed grouse.  Fallen rotting timber attracts beetles, food for woodpeckers and other birds.  The extinct ivory-billed woodpecker formerly relied on vast tracts of timber with freshly created windthrows from annual storms.  Unlike extant woodpeckers, they depended upon early colonizing, shallow burrowing beetles.  Snakes and lizards lay their eggs in rotting timber.  Bears tear up these logs, looking for beetle larva, termites, and reptile eggs.  The pits created when trees are uprooted fill with water following heavy rains, and they serve as breeding pools for amphibians.  Most of the organisms that live in southeastern North America evolved to thrive in canopy gaps resulting from wind storms.  Plants able to resprout after sustaining wind damage have a competitive advantage over those species easily uprooted and killed, and the animals that browse and can digest those plants also enjoy a competitive advantage.

One study estimated wind felled 20 square miles of forest per year in pre-settlement forests of Wisconsin.  They also estimated the recovery time for northern hardwood-hemlock forests to erase the windthrow scar is 1210 years.  The recovery time in southeastern forests is probably quicker due to the longer growing season. A tornado can leave a long-lasting impact on the landscape, and wind may be a critical element, along with megafauna foraging and fire, that may explain why Ice Age environments were so much more open than they were in late Holocene environments.

Reference:

Canham, Charles; and Orie Loucks

“Catastrophic Windthrow in the Presettlement Forests of Wisconsin”

Ecology 1988

Advertisement

Pleistocene Prairie Chicken (Tympanuchus cupido) Fossils Found in Southern States

Photo of greater prairie chicken from google images.

http://www.youtube.com/watch?v=H7QBAqjyi5k&feature=player_detailpage#t=10s

Youtube video of greater prairie chickens drumming on leks to attract mates.

During the Pleistocene prairie chickens were a common bird, at least locally, in southeastern North America.  Fossils of this species have been recovered from sites in Tennessee, Georgia, Alabama, and Florida.  All of these fossils come from cave deposits where hawks and owls originally dropped their remains while feeding.  Bones from at least 4 individual prairie chickens were excavated from Kingston Saltpeter Cave in Bartow County, Georgia, suggesting they were more than just an occasional species, even though most of the bird fossils found there were from woodland species.

There were 3 subspecies of prairie chicken: the greater prairie chicken (T. cupido americanus) which formerly ranged thoughout the tall grass prairie region but has been declining; the endangered Attwater’s (T. cupido attwateri) which lives on the coastal prairies of Texas; and the extinct heath hen (T. Cupido cupido) which lived along the coast of New England.

Specimen of a heath hen. I think heath hens should be considered a different species from the greater prairie chicken based on habitat preference and a DNA study.  Heath hens differed slightly in appearance as well.  They have a more reddish hue, their tarsi are shorter, and they have 5 neck feathers instead of 10.

The difference in habitat preference between greater prairie chickens and heath hens is astonishing.  Despite being considered the same species, their habitat preferences were the exact opposite.  Prairie chickens need completely open grassland of at least 130 acres in extent or their populations will decline.  They can’t even tolerate forest edges.  Trees provide perching platforms for predatory raptors that decimate them.  By contrast heath hens inhabited impenetrable coastal thickets consisting of bayberry, blueberry, beach plum, pine, and shrub oak.   A DNA test shows there were 6 degrees of mutational differences  between prairie chickens and heath hens.  The authors of this paper (referenced below) were hesitant to suggest a new species because their study was based on specimens of the last individuals which were residents of the heath hen’s final refuge on Martha’s Vinyard.  No museum specimens of mainland heath hens exist.  Therefore, the widely differing genetic lineage might be just for the final isolated island population.

Market hunters destroyed heath hen populations on the mainland (reportedly they were tasty).  In a desperate attempt to bolster the final intact population of heath hens on Martha’s vinyard, naturalists introduced greater prairie chickens, but their efforts failed because the introduced birds couldn’t adapt to the different environent.  A fire in 1916 followed by heavy raptor predation and introduced diseases wiped them out by 1932. The Nature Conservancy is considering introducing prairie chickens to Martha’s Vinyard, but this plan should be abandoned–I think it’s doomed to failure because they’re not the same species.

I believe heath hens were genetically isolated from greater prairie chicken populations during the LGM.  A lobe of the Laurentide glacier separated the two populations.  Heath hens survived on the unglaciated continental shelf and islands along the New England coast and gradually adapted to thickets rather than grasslands in a process that took thousands of years.  There’s not much chance of greater prairie chickens being able to adapt to coastal thickets in just one generation.

