Bird Upland GameLinking Management: Theory and Practice in South Texas Leonard A Brennan, Fidel Hernández, William P Kuvlesky, Jr., and Fred S Guthery CONTENTS Emerging Links between Theory and Management Harvest Management Empirical Example Research and Data Needs Culture and Perception Hunter–Covey Interface Models Static Models Dynamic Models Assumptions Role of Heat Continental Scale Local Scale Nest Predation Dynamics and Usable Space Predictions from Simulations Habitat: The Conceptual Link between Theory and Application Acknowledgments References 66 66 67 68 68 69 69 69 69 70 70 71 73 73 76 76 77 During the past 50 years, the South Texas landscape has provided a rich laboratory for developing a scientific basis for game bird management, primarily based on studies of northern bobwhites (Colinus virginianus) (Brennan 1999; Hernandez et al 2002) Beginning with landmark natural history investigations (Lehmann 1984) and continuing with a series of rigorous empirical studies and development of models by Guthery and his associates (Guthery 2002), the emphasis on quail research innovation in the United States has clearly shifted from the southeastern states to Texas and Oklahoma Few, if any, wildlife species other than perhaps white-tailed deer (Odocoileus virginianus) have received research attention comparable to the northern bobwhite For example, the recent Technology of Bobwhite Management (Guthery 2002) contains conceptual, quantitative, and theoretical models of all aspects of bobwhite life history: Energetics and energy-based carrying capacity, 65 © 2008 by Taylor & Francis Group, LLC Wildlife Science: Linking Ecological Theory and Management Applications 66 the physiological need for water, population ecology, viability and production, harvest theory, hunter– covey interface (HCI) theory, and theory of habitat management Linking such theoretical models with management practices remains a major challenge, but has the potential to address and clarify issues that have important implications for game bird management policy and regulation EMERGING LINKS BETWEEN THEORY AND MANAGEMENT Presently, there are at least four emerging examples of links between theoretical models and game bird management practices in South Texas: Various scenarios using harvest theory management models seem to be useful in clarifying the implications of an agency proposal to double the daily quail harvest in Texas Empirical data suggest that theoretical HCI models provide meaningful results that can help manage harvest pressure while optimizing hunting opportunities Role of heat in regulating bobwhite populations has been given a new prominence in quail management, both from a climactic perspective and from understanding how operative temperatures can make vast areas potentially lethal to bobwhites during significant periods of the day Simulation modeling to assess interactions among predation dynamics, usable space for nesting, along with precipitation and heat, indicate that while modest increase in bobwhite population might be gained from nest predator control, annual production can be devastated by only moderate decrease in nesting cover A unifying theme emerging from these four examples is that concepts related to habitat theory and usable space remain the cornerstone of successful bobwhite management applications The objective of this chapter is to briefly review and describe how these four examples form a linkage between theoretical ideas and applied management actions for sustaining wild quail populations and quail hunting in South Texas Given the space constraints of this chapter, it is impossible to address each topic in depth Rather, we focus on these four issues and the ways they serve as examples of how theory and management can be linked for wildlife conservation HARVEST MANAGEMENT The Texas Quail Conservation Initiative (TQCI) has been developed to stabilize, increase, and restore wild quail populations (Brennan et al 2005) The foundation on which the TQCI is based is habitat The initiative recognizes that a great deal of usable space for quail has been lost across the Texas landscape and that successful efforts to reverse the quail decline will mean delivery of programs that result in net gains of habitat for quail Despite the appropriate focus of the TQCI on habitat, administrators and politicians in Texas attempted to promote doubling the daily quail bag limit from 15 to 30 birds as an incentive to inspire private landowners to implement habitat management Rather than inspiring people to implement habitat management, the proposal to double the daily bag limit for private property owners who implement quail management backfired on its proponents It sparked a vociferous public debate about the impact of hunting, including bag limits, season length, and associated issues Although the intensity of opposition to the 30-bird daily bag limit caused the proposal to be withdrawn, it pointed to a number of issues pertaining to quail harvest in