(Editor’s Note: This is the final installment of our four-part series examining the broad scope of problems facing ducks, duck hunters and waterfowl managers. The author, Mickey Heitmeyer, one of the world’s leading authorities on mallards, offers his conclusions and recommendations. Because of its semi-technical nature, it is recommended that you first print the article, then read from the printed page.)

* * *

“Survival, particularly in the first year of life, is the major source of

variance in reproductive success. Yet the major source of mortality is

the North American hunter, an instrument of selection for only the past

hundred generations of snow geese. He (or she) may be a relatively

undiscriminating selective agent in terms of phenotypic characteristics

that were important during most of the bird’s evolutionary history”

(Cooke and Rockwell 1988)

Waterfowl are among the most successful groups of birds in the world. The diversity of species, their mobility and awe-inspiring migrations, the amazing architecture of morphology and physiology, and an uncanny timing of annual cycle events that coincide with periods of increased resource availability and reduced mortality agents across the North American continent are mind-boggling. But the gnawing question remains: Have we so altered their environment; the timing, sequence, and successful completion of annual events, and their exposure to mortality agents that by default we have jeopardized their resiliency and capability to sustain populations? Like it or not, one of the factors that contributes to this dilemma is sport hunting. We, as hunters, must demand that the best scientific information be used to design, implement, and regulate waterfowl hunting seasons. Further, we must evaluate both science and regulations to see if our old friend (or foe) heterogeneity and the “super hen” are included.

Adaptive Harvest Management

In 1995, the U.S. Fish and Wildlife Service adopted the concept of Adaptive Harvest Management (AHM) to regulate duck harvests in the United States. Most waterfowl biologists initially embraced this “new” approach because it explicitly admitted that the consequences of hunting regulations could not be predicted with certainty, and it provided a framework for learning and making decisions in the face of uncertainty. It also incorporated the virtue of “adapting” or changing decisions as new information is learned.

AHM was developed with the intent of dealing with uncertainties by: 1) offering a limited number of regulatory options (liberal, moderate, restrictive), 2) incorporating statistical models that would test hypotheses about harvest and environmental factors, 3) using an objective function that seeks to optimize the balance between maximizing harvest and maintaining a “viable” population over the long term, and 4) monitoring changes in population size to determine if regulatory changes are needed. AHM specifically sought to answer questions about whether harvest was “compensatory” or “additive” and whether density of birds had any effect on annual recruitment and survival.

All well and good, except for a few fundamental issues. First, the models did not initially define a “viable” population. Second, AHM regulatory options were based only on population changes of mid-continent mallards. Third, the only environmental variable used in models was the number of May “ponds” counted in the prairie-pothole breeding region of Canada. Further it assumed that duck production remains highly correlated with prairie pond numbers as historically was the case (at least since breeding surveys began in 1955) despite the fact that prairie ponds at the end of the 20^th century have become highly degraded and are within modified landscapes where ecosystem processes have been highly compromised. Fourth, only four hypotheses (the various combinations of compensatory/additive and density dependence/independence) were deemed important and testable. Fifth, the objective function, which sought maximum harvest, was assumed to be the desire of the American duck hunter. And, finally, as you might suspect, analyses were based on “average” population parameters and did not incorporate heterogeneity (differences) within or among species.

Since initial implementation of AHM, some attempts have been made to address these problems. First, population viability now is assumed to be the species goals set by the North American Waterfowl Management Plan (NAWMP). The question is: Do these NAWMP goals actually represent true population viability in today’s modified landscapes? NAWMP goals are specifically stated to be numbers present under “average” long-term environmental conditions. Breeding mallard numbers have been at or above NAWMP goals in some recent years, but they certainly were not achieved under average environmental conditions (the late 1990s had nearly unprecedented “wet” conditions and high May pond numbers on U.S. breeding grounds). Second, some attempts have been made to include “other” mallard stocks such as those breeding in eastern areas, but, to date, inclusion of other stocks and species is limited. Third, biologists have repeatedly argued for inclusion of more comprehensive, and contemporary, measures of habitat conditions throughout breeding, migration, and wintering areas, rather than the singular Canadian “May pond” numbers used in predictive equations. But to date inclusion of other habitat measures and geographical regions has not been incorporated. Fourth, in a related issue, no changes in hypothesis testing has occurred, despite increasing evidence of other important population phenomena. For example, effects of density dependence/independence are not functions solely related to breeding habitat conditions.

