- The Individual Cowbird Behavior Model (ICBM) -
The golden-cheeked warbler (Dendroica chrysoparia) and black-capped
vireo (Vireo atricapillus) are two endangered species that breed at Fort Hood,
Texas. Fort Hood personnel and other biologists have developed a variety of
programs to help ensure the viability of populations of these songbirds as directed
by the Federal Endangered Species Act. One such effort is the trapping of brown-headed
cowbirds (Molothrus ater), a brood parasite that lays it's eggs in the nests
of other bird species. Brood parasitism by cowbirds can dramatically reduce
the reproductive success of the host, and may contribute to population declines
observed for numerous passerine species.
The Individual Cowbird Behavior Model (ICBM) uses state-of-the-art computer
technology to simulate the spatial distribution and movements of cowbirds on
Fort Hood. By determining locations where foraging cowbirds aggregate, ICBM
can be applied by land managers to help guide decisions regarding the placement
of traps on the landscape. By reducing parastism by cowbirds, managers may better
protect these endangered species and other songbirds on Fort Hood. A related
model, the Fort Hood Avian Simulation Model -- FHASM, simulates the effects
of land-use practices and management activities on the population dynamics of
both the black-capped vireo and golden-cheeked warbler.
The cowbird gets its name from its association with cattle.
Insects that are exposed by grazing cattle are eaten by foraging cowbirds..
Rangeland in central Texas provides abundant feeding habitat for the cowbird.
A brood parasite, the brown-headed cowbird lays its eggs in the nest of another
bird, and then abondons the nest and allows the host to to care for its young.
Cowbird hatchlings typically grow more rapidly than the host's own hatchlings,
and may prevent the host from successfully fledging it's young. Cowbirds parasitize
many different host species, so can breed in a wide range of habitats. The daily
movements of cowbirds between breeding and feeding areas can be extensive (up
to 14 km), but cowbirds prefer to breed near ecotones formed by shrubland or
forests adjacent to grasslands.
Cowbird control efforts have been implemented on Fort Hood for
about a decade, and have proven very effective. For example, parasitism rates
for the black-capped vireo have declined from over 90% to less than 20% during
this time period. While successful, control efforts must continue indefinitely
to maintain beneficial effects for endangered species. Maintaining a rigorous
trapping program is a time-intensive project. Thus, the first objective in developing
ICBM was to determine if locations where cowbirds aggregate to feed cloud be
predicted. The second objective was to provide a mechanism by which the effectiveness
of alternative trapping stategies could be evaluated. Finally, an easy to use
WWW interface
was developed to allow the model to be run without a detailed understanding
of the underlying structure.
GRASS geographic information system (GIS) is used to capture spatially explicit habitat characteristics.
Vegetation is represented in GRASS maps, drawn from April 1996 Landsat TM imagery through neural network classification. The neural network was trained with 152 LCTA points from 1995. 91 showed grassland communities, 61 showed non-grassland, based on the PCC classification methodology (Tazik et al. 1992).
750mX750m (56.25 hectares) cells divided into 225 smaller cells of 50m resolution. Small cell data were summarized and attributed to their larger cells. Each large cell had a value for number of unique patches, representing the fragmentation of grasslands.The distance of all smaller cells from their nearest grassland neighbors was calculated and averaged for each large cell.
SWARM is an object-oriented simulation environment, capable of modeling individual, discrete entities and events in an ecosystem. Swarms are groups of entities, which may be nested hierarchically, all utilizing the same clock. "Agents" are entities that can generate events which affect themselves and other agents. ICBM uses this environment to simulate the interactions between "swarms" and "swarm objects":
The Observer Swarm consists of interface control panels, display and animation, and overall control of simulation (by the trap manager, for instance). The Model Swarm contains equations and ecological and behavioral information. Each object develops its own unique history according to the rules assigned to that object.
ICBM simulates the landscape of Fort Hood and all adjacent areas
within 7 km of the installation boundary. The simulation runs on a
time-step of one day, over a 100-day breeding season of the endangered
species during April, May, and June.