Pleistocene prairie chicken populations in southeastern North America were likely more adapted to grasslands, like modern day greater prairie chickens, though they may have had more variation in habitat tolerance than their modern descendents because the overall population of the species was so widespread.  Grasslands were common in the south during Ice Ages, especially during cold arid stadials, and this habitat persisted during interstadials as well.  Though most of the fossils of bird species found in Georgia, Alabama, and Tennessee are from woodland species, prairie chickens are not the sole grassland species.  Fossils of upland sandpipers (Bartramia longicauda) and magpies (Pica pica) also consistently are represented in avian fossil records here, indicating the presence of extensive grasslands.

I came across a study of presettlement land surveys in southeastern Arkansas that shows how grassland habitat large enough to support stable prairie chicken populations could have been distrbuted in the mostly woodland dominated environments of north Georgia and Alabama during the Pleistocene.

Map of Ashley County based on surveyor information from the below referenced paper written by Don Bragg.  There were sizeable prairies within the mostly wooded landscape.

Before European settlement the Ashley County region of Arkansas consisted mostly of upland pine-hardwood forests and bottomland swamps.  However, there were extensive tracts of grasslands large enough for early surveyors to name.  As I noted earlier, studies show that prairie chickens require treeless grasslands of at least 130 acres or their populations will decline until they eventually are extirpated.  Surveyors found at least 5 prairies in Ashley County.  Pine Prairie was 6800 acres, Twin Prairie was 2300 acres, Fountain Prairie was 2000 acres, Brushy Prairie was 500 acres, and Little Prairie was 300 acres.  A map of what’s now Bartow County, where prairie chicken fossils were found, would have had a similar distribution of prairies in a landscape composed more of mixed pine and oak forests and riverine woods.

In Ashley County (and probably the Pleistocene upper south) there were 4 different kinds of prairies.  Upland prairies are created and maintained by frequent fire and megafauna grazing.  Bottomland prairies form from a combination of fire and flood, both of which kill trees.  Alkaline prairies are areas with poor drainage.  As rainwater evaporates rather than draining away, basal salts accumulate, making it difficult for trees to grow, so grass dominates.  These are also known as lick prairies because ungulates are attacted to the accumulation of mineral salts.  Oak barrens are like savannahs with widely spaced post oaks and shortleaf pines.  Light fires kill all but the most fire tolerant species of trees here.  A rare unusual type of barren–hickory/dogwood–occurs in this region too.

Surveyors mapped the Ashley County district between 1818-1855.  They didn’t count every single tree, so studies based on their data are inexact but give good general information.  They marked plats of land by choosing and marking several witness trees on each plat.  The kinds of trees they marked are preserved in written records.  Black oak made up 18%, pine 17%, post oak 11%, white oak 9%, hickory 7%, sweetgum 7%, and all other species 31%.  An exact list can be found on a link to the paper in the references below.  The surveyors didn’t use the largest specimens as “witness” trees, but they still recorded some large trees, including a 12 foot in diameter cypress, and 6 foot in diameter black oak and loblolly pine.  Wild peach and apple trees occurred near abandoned Indian villages.

Ashley County surveyors noted a number of interesting landsapes created by disturbances which I summarize below.

Map of windthrows in presettlement Ashley County from the below referenced paper.  Some of the tracts were obviously made by tornadoes.

Three areas of 2200, 1700, and 400 acres consisted of fallen timber overgrown with vines and brush.  Many smaller areas of fallen trees less than 20 acres in extent were also recorded.  Tornadoes, thunderstorm down bursts, and ice storms caused these natural formations (or malformations).  Fallen trees provide all sorts of habitat and forage opportunities for many species of animals. 

During winter floods near rivers made it difficult for surveyors to do their work.  Floods killed trees and created standing deadwood.  Fallen woody debris often blocked rivers, preventing navigation. 

Surveyors recorded devastating unchecked fires.  One found an area of burned land 6 miles long. 

The earthquake of 1811-1812 formed elliptical depressions that surveyors called “earthquake swamps.”  The land actually sank below the water line, killing every tree in the depression. 

Surveyors reported pimple mounds–high circular swells of 6-30 feet in length.  Geologists later studied these mounds and determined they were formed 700-2400 years ago when severe extended droughts left many places bare of vegetation.  Winds blew sand into little dunes that later became covered with vegetation after the rains returned.

Photo of a pimple mound from google images.

References:

Bragg, Don C.

“Natural Presettlement Features of the Ashley County, Arkansas Area”

American Midland Naturalist (2003) 149 (1-20)

http://www.srs.fs.usda.gov/pubs/ja/ja_bragg003.pdf

Packovacs, Eric P. ; et. al.

“Genetic Evaluations of a Proposed Introduction: The Case of the Greater Prairie Chicken and the Extinct Heath Hen”

Molecular Ecology (2004) 13 1759-1769

http://fds.duke.edu/db/attachment/1024

Ryan, Mark R.

“Breeding Ecology of Greater Prairie Chickens (Tympanuchus cupido) in Relation to Prairie Landscape Configuration”

American Midland Naturalist (July 1998) 140 (1)