Lastly, heterogeneity in population parameters (such as identified in life cycle models in Part III of this series) has not been recognized. Unfortunately, AHM is still too “simple.”

For many obvious reasons, simplifying harvest regulations is important both for the science of learning and for compliance by hunters. However, when simplicity trumps extensively documented knowledge of the multiple biological factors affecting population dynamics, it is unacceptable. The potential risk of making long-term mistakes with hunting regulations simply is too great. Mathematicians call these errors “Type I” and “Type II.” A Type I error rejects the null hypothesis (e.g. hunting is additive) when it is in fact true. Conversely, Type II errors accept the null hypotheses (e.g., density dependence) when it is in fact false. Either error type is dangerous for waterfowl management, and the failure to incorporate heterogeneity in population analyses increases the probability of both.

Additive vs. Compensatory Mortality

The question of whether duck hunting harvest is “compensatory” or “additive” (or both) to natural mortality from non-hunting causes (see Part III of this series) has been widely debated among waterfowl managers for nearly 30 years, yet we still are uncertain about it. Simply put, and as defined by Williams 2002 et al. (see reference list), the “compensatory” mortality hypothesis states that hunting harvest reduces density of ducks and thereby reduces natural mortality. In essence, if compensation occurs, harvest (at least up to some point) does not cause higher total annual mortality of populations. The alternative “additive” mortality hypothesis states that annual mortality from hunting adds to mortality from non-hunting causes and creates higher total mortality in populations. Based on the preceding discussions of heterogeneity, I submit that both additivity and compensation exist. In fairness, the first authors writing on this subject also believed both “compensation” and “additivity” existed. They hypothesized that harvest was compensatory up to some unknown “threshold” point, and then harvest exceeding that threshold became additive. This undoubtedly is true. At some point, if we kill too many ducks the population is going to decline no matter what (if any) compensation exists.

But the compensatory/additive concept is more complicated than a “size of the pile” threshold. It includes a “who is in the pile” factor. The preceding parts of this series have identified the substantial individual variation in populations, both in recruitment and survival. If a female that produces no or few recruits (for many reasons, including age, body condition, genetics, etc.) is killed, her mortality may be relatively unimportant to subsequent population changes and thus likely is compensatory (at least in part). In contrast, if a “super hen” is killed, it invariably will be an additive death (regardless of the number of other ducks killed) unless the bird is at the end of its lifespan and reproductive capability. Many other factors are involved also. For example, a bird that has consumed lead shot, one that is diseased, or one that has some morphological and physiological defect probably will die, and not produce recruits, so killing this bird probably is compensatory. Further, to a degree, harvesting a juvenile may be more compensatory than killing an adult. However, additivity is exacerbated if too many juveniles are killed (the source of the next generation) and if the juvenile is a progeny of the “super hen” genetic stock.

Obviously, much more information is needed before we completely accept, or dismiss, either compensation or additivity. But, heterogeneity of different species and population segments must be acknowledged and considered in analyses.

Early and Late Season Frameworks

The problems of early and late hunting seasons have already been discussed in this series. Much of the debate on this issue involves mallards because of their abundance and hunter preference for them. Early seasons kill a higher proportion of juveniles and late-nesting females. Late seasons kill a higher proportion of adults and paired birds. Both consequences can be detrimental to the reproductive potential of mallard populations depending on the extent that early and late seasons kill more successful “super hens.” Extended seasons likely are detrimental to other species also, especially early seasons on breeding areas where the multiple species collections of “brown” juvenile birds and females makes hunting more indiscriminate among species and sexes.

