By Mark Petrie, Ph.D., and John Coluccy, Ph.D.

"Rained all of the after part of last night; rain continues this morning. I slept but very little last night for the noise kept up during the whole of the night by swans, geese, white and gray brant, ducks, etc., on small sand island close under the port side; they were immensely numerous, and their noise horrid." -William Clark, 1805

For many of us who have suffered a slow day in the duck blind, Bill comes off as a bit of a whiner. In fact, we're reasonably sure that most duck hunters would relish the conditions described by William Clark as he struggled to remain dry and catch a nap along the banks of the Columbia River. Other eyewitness accounts of early duck populations told of "bank to bank canvasback on the Susquehanna River" and "ducks so thick they blotted out the sun." Nice images to be sure, but they fall a little short when trying to relate historic duck populations to the modern hunting experience. For the purposes of this article, we decided to take a shot at reviving those long-lost waterfowl populations using a little science, conjecture, and some plain old feeling around in the dark.

Just how large were duck populations when Lewis and Clark left St. Louis to explore the uncharted West? We need to answer three questions before offering a guess. How large are duck populations today? How are these populations distributed among breeding habitats in Canada and the United States? And finally, how did European settlement reduce the number of birds breeding in these habitats and the young they produced?

Let's Count Ducks

Breeding waterfowl in North America are remarkably well surveyed. The traditional survey area, or TSA, includes the prairies, Canada's western boreal forest, and Alaska. Breeding surveys are also conducted in eastern Canada, the northeastern United States, and parts of the Midwest, as well as in California, Washington, Oregon, and Nevada. In 2012, these surveys counted 54 million breeding ducks (total ducks, not pairs), with 48.6 million in the TSA. Sixty percent of these birds were found on the Canadian and U.S. prairies. Another 30 percent of breeding ducks were counted in Alaska and Canada's western boreal forest. This broad region plays an important role in maintaining continental waterfowl populations by supporting relatively stable numbers of breeding birds, which make a fairly consistent contribution to the fall flight each year. Surveys outside the TSA claimed the rest.

To estimate how large presettlement duck populations might have been, we need to mentally rebuild the duck factory. Prairie breeding ducks use three types of wetlands. Temporary wetlands are the most common and typically hold water for only one to three weeks in spring. In dry years most temporary wetlands hold no water at all. Seasonal wetlands are the second most abundant breeding habitat and are wet anywhere from three weeks to three months each year. Semipermanent wetlands are the least common, though they may hold water for several years at a time.

There is a strong relationship between duck numbers and the abundance of temporary and seasonal wetlands. Every temporary or seasonal wetland lost on the prairies can mean one less pair of breeding ducks.

Prior to settlement, 95 percent of all prairie wetlands were temporary or seasonal. Knowing how many of these wetlands were lost provides a clue to the number of ducks the prairies once supported.

Iowa and Minnesota don't automatically come to mind when we think of the Prairie Pothole Region. However, both states were once a vital part of North America's duck factory. In fact, Iowa may have once hosted 2 million pairs of breeding ducks, about the same as today's densities in North Dakota. Unfortunately, the state has lost 95 percent of its temporary and seasonal wetlands, and now supports just 150,000 breeding pairs. The situation in Minnesota isn't much better. Eighty percent of this state's original temporary and seasonal wetlands are gone. Prior to settlement, western Minnesota may have supported 2 million pairs of breeding ducks compared to 375,000 today.

What about the rest of the U.S. prairies? North Dakota now supports 2 million pairs of breeding ducks in good years, though the state has lost half of its temporary and seasonal wetlands. South Dakota has lost a third of these habitats and can host 1.75 million breeding pairs under the right water conditions. North Dakota may have once supported 4 million pairs of breeding ducks, while South Dakota hosted another 2.7 million pairs. In total, the U.S. prairies may have historically supported 10.7 million duck pairs, compared to just 4.25 million in recent years.

Looking north, duck counts on the Canadian prairies have traditionally averaged twice that of the United States. Thus we might expect the Canadian duck factory to support roughly 8.5 million breeding pairs today. Unfortunately, wetland loss on the Canadian prairies over the past 25 years has far exceeded that of the United States, so it's unlikely that Canada still supports double the number of ducks. A better estimate is somewhere around 7 million breeding pairs. Canada's prairies have been stripped of 60 to 70 percent of their temporary and seasonal wetlands. With these habitats intact, the region may have hosted up to 20 million pairs of breeding ducks in wet years. Together, the U.S. and Canadian prairies may have supported 30 million pairs of breeding ducks compared to just 11 million today.

What about breeding duck populations outside the prairies? Canada's western boreal forest is now experiencing a variety of natural resources development. While this is cause for serious concern, breeding duck numbers in this region have managed to hold their own so far. Dabbling duck populations in Alaska have actually increased over the past 50 years and may now exceed presettlement levels. In eastern North America, long-term changes in duck populations are harder to discern. Beavers, which create much of our eastern breeding habitat, are at record numbers thanks to favorable forestry practices and a weak fur market. As a result, habitat loss on the prairies is almost certainly responsible for most of the decline in breeding duck numbers during the modern era.

