Historical Agriculture and Soil Erosion in the Upper Mississippi Valley Hill Country

Front matter

FOREWORD BY ANDREW GOUDIE

TG Dye : "rt IM RS te : i

STANLEY W. TRIMBLE

Dedication

First, I wish to dedicate this book to the late Dr. Stafford C.Happ and his colleagues, whose data and assistance were vital to this study. They were ahead of their time, and their contributions to soil conservation and to fluvial geomorphology are just now being more fully recognized.

Equally, I dedicate this book to my family, who so willingly supported me in the long endeavor described herein.

Thirdly, I dedicate it to the people of the Hill Country, who persevered in the adversity of a grim period, helped solve the problem, and continued to improve their countryside.

Contents

FigUTES ——————— RESSE xi PORE WO EEE E E EEE E E E bevstetss satin xxix lj rn I nre E EEEE EE SEEE E ETE E E M xxxi BOB TAPIA. M ——————— Xxxix Acknowledgments d ———————— xli litro duc HON o eee ehm i ertet retten bee a P esed be este epu o AAEE xliii Phiyslog raphy e ——— 5 Gur C —————————————————ÁÁÁ—— € 8 Murano m ————— 14 mu ————————————Á— € 19 CONCIUSION ss. ———————————————M 22 Barly Settlement d —————— 24 Historical crops and other agricultural land use... 33

(0 ———————————————Ó—M 33 Grazing and animal husbandry acier rmn rettet tarta Pe esee enn 37 Conservation: Tcu —————— aE 47

Chapter3 The systematic effects of historical agriculture on the physical

Foreword

Andrew S. Goudie University of Oxford

That great British Geographer, the late Sir Henry Clifford Darby (1909-1992), often asserted that the two most important thrusts of geography were historical geography and geomorphology. This was because of his concern with understanding the evolution of landscapes, and he himself wrote in detail about the evolution of some great examples of these, including the English Fenland (Darby, 1956). Since that time there has, building upon the work of G.P. Marsh (1864) in the nineteenth century, been a fine tradition of tracing the way in which humans have modified the face of the Earth by, for example, draining wetlands and clearing woodlands. Notable here was the magisterial work of the late Michael Williams (1970; 2003). Environmental history has become a fertile field. Stan Trimble has already shown this in the context of the Southern Piedmont in the USA (Trimble, 2008). This present study examines the ways in which another of the great landscapes of the USA has been transformed by human activities. It is based on almost 4 decades of dedicated research and has many of the characteristics of a gripping detective story, based on meticulous forensic investigation. It demonstrates the unfashionable virtue of long immersion in a region, of the use of archival material, of intensive fieldwork, and of the value and magnificent collection of repeat photographs. It is an exemplification of true geography, which makes a major contribution, by both methodology and substance, not only to historical geography but also to geomorphology, It will also be avidly devoured by local historians.

To gain an impression of the value of this work, one needs, paradoxically, to start by looking at the volume’s conclusions. It anwers the question as to why it was that the region’s farmers, unwittingly, caused landscape destruction within a few decades: a climate that was more erosive than Europe, the presence of steep, erodible slopes, and the nature of the crops grown. It also demonstrates how even with simple technology—horse-drawn implements—farmers can cause severe erosion and gully formation. It also raises the question as to whether land cover change or climate change was the main cause of the degradation that occurred and indicates very clearly that it was the former rather than the latter. However, another conclusion is that in addition to causing this spasm of degradation, the region's farmers, by instituting new land management and conservation strategies, were also responsible for 'the almost miraculous recovery of the region' at the time that there was an increase in agricultural productivity. It has been the change now that it is difficult to discern 'the environmental violence and human frustrations that once existed there.’ There were aided in this by well-designed government programs. The book also demonstrates the importance of forest-regrowth—the rebirth of America's forests—and its xxix

Xxx Foreword attendant environmental improvements. The geomorphologist will learn much from the complexity of stream processes that is evident and from the linkages that occur in different parts of the river basins between erosion and deposition. They will also be fascinated by the rather limited impact that the great floods of 2007 had on the landscape. There is a lesson here, and in the book as a whole, with respect to our current preoccupation with the effects of future anthropogenic climate changes. We will do well to remember the importance of future land cover changes as well.