The Individual Cowbird Behavior Model (ICBM) can be thought of as two submodels, each using a variety of analytical tools, hardware and softwares located at the University of Illinois. The www Interface allows an off-campus user, such as the trap manager at Fort Hood, to access the model. A number of user-specified variables enable the trap manager to tailor the program to fit current needs and available resources.
Brood parasitism by brown-headed cowbirds (Molothrus ater) reduces the reproductive success of two endangered passerines at Fort Hood, TX. A control program has focused on trapping cowbirds at feeding locations, areas almost always associated with cattle. We developed an individual-based model to predict visitation rates for all feeding locations. The model captured the spatial arrangement of habitats, daily movements of cattle, and daily movements of cowbirds from their breeding site to a feeding location occupied by cattle. We simulated four types of cowbird movements: 1) random walk, 2) direct return to previous locations (i.e., memory), 3) return to previous locations with en route assessment (i.e., memory and current state of system), and 4) omniscience. We found straight-line distances between breeding and feeding locations produced by Type 3 closely matched independent telemetry results, and search distances approached those of omniscience. The map detailing visitation rates across the landscape may prove useful in determining future trap locations.
Abstract Not Available Yet
ICBM was used to simulate the location of breeding and feeding cowbirds . Using data from warbler and vireo breeding habits, this model identifies cowbirds likely to breed in endangered species' areas. It then considers how many cowbirds will be trapped in a given scenario, and what particular species, if any, might have otherwise been threatened by each individual cowbird.
Conclusions
This part of ICBM identifies feeding/breeding locations visited by cowbirds for a given landscape. Visitation is predicted based on natural landscape features from satellite maps, as well as user-specified variables such as initial cowbird and cattle densities. Visitation maps are input into the Trapping submodel which then determines the efficiency of various trapping scenarios.
I.
Habitat suitability for cowbird breeding is determined by (1) the amount of non-grassland habitat, and (2) the average distance from grassland edge. Suitability plus a user-specified saturation level, relative to total capacity of the area to support cowbird reproduction, will determine the number and placement of cowbirds.
Cowbird competition is likely to occur when their breeding habitat overlaps that of the threatened species. Cowbird habitat is therefore classified where they overlap Vireo and Warbler breeding areas.
|
Cowbird Breeding Suitability Map (Click Picture To Enlarge) |
|
Cowbird Breeding Locations Map (Click Picture To Enlarge) |
II.
Cowbird feeding areas depend primarily on the presence of cattle, but also distance from breeding habitat, their current location, and the rules of movement as described below.
Cowbird movement can be controlled by applying one of the following rules. When tested against actual field studies, the memory with drop-down rule compared most favorably. The user may specify one of the others when submitting a hypothetical situtation.
Next nearest Rule -- Cowbirds randomly move from breeding territory to nearest feeding area, then to next nearest feeding area, etc. until a cattle herd is located.
Memory Rule -- move to previous successful feeding areas in reverse chronological sequence until a cattle herd is located. If no cattle found at memory locations, follow next nearest rule
Memory with drop-down Rule -- Cowbirds act as above, but assess feeding areas en route to memory locations.
Omniscient Rule -- Cowbirds travel directly from breeding territory to nearest feeding area with a cattle herd.
Because cowbirds do not adhere to the fenced boundary of the military installatation, ICBM includes objects up to 7 kilometers from the installation boundary, Cattle do, however, so there can be no in-out exchange of cows. Off-base corrals are assumed to be distributed at the same density of those on-base, and serve the same number of cattle.
III.
Cattle grazing quality depends of these four landscape elements: (1)
good amount of grassland habitat; (2) some patchiness of grassland habitat;
(3) near a corral; and (4) near permanent water.
Initial cattle placement is determined on-post by the locations of 30 known corrals. Off-post, a hypothetical distribution has a density relative to on-post, at random locations stratified by habitat suitability. The number of herds, each consisting of 30 individuals, can be specified in the interface.