Accounting for Recruitment

Although we can speculate about how many breeding ducks the prairies once supported, this doesn't tell us how big duck populations actually got. Most duck hunters are familiar with the term "fall flight," the size of the duck population going into the fall. Fall flight estimates are based on the size of the breeding population, habitat conditions, and the number of young produced that survive into fall. Since the 1950s, years with good breeding conditions on the prairies and elsewhere have produced fall flights of 100 million birds or more.

That brings us to the topic of "recruitment." Waterfowl managers define recruitment as the number of young added to the population by reproduction from adults in the spring. Recruitment depends on the percentage of hens that successfully hatch a nest-or hen success-and duckling survival. To answer the fall flight question, we need to consider possible changes in these variables.

Hen success is a function of nest success and renesting intensity. Nest success is simply the probability that a nest will hatch. Renesting intensity is strongly dependent on breeding habitat conditions and varies considerably among species, with mallards being the most likely renesters. Hens renest more often in wet years because more habitat is available to support late-hatched ducklings. We don't know if the relationship between renesting intensity and breeding habitat conditions has changed over time, but ducks still renest more frequently when it's wet and less frequently when it's dry. Consequently, we assume that long-term changes in hen success are probably due mostly to changes in nest success.

There is some evidence that nest success on the prairies declined by as much as 50 percent between the 1930s and the 1970s. But nesting ducks have never had it easy. Hen mallards marked with radio transmitters have renested as many as five times. This behavior evolved thousands of years before the prairies were cultivated and strongly suggests that nest predation was always high. To be conservative we'll assume that nest success before settlement was double that of today.

Changes in duckling survival are the last piece of the recruitment puzzle. Although there are no long-term studies of duckling survival, we do know that mink are a major duckling predator. Mink numbers crash during times of drought, though some mink are able to persist where permanent water is available. Irrigation projects and cattle operations have probably increased the amount of permanent water on the prairies, which may allow mink populations to quickly rebound when wet weather returns. As the prairies emerged from drought, presettlement duck populations may have enjoyed a longer holiday from mink predation than the birds do today.

Habitat loss on the prairies may also have reduced duckling survival. Hens frequently move their broods from wetland to wetland because of changes in food availability. With fewer wetlands on the landscape, hens and ducklings often have to move farther overland in search of food, which increases their exposure to upland predators. In fact, research indicates that duckling survival is higher on landscapes with more seasonal wetlands. As a result, the widespread loss of seasonal wetlands on the Canadian and U.S. prairies may have reduced duckling survival while also decreasing the number of breeding pairs.

Crunching the Numbers

Our understanding of wetland loss suggests that the duck factory could once support three times as many breeding ducks as it does today. To keep it simple, let's assume that the number of ducks breeding outside the prairies has not changed. We'll also assume that recruitment in these areas hasn't changed. In addition, we know that the largest fall flights occur in years when the prairies are universally wet, and that up to two-thirds of all ducks are counted on the prairies in wet years.

To estimate how big the fall flight might once have been, we can use modern-day duck populations and work backward. Let's assume that there are 50 million breeding ducks (total ducks, not pairs) distributed across the United States and Canada. It's been a very wet spring and 33 million of these ducks have settled on the prairies, while another 17 million can be found in Alaska, the western boreal forest of Canada, and areas outside the TSA. Given such conditions, we may have a fall flight of 100 million ducks, which we will use as our benchmark for modern day duck populations when the prairies are wet.

Now turn the clock back to 1805, when Clark was trying to get some sleep along the Columbia River. Assuming that the prairies have also been very wet, we have 100 million total breeding ducks (including both paired and unpaired birds) on the prairies. Duck numbers outside the prairies are unchanged at 17 million. This means that 200 years ago, we had 117 million breeding ducks compared to 50 million today (or 2.3 times as many ducks). At face value, this translates into a fall flight of about 230 million birds (2.3 x 100 million).

But we haven't yet accounted for changes in recruitment. Let's assume that recruitment on the prairies was twice what it is today, mostly due to higher nest success and duckling survival. Not only have we tripled the number of ducks on the prairies; they're twice as successful at producing young. We'll spare you the math, but this pushes our presettlement fall flight estimate to around 430 million birds, even if we assume that duck breeding populations and recruitment outside the prairies have not changed.

It's important to recognize that fall flights in excess of 400 million birds were probably rare. Our suggestion that the prairies could once support 30 million breeding pairs or 100 million total ducks assumes that most temporary and seasonal wetlands held water. In many years, large parts of the prairies were dry and the full carrying capacity of the duck factory would not be realized. Moreover, these incredibly wet years would have to be preceded by years of very good production if there were to be enough ducks to fill all those wetlands. Still, it's fun to imagine what it would be like to experience such a fall flight while sitting in a duck blind. We doubt that you'd complain about the noise.

Dr. Mark Petrie and Dr. John Coluccy are directors of conservation planning at DU's Vancouver (Washington) and Great Lakes/Atlantic offices, respectively.