The “tit-for-tat” politics that caused the current extended beginning and ending dates for duck season frameworks are disgusting. Initially spawned by insolence from certain Mississippi politicians about the need for late January seasons, the pendulum swung to reciprocation demands for early seasons in northern states (and Canada). If, by default, either early or late seasons increases the mortality of “super hens,” then changes must be made.

I know this conclusion will not make hunters in either the north or south happy. It speaks of inequality in opportunity by geography. But who said the duck pie must be divided equally? God certainly didn’t. Ducks are not distributed equally or randomly, nor should hunting be so. Further, in my opinion, the continuation of the contemporary lengthy season frameworks (both early and late) runs the high risk of changing waterfowl distribution in the future (e.g., next section on spinning-wing decoys).

Obviously, we need increased support for wetland and waterfowl conservation throughout North America from all segments (and geographies) of society. We also need stronger conservation leadership. Politicians and administrators fear the controversy of inequality lest a minority group cry oppression and seek to boot them out of office. But the reverse is true also. If the majority need is not met, then all suffer. Who has, or is willing to accept, the fiduciary responsibility for waterfowl in North America? For years duck hunters throughout North America have supported habitat programs in northern breeding areas (sometimes at the expense of their own local habitat programs) because they understood the need to produce ducks that would eventually migrate to areas where they could be harvested. Also, some northern hunters have contributed to habitat programs on wintering and migration areas because they understood that if nonbreeding habitat wasn’t present there would be no ducks to return to breeding areas. Some geographical priority must be given to habitat conservation programs in both breeding and nonbreeding areas. Likewise, some geographical priority (that requires different regulations) must be included in harvest regulations. If that means changing the framework system, so be it.

Spinning-wing decoys

Another major issue that must be addressed involves the effects of spinning-wing decoys on duck populations. Many other hunting issues deserve discussion also [e.g., refuges, zones and splits, species-specific regulations (or the lack thereof), etc.], but space is limited and the reader may be tired of this by now. Perhaps later.

The use of electronic duck decoys, especially spinning-wing decoys (SWDs), became widespread across North America during the late 1990s. Today, more than 70% of duck hunters use them, at least occasionally, across the U.S. (percentages are higher in some states and locations). Why? Because they work. The greatest test tube in the world is a duck hunter. Duck hunters are relentless in the pursuit of things that will help them be more successful. The immediate and continued acceptance of, and massive expenditures for, SWDs by duck hunters have shown that their use increases the vulnerability of ducks to being killed. No further statistics or scientific studies are needed to believe this. If you don’t accept this, then you haven’t spent much time in a duck blind.

The scientific community believes this also. A California study showed that in paired tests (one group using, and one group not using SWDs) harvest rates of hunters using SWDs was six times higher in early season, four times higher in mid-season, and two times higher in late season. A Manitoba study showed mallard harvest was five times higher for hunters using SWDs in marshes and 24 times higher in agricultural fields. A Missouri study found daily success for hunters using SWDs increased 13-19 percent, and in paired tests, harvest was 1.5 times higher for hunters using SWDs early and late season and three times higher in mid season. A Nebraska study using paired tests showed that hunters using SWDs harvested twice as many ducks in marshes and the mallard harvest was three to four times higher in late season. A Minnesota study found mallard flocks were 2.9 times more likely to respond to hunters, and sizes of flocks were 1.25 times bigger when SWDs were used. Mallard kill/hour/hunter was 4.7 times higher when SWDs were used. Juveniles were especially susceptible to SWDs. It was estimated that if 47 percent and 79 percent of Minnesota hunters (actual percentages now are above that) had used SWDs in 2000 and 2002, the mallard harvest in the state would have doubled.