Cattle movement simulations capture documented behaviors. Bailey, Walker and Rittenhouse (1990) concluded that cattle display a "win-switch" strategy. That is, they leave a grazing area before forage quality declines to the level predicted by optimal foraging theory. In addition, cattle may utilize spatial memory to avoid recently grazed area. Therefore, a potential feeding location also reflects time since previous visit and distance from current location, in addition to primary grazing quality characteristics
Maps (Click Picture To Enlarge)
 |
Habitat Suitability for Cattle |
 |
Initial Cattle Herd Locations |
 |
Cumulative Cattle Herd Visits |
 |
Predicted Cowbird Visits |
Conclusions
Cowbird Trapping quantifies cowbird captures based on a set of user-defined management decisions. A satisfactory strategy should enable managers to trap more cowbirds with less effort.
The non-spatial attributes of the two types of traps is fixed, but a user must specify how many of each type are located on the landscape.
Spatial attributes of traps are input by the user or via maps previously generated either by the user or the Behavior submodel:
Timing of trap movement is set by the user interface. These options apply only to the Hybrid traps because the Megas are immobile. Due to resource limitations, the third and fourth are unlikely options, which yield no obvious improvement in capture success anyway.
Trap placement rules are chosen from the following options. Placement sites must be accessible by maintenance personnel.
After submitting the required parameters to the interface, and the model has been run, the following information can be summarized and reported:
Future research may include the following:
The following works were consulted in preparing the final technical report. We regret omissions, if any.
Bailey, D.W. "Daily Selection of Feeding Areas by Cattle in Homogenous and Heterogenous Environments", Applied Animal Behavioral Science, vol 45 (1995), pp. 183-200.
Bailey D.W., J.W. Walker, and R.L. Rittenhouse. "Sequential Analysis of Cattle Location: Day-to-Day Movement Patterns", Applied Animal Behavioral Science, vol 25 (1990), pp. 137- 148.
Bailey, D.W., J.E. Gross, E.A. Laca, L.R. Rittenhouse, M.B. Coughenour, D.M. Swift and P.L. Sims. "Mechanisms that Result in Large Herbivore Grazing Distribution Patterns", Journal of Range Management, vol 49 (1996), pp. 386-400.
Brittingham, M.C. and S.A. Temple. "Have Cowbirds Caused Forest Songbirds to Decline?" Bioscience, vol 33, no. 1 (1983), pp. 31-35.
Charnov E.L., "Optimal Foraging: the Marginal Value Theorem", Theoretical Population Biology, vol 9 (1976), pp. 129-136.
Conroy, M.J., Y. Cohen, F.C. James, Y.G. Matsinos, and B.A. Maurer. "Parameter Estimation, Reliability, and Model Improvement for Spatially Explicit Models of Animal Populations", Ecological Applications, vol 5, no.1 (1995), pp. 17-19.
Cook, C.W. "Factors Affecting Utilization of Mountain Slopes by Cattle", Journal of Range Management, vol 19 (1966), pp. 200-204.
Cook, T.L., J.A. Koloszar, and M.D. Goering. 1995 Annual Report: Behavior and Movement of the Brown-headed Cowbird (Molothrus ater) on Fort Hood, Texas, report by The Nature Conservancy of Texas (January 1995).
Coughenour, M.B. "Spatial Components of Plant-Herbivore Interactions in Pastoral, Ranching, and Native Ungulate Ecosystems", Journal of Range Management, vol 44, no. 6 (1991), pp. 530-542.
Dwyer, D.D. Activities and Grazing Preferences of Cows with Calves in Northern Osage County, Oklahoma, Oklahoma State University Agricultural Experiment Station Bulletin B-588 (1961).
Friedmann, H. Host Relations of the Parasitic Cowbirds, U.S. National Museum Bulletin Number 233 (1963).