The book's preface, however, must not be skipped. It is there, in a handful of graphic introductory paragraphs, that the justification of this remarkable work is outlined—the story of ‘the fall and recovery of paradise"

Darby, H.C. 1956. The Draining of the Fens. Cambridge: Cambridge University Press.

Marsh, G.P. 1864. Man and Nature. New York: Scribner.

Trimble, S.W. 2008. Man-induced Soil Erosion on the Southern Piedmont (2nd. edition). Ankeny, Iowa: Soil and Water Conservation Society.

Williams, M. 1970. The Draining of the Somerset Levels. Cambridge: Cambridge University Press.

Williams, M. 2003. Deforesting the Earth. From Prehistory to Global Crisis. Chicago: The University of Chicago Press.

Preface

Imagine that as a small child in the mid-19th century, your family left New England, or upstate New York, or Norway, or any of several places of origin in the United States or Western Europe. They settle in what is now called the Hill Country of southwestern Wisconsin, where your father builds your house in a small, beautiful valley with a cold, clear, ever-flowing spring. A narrow creek filled with brook trout flows near the house. Your neighbors and dear friends who migrated with you build a large house on a low terrace overlooking the broad, fertile floodplain of a larger stream nearby.

The decades pass. The land is well watered and bountiful. Crops are good and the amenities of “civilization” are increasingly available. Roads expand, and railroads and small industries such as mills and cheese factories and even breweries are constructed. Churches are built and other civic and social organizations are formed. Villages and small towns prosper. Life is good. The Hill Country is a sort of rural paradise, an Arcadia in America.

But by about the sixth or seventh decade of your life, things begin to change drastically. The once-clear spring, your only source of potable water, is no longer as dependable and sometimes dries up. The beautiful creek flowing past your house has become turbid. It dries up on occasion, but then rages when even moderate rainstorms bring down torrents of mud, sand, and boulders, all threatening your house and farmstead. Of course, the brook trout are long gone. Even after you have used the stream-transported boulders to build a protective dike, your house and farmyard often flood, and running the farm becomes difficult. Upland fields above the house, once so fertile and productive, now have deep gullies, some gushing even more water and debris into the farmyard. All of this makes it more difficult to farm, your income declines greatly, and so does your whole quality of life. You know of other people in similar situations, some of whom have abandoned their house and farmyard sites. Your family must decide whether to stay and fight the elements, or just move the farmstead to a higher place. Since you are now the family elder, they look to you for advice.

Your neighbors on the larger stream are having their problems, too. Once above all floods of the larger stream, their terrace location is now being flooded, bringing great hardship. Even worse, the floodplain is aggrading with silt and sand, bringing floodwaters ever closer and sometimes destroying their crops and fences and making the land less fertile. Floodplain pastures are increasingly swamped, and animals must now be pastured behind the house on the steep, fragile hillside, which begins to gully, bringing water and debris onto their terrace location and into their house. Their farm income decreases and their farmstead, once a showplace, now appears derelict. They, too, feel threatened and consider moving to higher ground or leaving the region altogether.

Matters on the general landscape are hardly better. Hillside gullies bring water and debris with every storm, sometimes closing roadways and railroads. Flooding tributaries rip away roadways, demolish bridges, and undercut telephone and power lines. Farther xxxi xxxii Preface downstream, bridges and roadways are being threatened with every storm, and some are even buried. Mills are washed away or buried so flour, cornmeal, and sawed lumber are more difficult to procure. People lose their homes and sometimes drown in floods. In some locations, entire villages are being flooded and even buried. A greatly increased proportion of township, county, and state tax money must go to road repair and for public safety in general, given the increasing hazards. With transportation and communications so tenuous, it becomes more difficult to obtain supplies, to market crops and milk, to communicate and socialize with distant friends, and even to go to church.

What sounds like the plot of a horror movie or a bizarre story by Franz Kafka was reality in much of the Hill Country along the upper Mississippi River. And, indeed, much of it happened within the lifetime of many people there today. It is a sad and often tragic story, but it is more than counterbalanced by the almost miraculous and mostly happy story of the recovery, again which took place within a lifetime. Both the fall and recovery of paradise are the stories I want to tell.