Interestingly (and I don’t think coincidentally), the distribution of mallard harvest has changed in the last six years since use of SWDs has become widespread. For example, in the combined Central and Mississippi Flyways, distribution of mallard harvest was 22.6 percent, 22.4 percent, and 55.1 percent in northern, mid-latitude, and southern states, respectively in the 1990s (pre-SWD). Since 2000, distribution of mallard harvest is closer to 30 percent, 25 percent, and 45 percent in northern, mid-latitude, and southern states, respectively. Additionally, the age-ratios (young per adult) of mallards harvested in these states changed. Pre-SWDs age ratios of mallards were 2.5:1, 2:1, and

1:1 in northern, mid-latitude, and southern states, but since 2000 the age ratios have averaged nearly 3:1, 1.5:1, and 0.5:1 in these same areas. The bottom line is that more birds, and more young birds, are being shot in northern states than ever before. Another interesting coincidence (maybe cause-and-effect?) is that hunter numbers and their reported number of days hunted per year have increased dramatically in many northern states (e.g., North Dakota) in the Central and Mississippi Flyways since SWDs emerged.

Certain reviews of mallard harvest rates in the United States in the last two decades have indicated that total harvest rates have not changed since SWDs have been in existence. There seems to be a paradox here. How can numerous studies, and the hunter “test-tube” of high use of SWDs (greater than 70 percent) and increased number of hunters and days spent afield, all indicate major increases in vulnerability and total harvest of ducks, yet total harvest estimated from banding analyses and surveys of hunters remain the same? Is there some problem with the adequacy of banding analyses, sampling protocols, and surveys? I don’t know. Some suspect a problem in estimating band recovery rates that occurred after the inscription on bands was changed to include a 1-800 phone number in the 1990s. Reward bands have been placed on mallards and wood ducks over the past few years to identify the “new” reporting rates of bands associated with the 1-800 number and these data should provide more confidence in the precision and stability of recovery rates (that are used to calculate harvest and annual survival rates). Biologists typically assume that biases in sampling, banding, and analyses remain consistent among years/areas and therefore, even if the precision of estimates are “off,” that the trends in data remain useful. This may be true for some analyses, but if substantial heterogeneity exists among species, years, methods, etc., then the biases may be significant. This may be especially true for an estimate such as harvest rate, because even a slight deviation from “true” mortality rates of “super hens” can cause large variance in modeling population estimates (see Part III of this series).

Even if total harvest rate of mallards in the U.S. (or within flyways) has not increased (unlikely from my point of view), the changed vulnerability of birds, and the changed distribution of harvest across latitudes after SWDs came on the hunting scene can still have demonstrable negative effects on populations. I offer an example that this time includes “heterogeneity”, instead of “size of pile” statistics. Consider a somewhat hypothetical (much data are real, however) harvest scenario for midcontinent mallards pre- and post-SWDs (Table 1in Addendum). In this population, we have 10,000 female mallards (5,000 adults and 5,000 juveniles) present at the beginning of duck season. One-thousand subsequently are shot by hunters (a 10 percent harvest rate). Distribution of harvest pre-SWDs is 22.6 percent in northern states, 22.4 percent in mid-latitude states, and 55.1 percent in southern states. This leads to 585 juveniles and 416 adults being killed (a population average of 1.4:1 age ratio). Post-SWDs, we assume no change in total harvest rate (still 1000 total mallards shot), but we acknowledge the changed distribution of harvest and increased vulnerability of juveniles in various latitudes (Table 1). Now, post-SWD, 540 juveniles and 459 adults are shot. Consequently, even though total number of mallards shot (harvest rate) was the same pre- and post-SWD, the harvest of adults increased 10 percent. Simple math tells us more adults were shot in southern states post-SWDs because the population that migrated to the south contained more adults (i.e., a high proportion of young were shot in northern states and Canada).

A 10 percent increase in harvest of adult females may not seem significant to some, but if one follows the simple life model examples used in Part III of this series, this 10 percent increase probably includes many super hens (adult pairs that have lived to reach southern wintering areas) and their death could substantially decrease subsequent recruitment. Also, by simple math, more dead adults in the south (caused by higher harvest of young up north) could quickly decrease traditions of use by birds migrating to more southern areas and compound changes in both future bird and harvest distribution. Messing with evolutionary patterns of waterfowl migration and distribution, particularly in today’s highly modified landscapes, is scary, often latent, stuff.