general and shifted the debate to a new direction Opponents of the 30-bird daily quail bag limit argued that there was no biological basis for such a proposition, which seems like an appealing argument until one considers that the present 15-bird per day limit in Texas also has no biological basis, as most other statewide daily bag limits for quail © 2008 by Taylor & Francis Group, LLC Upland Game Bird Management 67 (Williams et al 2004) Fifteen birds per day is simply a number that is deemed an intuitive, happy medium somewhere between zero (no hunting allowed) and infinity (no limit required) that provides opportunities to maximize recreational quail hunting opportunities in a manner assumed consistent with wise use and sustainability In Texas, the 30-bird bag limit debate, as it became known, veered into conceptual territory that began to address the question: What, exactly, were the biological bases for setting a daily quail bag limit, or even a season length? Although quail harvest management remains a contentious issue, recent syntheses of past work, along with refinement of associated additive harvest models, appear to have added a bit of clarity and new direction (Guthery 2002) to the debate In its simplest form, the additive model appears as Qa = Vo + So Ho , where Qa is the total mortality from start to end of the hunting period, Ho is the hunting mortality rate for a given hunting effort in a population not subject to natural mortality, Vo is the natural mortality rate in a population not subject to hunting mortality, and So is the survival rate in a population not subject to harvest Despite being imperfect, the additive model can be used as a basis for approximating harvest rates on specific areas Managers will need data on fall density, spring density, and fall-to-spring survival (an approximate guess based on the previous year or a range of historical data) to make meaningful harvest prescriptions that result in a predetermined spring density Some simple assumptions about percent summer population gain from breeding also need to be made EMPIRICAL EXAMPLE Using a modification of the additive model, harvest prescriptions for specific areas can be calculated based on H = (T − N)(1 − N), where H is the harvest rate, T is the total mortality over winter = (fall density − spring density goal)/(fall density), and Nis the nonhunting winter mortality Assuming a fall density of bird/acre (2.4 birds/ha) and a spring density goal of 0.5 bird/acre (or about 1.2 birds/ha) T = (1 − 0.5)/1 = 0.5 N = 0.4 Therefore, H = (0.5 − 0.4)/(1 − 0.4) = 0.17 or a harvest rate of 17% If fall density estimates showed 1000 birds present on a 1000-acre pasture, the recommended harvest for that given hunting season would be 170 quail Theoretically, it should not matter much when those 170 birds were harvested during the hunting season, although it would probably be preferable to so during the early to middle part of the season to avoid potentially additive effects of late-season hunting mortality In contrast, given the current 15 birds per day bag limit, and a 120-day quail season in Texas, it would take only 33 hunter-days at 15 birds/day to legally inflict local extinction on the quail population in the 1000-acre pasture Thus, implementing harvest management prescriptions, that is, harvesting a given number of birds during the season based on calculations such as those shown above, on specific areas indicates that not only would losses to hunting be potentially lower than under the present policy, but hunting losses could also be tailored to year to year variation in rainfall and associated habitat conditions that influence summer gain (Table 4.1) © 2008 by Taylor & Francis Group, LLC Wildlife Science: Linking Ecological Theory and Management Applications 68 TABLE 4.1 Quail Harvest Prescriptions on a Management Unit under Various Annual Environmental Conditions Recall from the text example the 17% harvest rate prescription to achieve a breeding population density of 0.5 bird/acre Now, consider the following three different potential production scenarios for year 2: Example 1: Good production Breeding season gain of 100% = +500 birds Start of next hunting season = 1000 birds H = (0.5 − 0.4)/(1 − 0.4) for a harvest rate of 17% or 170 birds out of 1000 Example 2: Excellent production Breeding season gain of 150% = +750 birds Start of next hunting season = 1200 birds Assuming a spring density management objective of 0.5 bird/acre, and a fall density of 1.2 birds/acre: T = (1.2 − 0.5)/(1.2) = 0.58, and N = 0.4 (as noted in text); therefore, H = (0.58 − 0.4)/(1 − 0.4) = 0.8/0.6 = 0.3 for a harvest rate of 30% or 360 birds out of 1200 Example 3: Low production Breeding season gain of 50% = +250 birds Start of next hunting season = 750 birds Assuming a spring density management objective of 0.5 bird/acre, and a fall density of 0.75 birds/acre: T = (0.75 − 0.5)/(0.75) = 0.33, and N = 0.4 (as noted in text); therefore, H = (0.33 − 0.4)/(1 − 0.4) = −0.07/0.6 = −0.12 or no harvest Note that under the three scenarios presented, both the total annual bag and harvest rate, covary as fall density changes in relation to variation in breeding season gain RESEARCH AND DATA NEEDS The data needed to implement prescriptions for quail harvest management are at once both simple and complex They are simple in the sense that specific harvest quotas to achieve a breeding population density are based on estimates of population density and fall-to-spring survival and some measure of summer gain in relation to environmental conditions They are complex in the dual sense that (1) many factors can influence and bias estimates of population density and fall-to-spring survival, and (2) most field managers find collecting such data challenging at best, and intimidating at worst Of course, there is also the potential for resentment among managers who would consider collecting such data unnecessary, because they are entrenched in status quo CULTURE AND PERCEPTION Application of harvest management prescriptions has not gained widespread acceptance as a means of regulating quail hunting First, it is a relatively new idea that is not well known by managers even though the basic ideas behind it are based on landmark quail research (Errington 1945; Roseberry and Klimstra 1984; Guthery 2002) Second, the incumbent data and associated work required for collecting such data are a significant obstacle for many managers Third, the theoretical lack of a daily bag limit gives lay people the impression of unlimited harvest opportunities when in fact annual harvest will usually be more conservative than what is currently allowed Some see a seasonal quota based on harvest management prescriptions as an opportunity for hunters to be legally sanctioned game hogs, when actually just the opposite is the case Despite the obstacles noted above, harvest management prescriptions will need to be considered as we move towards new definitions of quail management in the twenty-first century For example, © 2008 by Taylor & Francis Group, LLC Upland Game Bird Management 69 Williams et al (2004) presented a compelling case for such a change in perspective They argued that quail habitat management efforts need to be scaled up from individual efforts on single properties to coordinated efforts on regional cooperatives, and that quail hunting pressure needs to be scaled down from statewide bag limits that have no biological basis to localized harvest prescriptions based on meaningful and conservative seasonal quail hunting quotas designed to sustain populations HUNTER–COVEY INTERFACE MODELS Quail hunting is a complex behavioral process that until recently did not seem amenable to quantification However, application of static and dynamic approaches that identified data needed to describe the HCI have appeared during the past decade (Radomsky and Guthery 2000; Guthery 2002) and produce meaningful results and predictions when used with empirical data (Hardin et al 2005) The models on which HCI theory is based contain some of the same variables as those of operations research in warfare Fortunately, for hunters and pointing dogs, quail are quarry that not return fire However, hunting is hunting whether you are after a submarine or a bobwhite The potential impacts of “friendly fire” (e.g., politicians accidentally shooting attorneys while quail hunting) are not part of HCI theory STATIC MODELS Application of static models based HCI theory has the potential to complement harvest management prescriptions in that such models provide a mechanism to optimize hunting opportunities in a spatial context through time That is, by increasing or decreasing the rate at which the hunt takes place — and by extension the rate at which coveys are encountered and birds are killed — harvest, and harvest opportunities, can be modulated to potentially achieve a management outcome (Figure 4.1) However, HCI is only about a decade old, and has been tested only on one area in South Texas (Hardin et al 2005), although a less formalized application of spatial data has been used to quantify willow ptarmigan (Lagopus lagopus) hunting in Norway (Brøseth and Pederson 2000) HCI theory has yet to become assimilated into the mainstream of quail management, despite having potential as a tool for managing harvest pressure on intensively hunted areas The quantitative nature of HCI theory further contributes to maintaining its obscurity Perhaps such issues will change over time as wildlife managers become more comfortable with quantitative techniques or software becomes available that will put a user-friendly face on implementing HCI theory for management DYNAMIC MODELS In HCI theory, dynamic models are used to assess the role of how quails learn to avoid predators and thus influence the outcome of the hunt Any seasoned quail hunter appreciates that quail tend to “flush wild” or “run wild” or otherwise seemingly increase their ability to avoid being shot at as the quail hunting