Background and perspective

Everyone knows about the buried civilizations of the ancient world. In the Middle East, soil erosion was so severe that cities were buried and ports became landlocked. Archaeologists have been excavating the Greek and Roman ruins for centuries, and we marvel at these unearthed treasures and the environmental processes that caused these interments. But very few know that civilizations in mid-America have been entombed, not over the time span of centuries and millennia, but during just short decades.

The Upper Mississippi Valley Hill Country, reaching from Hastings, Minnesota, downriver past Dubuque, Iowa (Figure 0.1) is one of the most beautiful landscapes in the world. It has many names—the Paleozoic Plateau, Coulee Country, and Bluff Country. The U.S. Department of Agriculture (USDA) terms it the Northern Mississippi Valley Loess Hills. Often perceived as an unglaciated island in a sea of glacial drift and thus generally termed the “Drifltess Area," it is better seen as an island of hilly country, largely unglaciated, surrounded by a sea of gently rolling landscape, all glaciated. Extravagantly praised by early explorers before European* settlement, it is now a serene, bucolic region of wooded hills, green vales (termed coulees), clear streams often filled with brook trout, and model farms. Locals, with good cause, call it "God's Country.” Like good music or poetry, it is an entity that sooths the soul and reassures one of the goodness of the world.

But it was not always so. In the early 20th century, the region went through a dark and desperate period of soil erosion, stream sedimentation, and flooding, much as happened millennia earlier in the Mediterranean area. Roads, bridges, farms, and villages were literally buried, but here in a much shorter time. Aquatic habitats disappeared as once-limpid brooks became turbid and began to resemble gravel roads or huge ditches. Deep, raw gullies fingered far up valleys and hillsides. Frequent floods covered lowland fields with water and sediment, and even blocked roads or eroded them away, often cutting off communication between communities. In a real sense, the landscape was truly out of control. All this was a disaster on the scale of a huge hurricane, a tsunami, or an earthquake—but instead of one huge event, it occurred in increasingly severe increments—a disaster in slow motion. One might think that the relatively slower development of this disaster would have given people more chance to escape it. But no one knew the eventual dimensions

* "European" here refers to the culture involved, especially the agriculture. It may include recent European immigrants, "old Americans," Native Americans, African Americans, and others.

Preface xxxiii

= x

S-O N S A

T ae Theilman \ Z Acad X P ver als Millville D g ae $e ue p winter | {Beaver UA

1 @WW Falls Ay \Galesville Elba Winonae

Stockton

WINONA

Fillmore i AA < HOUSTG N ! irai d MNesoja ILC UAE |o Bs OWA : kt Pu d NEN A | RICHLAND

Lansing

È eh $ Soldiers z Up? Village Creek. & Grove! Richland winESHIEK | ALLAMAKEE) $96" Mh Center Waterville ' wisonsin | ! Yellow CRAWPORD;

Volney i 1 lon. Prairie B E TSUSION Y o8 | v i Z McGregor GRANT i X Mey, i % 1 Clayton H . " . " " 1 Ikad i Upper Mississippi River wW AAN pis P —M Hill Country ó e | LAFAYETTE 1 SLittleport t pon eR d ig emn From Read (1940) and H. Johnson (1976) 1... bis i E KG}

— — -From Trewartha and Smith (1941)

Galena JO DAVIESS N 0 30 mi ———3—3—3 0 50 km

Figure 0.1 The Upper Mississippi River Hill Country. The delineations shown were used because of previous historical work done on the region. This is also quite similar to the delineation of the region into a Major Land Resource Area (MLRA) by the USDA-NRCS (Argabright et al., 1996). Termed by the USDA as the Northern Mississippi Valley Loess Hills, it would exclude most of Dunn County, Wisconsin, and Carroll County, Illinois, but include Clinton County, Iowa.

xxxiv Preface

Figure 0.2 The Great Flood of 1907 in lower Chaseburg, Wisconsin, the highest on record up to that time. The photo was made looking east on Mill Street near the front of the present Hideaway Tavern. Since that time, the floodplain has aggraded to a level about 4 or 5 ft (1.2-1.4 m) above the flood level of 1907. The remains of lower Chaseburg were purposefully demolished in 2009 because of the continued flooding.

of the problems—the magnitude in space and time—so they had little idea about how to react. Few people had the financial and logistical ability to do much about it, but many eventually had to react when the water and sediment literally came in the front door.