The problem with ignoring heterogeneity in waterfowl management decisions raises its ugly head again in this SWD example. In addition to making fundamental mistakes about population consequences, it also demonstrates the potential for many unintended consequences to occur. A flow-chart (Figure 1) shows the sequence of effects of using SWDs, some social, some biological, and all interrelated.

Furthermore, use of SWDs raises many issues related to the ethics of fair chase. Many of these ethical issues are similar to those dealt with years ago (and with much less biological data involved) when decisions were made to ban live decoys, baiting, electronic callers, and night hunting. I won’t discuss this ethical dilemma here, but leave it to say, that in my opinion, this time we have both ethical and biological concerns with continued use of SWDs.

Conclusion

The objective of this series was to identify the importance of understanding, and incorporating, heterogeneity into future waterfowl population and habitat management programs. Specifically, I believe we must seek to better understand and protect the “super hen” component of all species and populations.

Heterogeneity is not new to waterfowl biologists and it has been incorporated into some programs (e.g., different harvest regulations for mallard stocks and sexes). I respect the ongoing efforts to improve management capability. Increasing the complexity of analyses and understanding is difficult, but I think possible. Having worked in the waterfowl business for nearly 30 years, I appreciate some of the difficulties in doing so, especially in today’s social and politically-oriented environment. Unfortunately: 1) waterfowl and wetland ecology curricula and research at universities are declining, highly fragmented, and less applied; 2) state, provincial, and federal resource agency biologists have less time, money, and motivation to work on waterfowl; 3) agendas of most non governmental groups mainly are focused on raising money and seldom invest in science or improved understanding and application of complex or controversial issues; 4) the “farm-system” of developing future waterfowl biologists that are passionate and hunter-oriented is sparse; and 5) waterfowl hunters are increasingly disengaged from regulation and management programs.

Collectively, the waterfowl community of hunters, biologists, administrators, and policy-makers can help correct some of the declining emphasis on waterfowl/wetland programs and improve current waterfowl management by including heterogeneity concerns. Many needs have been identified in this series; and at the very least, I believe we should:

• Continue to expand long-term studies of key waterfowl species that attempt to understand lifetime reproductive success (LRS) and identify heterogeneity in population segments.

   

• Clarify reasonable indicators (e.g., early paired adults, better body condition, advanced plumage development, etc.) of “super hens” among species and populations.

   

• Develop monitoring programs to determine seasonal, annual, and geographical variances in “super hen” indicators.

   

• Develop population dynamics models that include multiple population segments based on known or potential varying reproductive and survival probabilities, including analyses of LRS for key species and populations.

   

• Support habitat and harvest management programs for multiple species and populations and not assume all duck species respond to programs like mid-continent mallards.

   

• Conduct specific analyses of harvest age-ratios and other “super hen” indicators related to temporal, spatial, and harvest regulations variables (such as SWD effects).

   

• Update AHM models to reduce problems outlined earlier in this Part IV AHM section.

   

• Improve education of, and communication between, hunters, biologists, and administrators on waterfowl management issues.

   

Obviously, hunters play a critical role in the future of waterfowl management. If changes in hunting regulations are warranted, and can reduce excessive mortality of “super hens,” then they offer a quick, less expensive, and probably a hunter-acceptable option compared to the longer term, expensive, and politically challenging work of changing habitat and predator populations over widespread geographical areas throughout breeding, migration, and wintering areas. Certainly, we must continue to invest for the long term in habitat programs, but we should not dismiss the contributions (short- and long-term) related to harvest regulations. At the very least, if one believes that heterogeneity is an important consideration in waterfowl management, then shortening season length frameworks, reducing hen bag limits and banning SWDs are “no brainers.”

We owe it to our wonderful waterfowl to use every tool and every piece of information possible to make better, more informed, decisions. May more “super hens” fly each year.