season progresses It has been noted that there are at least four different scenarios (Figure 4.2) that are related to the rate at which quail learn and the intensity at which hunting occurs (Radomsky and Guthery 2000; Guthery 2002) Empirical data indicate that bobwhites in South Texas most likely fit a “low hunting intensity–low learning rate” scenario (Hardin et al 2005) (Figure 4.3) ASSUMPTIONS One of the most fascinating aspects of HCI theory is that it produces meaningful predictions and results when tested with empirical data, despite failure to meet underlying assumptions The key assumptions upon which HCI theory is based (Table 4.2) are seldom met under field conditions for a variety of reasons For example, the assumption that coveys are randomly distributed across the landscape can be met during some years but not others (Figure 4.4) Additionally, the assumption that © 2008 by Taylor & Francis Group, LLC Wildlife Science: Linking Ecological Theory and Management Applications 70 (d) 45 40 35 30 25 20 15 10 w=96 w=82 w=69 w=55 w=42 7.44 9.14 10.85 12.55 Daily harvest Daily harvest (a) 60 50 40 30 20 10 14.26 w=96 w=82 w=69 w=55 w=42 5.28 7.50 Velocity (kph) (e) 40 35 30 25 20 15 10 w=96 w=82 w=69 w=55 w=42 9.14 10.85 12.55 Daily harvest Daily harvest (b) 7.44 14.26 w=96 w=82 w=69 w=55 w=42 5.28 w=96 w=82 w=69 w=55 w=42 10.85 12.55 14.26 Daily harvest Daily harvest (f) 35 30 25 20 15 10 Velocity (kph) 14.16 7.50 9.72 11.94 14.16 Velocity (kph) (c) 9.14 11.94 50 45 40 35 30 25 20 15 10 Velocity (kph) 7.44 9.72 Velocity (kph) 45 40 35 30 25 20 15 10 w=96 w=82 w=69 w=55 w=42 5.28 7.50 9.72 11.94 14.16 Velocity (kph) FIGURE 4.1 Static model output from HCI analyses of quail hunting data (From Hardin, J B., et al 2005 J Wildl Manage 69: 498 With permission.) hunting patterns and areas covered during a hunt are not redundant is easily violated (Figure 4.5) If nonredundancy is more or less constant, this would not affect correlations between empirical and theoretical kill rates If hunts are applied randomly to space, or some semblance thereof, HCI ought to work, as it seems to work in line transect sampling Despite the difficulty in meeting assumptions, it seems that HCI theory is “robust” to violation of the assumptions upon which it is based This is a seemingly remarkable phenomenon that needs to be tested with sensitivity analyses and additional simulations ROLE OF HEAT Poultry scientists have long-recognized the deleterious effects of excess heat on laying chickens such as declines in feed intake, egg production, eggshell thickness, and quality of yolk and albumin inside the egg (Card and Nesheim 1972; North 1972) This was also noted in bobwhites more than three decades ago (Case and Robel 1974) Thus, it is curious that, until recently, quail researchers ignored heat as a factor that limits quail production (Guthery et al 2000; Guthery 2002) In this section, we explore two scale-dependent issues that relate to how heat affects quail populations, the first on a continental scale in the context of climate change and global warming, and the second related to how heat impacts usable habitat space on a local, fine-grained scale CONTINENTAL SCALE By disrupting breeding and habitat use on a local scale, global warming could potentially contribute to quail population declines on a continental scale Although largely overlooked by most quail © 2008 by Taylor & Francis Group, LLC Upland Game Bird Management 71 500 500 Low–low 400 300 Total 200 300 200 Naïve 100 Coveys Conveys 400 100 Experienced 0 15 30 45 60 75 90 Day 15 30 45 60 75 90 Day 500 500 High–high High–low 400 Coveys 400 Conveys Low–high 300 300 200 200 100 100 15 30 45 60 75 90 Day 0 15 30 45 60 75 90 Day FIGURE 4.2 Dynamic model predictions of HCI theory (From Radomsky, A A., and F S Guthery 2000 In National Quail Symposium Proceedings Brennan, L A., et al (eds), vol Tallahassee, FL: Tall Timbers Research Station, p 78 With permission.) researchers, global warming has been hypothesized to be potentially responsible for long-term quail declines (Guthery et al 2000) Habitat loss and fragmentation are most frequently invoked as being the cause of quail population declines, but some have noted (Guthery et al 2000) that there are other places where declines have occurred where there is no apparent loss of habitat Intellectually, longterm warming trends could potentially be responsible for quail population declines, at least from the perspective of climate change as one in a series of multiple working hypotheses (Chamberlain 1890) LOCAL SCALE Few people seem to appreciate that quail are sensitive to operative temperatures >38.7◦ C, that such temperatures occur regularly at southern latitudes, and that operative temperatures greater than this threshold can be lethal to quail (Figure 4.6) In the south, operative temperatures >45◦ C regularly occur at ground level during summer (Figure 4.7) At such temperatures, the heat intake exceeds heat loss in the quail body and death can occur in