Because all of this seems so unbelievable when related in the abstract, just consider one case, that of Ole Martiny in the once-idyllic village of Chaseburg, Wisconsin, on Coon Creek just south of La Crosse. After living in Chaseburg most of his life and knowing the village’s collective observation and perception of past stream behavior, he decided in 1903 to build his house on a low alluvial terrace by the creek. On July 21, 1907, an unprecedented flood, the largest since European settlement, occurred, rising four feet (1.2 m) on his house (Figure 0.2). Believing this to be simply a quirk of nature, he promptly raised the house 4 ft (1.2 m), thinking that would put him above any future floods. But the flooding and sedimentation continued and in less than two decades, had driven him from the house, the foundation of which now lies about 10 ft (3 m) under the present floodplain (see Chapter 4).

Eventually, there was some salvation for the region. The human hand, which had brought such destruction here, was with time just as capable in restoring the same landscape. And so it happened: the second half of the 20th century, continuing even now, saw the restoration of this region, and it was done by the progeny of the original settlers, in some cases, perhaps, just barely within the lifetimes of those original settlers. Soil erosion was curtailed, scarred hillsides were healed, streams once again assumed their sylvan nature and appearance, and brook trout again bred and flourished in now-clear streams.

All is not perfect, however, because there is a painful legacy of this sediment. Onceproductive agricultural fields in bottomlands are now marsh or even open water. Towns, villages, homes, and agricultural fields in some places lie huddled behind dikes to mitigate

Preface XXXU flooding caused by aggraded channels and floodplains. Sediment, eroded from slopes perhaps a century or more ago, deposited along the stream, and now termed legacy sediment, is presently migrating and creating more problems in some downstream locations. In several ways, the Hill Country is a region trapped in its own history.

A person visiting the region or even having grown up there may well not know this story. Those who did knew it only in part and, in any case, are already history themselves. As we now admire a wooded hillside, we may not know that gullies once cut across it; as we now gaze at a stream, we may not know that it once flowed at a level many feet lower than now, or as we look across a floodplain, we may not know that a farm or even a village lies buried there. And even if we know that, do we know how and why it happened? To reveal such landscape secrets, often literally buried, is the role of historical geography. And that is what this book is about—revealing the secrets of past landscapes and how they evolved, and why.

My love of this quintessential American landscape must be clear to the reader by now. Although I'm not a native son, this is my second home in the world. More than just telling you about it, I want to equip you with the means to understand and really appreciate it. But more than that, I hope to give you the means of going out on the landscape and visualizing what happened, right on the ground where it happened. It is a thrill to see and experience a landscape in its historical totality. Professional geographers used to say that "one learns geography through the soles of one's shoes.” I still believe that. Even those who cannot physically visit the region can use topographic maps and earth images from the internet to be there virtually.

I emphasize that this study primarily concerns watersheds tributary to the Mississippi River and deals only to a small degree with the Mississippi River itself. For that, one is directed to the excellent studies by Burke (2000), Anfinson (2003), and especially by Fremling (2005).

The approaches and tools of the geographer are eclectic. This research project, over the past 39 years of my career, has employed methods and techniques ranging from hydraulic engineering to rural sociology, with much archeology thrown in. But my primary goal here is to show the human side of these profound landscape changes. Over the past 39 years, I and others, in many publishing outlets and often in technical terms, have shown how humans have effected the changes, but now the emphasis here is also on how all these landscape changes, in turn, have affected humans. Of course, I will have to discuss many of the physical aspects to make all this understandable, but I will hold that to a minimum and then make reference to the appropriate publications so that those interested persons can read further.