Addendum

Table 1. Effects of a changed distribution of harvest (total and young, adult) caused by use of spinning-wing decoys (SWD) on the percentage of adults killed in a hypothetical mallard population.

Selected References

Cooke, F., and R.F. Rockwell. 1988. Reproductive success in a lesser snow goose population. Pages 237-250 in T.H. Clutton-Brock, editor, Reproductive success: studies in individual variation in contrasting breeding systems. University of Chicago Press, Chicago.

Fredrickson, L.H., and R.D. Drobney. 1979. Habitat utilization by postbreeding waterfowl. Pages 119-130 in T.A. Bookhout, editor. Waterfowl and wetlands - an integrated review. Proceedings of a symposium of the North Central Section of the Wildlife Society. LaCrosse Printing Company, LaCrosse, WI.

Humburg, D.D., T.W. Aldrich, S. Baker, G. Costanzo, J.H. Gammonley, M.A. Johnson, B. Swift, and d. Yparraguirre. 2000. Adaptive harvest management: has anything really changed? Transactions of the North American Wildlife and Natural Resources Conference. 65:7893

Johnson, F.A., J.A. Royle, and M.C. Runge. 2002. Framework-date extensions and the adaptive management of mallard harvests. Fish and Wildlife Service, U.S. Department of the Interior, Washington, D.C. 10pp.

Lercel, B.A., R.M. Kaminski, and R.R. Cox, Jr. 1999. Mate loss in winter affects reproduction of mallards. Journal of Wildlife Management 63:621-629.

Raveling, D.G. 1978. Dynamics of distribution of Canada geese in winter. Transactions of the North American Wildlife and Natural Resources Conference. 43:206-225.

Szymanski, M.L. and A.D. Afton. 2005. Effects of spinning-wing decoys on flock behavior and hunting vulnerability of mallards in Minnesota. Wildlife Society Bulletin 33: 993-1001.

U.S. Fish and Wildlife Service. 2005. Review of electronic motorized decoys for taking migratory game birds. Fish and Wildlife Service, U.S. Department of the Interior, Washington, D.C.

Williams, B.K., F.A. Johnson, and K. Wilkins. 1996. Uncertainty and the adaptive management of waterfowl harvests. Journal of Wildlife Management 60:223-232.

Williams, B.K., J.D. Nichols, and M.J. Conroy. 2002. Analysis and management of animal populations. Academic Press, New York. 817pp.

The author thanks many individuals from diverse agency, scientific, hunting, and nongovernmental groups for perspectives, debate, and review of this series. Robert Cox, Leigh Fredrickson, Dale Humburg, Karen Kyle, Jim Phillips, Gary Pogue, and John Stanard provided especially helpful editorial and context assistance. The genesis for the author’s interest in heterogeneity originated in many discussions in hotel rooms and flyway technical section meeting rooms, often hosted by the late Dennis Raveling.

Dr. Mickey Heitmeyer has investigated waterfowl and wetlands for nearly 30 years, and is widely recognized as one of the world’s leading experts on the biology of mallards. His professional background includes: technician and wetland manager for the Missouri Department of Conservation, research biologist for the University of California-Davis, Director of Research and Outreach for the California Waterfowl Association; Group Manager of Conservation Programs and Director of the Institute for Wetland and Waterfowl Research for Ducks Unlimited, Inc. and Ducks Unlimited, Canada. He is currently a Research Associate with the Gaylord Memorial Laboratory at The University of Missouri. He is an avid duck hunter and is owner of Dark Cypress Farms and co-owner of Cato Slough Hunting Club LLC ( www.catoslough.com ). Both are conservation properties in southeast Missouri that offer wetland management services and limited high-quality sporting opportunities. He and several other wetland scientists and managers have recently formed a nonprofit entity," Wetland Management and Education Services" (501 C3) and a consulting group, "Greenbrier Wetland Services" to provide wetland and waterfowl education, management, restoration, planning, and evaluation/monitoring services that have declined, or can not be provided in traditional agency, academic, and nongovernmental organizations. He can be contacted at mheitmeyer@earthlink.ne t