I hasten to point out that I am not the first to study these profound changes in this region. In particular, I mention the work of Carlos Bates and O. R. Zeasman in the 1920s, Stafford C. Happ and his associates and students in the 1930s and '40s, Richard Sartz and his colleagues in the 1960s and "70s, Hildegard Binder Johnson in the 1970s, and James C. Knox and his students starting in the early 1970s. My own work there dates from 1973. These and many other studies will be cited in due time, but this study will remain highly personal.

While I hope that professional earth scientists will find this work interesting and useful, my primary desire is to engage the interested and aware nonspecialist. Thus, there is a concerted attempt to make the work accessible to those in other fields. Professional earth scientists should consult the cited literature for more technical and comprehensive explanations and discussions. To further facilitate accessibility, I use both English (Imperial) and metric (SI) units of measure and provide a glossary of technical terms.

XXXVI Preface

This book has been in the making for a long time. My earlier studies of the Southern Piedmont (e.g., Trimble, 1969, 1970, 1974) prepared me well for this work because it was there that I fully realized the power of humans to drastically modify the landscape in both negative and positive ways. The effects of modern soil conservation measures, and the dedication of those who implemented them, impressed me greatly. I also learned much about physical processes on slopes and in streams and what drove those processes. A big lesson was the role that sediment storage gain and loss played in determining the sediment yield of streams (e.g., Trimble, 1975c, 1975d, 1976b). Perhaps most important, I saw in action what we now refer to as the "complex response," that is, different stream reaches respond in different ways and different times to what seem to be the same stimuli (Schumm, 1973). But the good news is that it is somewhat systematic and predictable. Discovering this was a great education in fluvial processes, as well as the human impact, and it carried over to the present study.

While working in the Piedmont, I met Dr. Stafford C. Happ (1905-1993), the person who, as a US Department of Agriculture (USDA) scientist, had designed many stream research projects in the Piedmont, the Hill Country, and elsewhere in the United States (Trimble, 2008). When advising me on the Piedmont work, he mentioned his pioneer work in Coon Creek, Wisconsin. As luck would have it, my first university teaching job was at the University of Wisconsin at Milwaukee. As soon as possible, I reconnoitered Coon Creek in depth and obtained a research grant from the US Geological Survey (USGS). Then, from several places, including the National Archives, I was able to dig out much of the relevant data. But the best thing I did was to ask Dr. Happ to advise me on the project. We began fieldwork in earnest in late May 1974 and for the next few summers, my family, my students, and I "camped" in a rented house either in Stoddard or in Coon Valley, Wisconsin. While nearly all of the local farmers gave us permission to enter their property, some were unsure about what we were doing out on their floodplains and in their streams, but when they saw us out from 7 am to 7 pm six days per week, they were convinced that this was serious business. Although he was 69 years old at the time we began, Dr. Happ often worked right along with us in temperatures that once went to 112°F (44°C) and sometimes in water up to our necks. And we were always pestered with insects like mosquitoes, sweat bees, horseflies, deerflies, chiggers, and deer ticks. Not only did he never complain, he could work harder and longer than any of us. When we finally quit for the day, he went to his cramped motel room and laboriously plotted the surveys we had done that day. While we did not always agree on research priorities and matters of precision, I must say that he was the best field person I’ve ever known, and the project would have been almost impossible without his not infallible but elephant-like memory concerning the location of profile ends, benchmarks, and physical conditions of the 1930s and ‘40s.

Happ’s research designs are given in detail in Trimble (2008), and briefly described in Chapter 3. These were highly labor intensive, being based largely on precisely cored and surveyed stream and valley cross-sectional profiles, and for the record, I wish here to briefly note what was involved in resurveying them. There were only a very few permanent benchmarks set, nearly all profiles were marked by blazed "witness" trees, and most benchmarks were merely common nails set low on trees. Such markings would be hard to recover in a dense forest, but there were very few trees in the bottoms during the late 1930s because the land was cultivated or grazed, so the markings were then fairly visible. The original intention had been to return within five or so years and resurvey, but World War II intervened and the projects were forgotten by all but Dr. Happ. By the mid-1970s, there was less grazing so that many bottomland areas had succeeded to undergrowth or young forest. This was 30 years before GPS so we had to find range ends by using

Preface xxxvii

Happ’s careful descriptions of trees and the surrounding landscape, and scaling from air photos on which Happ had precisely plotted the profiles. However, tree growth had covered the small nails and many marked trees had died or had been removed. Moreover, undergrowth was often so thick that even permanent markers such as steel pipes set in the ground were virtually invisible. Thus, finding some profiles was literally like looking for a needle in a haystack. On many occasions, I saw Dr. Happ spend hours with a small hand ax, carefully chipping away at the side of a tree trying to find the suspected nail, which he usually did, and I might add without disturbing the elevation on the nail.

To be useful, a resurvey had to be precisely on the old line so just finding the exact line was a major task. If we could find one end or any point along the line, we could use the precisely recorded azimuth to help find the line. To do this we calibrated our theodolite to Polaris and then corrected for the considerable change of magnetic declination since the surveys. Because of so many magnetic anomalies, however, we usually obtained only an approximate line by this method, so we surveyed trial lines, trying to conform to unchanged topography. This was highly problematic because many of the profiles were well over a thousand feet (300 m) long (one was over half mile, or 800 m) so that a small directional error could depart far from the original line. Moreover, all or parts of the line had to be cleared of brush and trees for a clear line of sight for our optical level. Most of my student volunteers had been through my field research class at UCLA so they knew how to survey. Thus, clearing the line usually fell to me while the students did the surveying, sometimes under Happ’s direction. To do this, I used a brush hook, which is a heavy but extremely sharp, scimitar-shaped blade mounted on an ax handle. The best I can say about it was that it kept me in good shape. The end result of all these problematics was that the resurvey of some profiles was exhausting and took more than a week. At first, we were squeamish about wading through stinking, anoxic swamps and the deep turbid streams of the lower valley, but soon paid it no mind. One student, however, a recent veteran of the Vietman war, in 1974 pronounced the Hill Country swamps worse than anything he had experienced in the jungles of Vietnam. On most days, we were soaking wet by 9 am, either from wading or sweating or both. We measured the severity of the day by the amount of water we drank. Luckily, no one ever fell out from the heat but several went to the emergency room for cuts and insect bites.

Note that Happ's design and our new survey data gave us stream and valley cross-sections for c. 1853 (based on borings to the old soil), c. 1938, and c. 1975. Other than resurveying the profiles, my contribution in this respect was to determine fluvial changes between those dates, thus establishing differential rates and then trying to connect the geomorphic changes to hydrologic changes driven by land use and climate. I did that to a large degree by floodplain excavations using archaeological techniques with datable landscape features such as bridges, roads, milldams, fences, and buildings, and by historic photos and descriptions. These receive much attention in this book.

My original intention was to report the work in journal articles, preferably Science, and I have done that now for over 35 years. However, while I was a visiting professor at the University of Chicago in 1978, Professor Karl Butzer heard a presentation gave on the work and invited me to write up the work as a possible book in the Geography Research Paper series sponsored by the Department of Geography there. I saw this as a heavy charge, and Istarted thinking in those terms, but it just seemed that there was always much more to do and to discover before writing a book. And indeed, I have found out more every year, and I believe that waiting truly has made the book immeasurably more complete and mature. Even as I write, I discover new insights and connections that had eluded me for 39 years.

xxxviii Preface

At one time, I perceived the book as a technical compendium of all the fluvial geomorphology work done in the region. But eventually, I saw that the story that must be told is the story of people interacting with their environment. Yes, I at least touch on most technical aspects of the work, but I have tried to hold such to a minimum and merely reference the more complete works.

My hope is to portray the totality of landscape, the myriad of both physical and human factors that have wrought changes on the Hill Country and that, in turn, have been affected by the complexity of both positive and negative feedbacks. With this, l'm put in mind of Barry Commoner' first "law" of ecology, that everything is connected to everything else (Commoner, 1971). Too often, we see things in a restricted, isolated, even myopic way, and my intention here is to pull as many of these diverse strands together as possible. In turn, I have borrowed from many fields of academic endeavor: hydrology, geomorphology, soil and water conservation, agricultural science, and rural sociology, to name a few. I hope that I have done them justice.