Friedmann, H. and L.F. Kiff. "The Parasitic Cowbirds and Their Hosts", Proceedings of the Western Foundation of Vertebrate Zoology, vol 2, no.4 (1985), pp. 225-302.
Gates, J.E. and L.W. Gysel. "Avian Nest Dispersion and Fledgling Success in Field-Forest Ecotones", Ecology, vol 59, no. 5 (1978), pp. 871- 883.
Graber, J.W. "Distribution, Habitat Requirements, and Life History of the Black-capped Vireo (Vireo atricapillus)," Ecological Monographs, vol 31 (1961), pp. 313- 336.
Hayden, T.J., D.J. Tazik, R.H. Melton, and J.D. Cornelius, "Cowbird Control Program on Fort Hood, Texas: Lessons for Mitigation of Cowbird Parasitism on a Landscape Scale, " in T. Cook, S. Robinson, S. Rothstein, S. Sealy, and J. Smith, eds., The Ecology and Management of Cowbirds (The University of Texas Press, Austin, TX). In press.
Hodder, R.M. and W.A. Low, "Grazing Distribution of Free-Ranging Cattle at Three Sites in the Alice Springs District, Central Australia", Australian Rangeland Journal, vol 1 (1978), pp. 95-105.
Holechek, J.L., R.D. Pieper, and C.H. Herbel. Range Management: Principles and Practices, Prentice Hall, NJ (1989).
Low, W.A., M.L. Dudzinski and W.J. Muller, "The Influence of Forage and Climatic Conditions on Range Community Preference of Shorthorn Cattle in Central Australia", Journal of Applied Ecology, vol 18 (1981), pp. 11-26.
May, R.M. and S.K. Robinson, "Population Dynamics of Avian Brood Parasitism," American Naturalist, vol 126, no. 4 (1985), pp. 475- 494.
Mayfield, H., "The Brown-headed Cowbird, with Old and New Hosts", Living Bird, vol 4 (1965), pp. 13-28.
McDougald, N.K., W.E. Frost, and D.E. Jones. Use of Supplemental Feeding Locations to Manage Cattle Use on Riparian Areas of Hardwood Rangelands, United States Department of Agriculture (USDA) Forest Service General Technical Report PSW-110 (1989).
Pulliam, H.R., J.B. Dunning, Jr., and J. Liu, "Population Dynamics in Complex Landscapes: A Case Study," Ecological Applications, vol 2, no. 2 (1992), pp. 165- 177.
Robinson, S.K., J.P. Hoover, .R. Herkert, and R. Jack. "Cowbird Parasitism in a Fragmented Landscape: Effects of Tract Size, Habitat, and Abundance of Cowbirds and Hosts", in T. Cook, S. Robinson, S. Rothstein, S. Sealy, and J. Smith, eds., The Ecology and Management of Cowbirds (The University of Texas Press, Austin, TX). In press.
Robinson, S.K., S.I. Rothstein, M.C. Brittingham, L.J. Petit, and J.A. Grzybowski, "Ecology and Behavior of Cowbirds and their Impact on Host Populations" in T.E. Martin and D.M. Finch, eds., Ecology and Management of Neotropical Migratory Birds: A Synthesis and Review of Critical Issues, Oxford University Press, New York, (1995).
Robinson, S.K. and D.S. Wilcove, "Forest Fragmentation in the Temperate Zone and its Effects on Migratory Songbirds," Bird Conservation International, vol 4 (1994), pp. 233- 249.
Rothstein, S.I., J. Verner and E. Stevens. "Radio-Tracking Confirms a Unique Diurnal Pattern of Spatial Occurrence in the Parasitic Brown-headed Cowbird," Ecology, vol 65, no. 1 (1984), pp. 77-88.
Santa Fe Institute, World Wide Web site describing the modeling environment called Swarm. The WWW address for this information is: http://www.santafe.edu/projects/swarm.
Senft, R.L., L.R. Rittenhouse and R.G. Woodmansee. "The Use of Regression Models to Predict Spatial Patterns of Cattle Behavior," Journal of Range Management, vol 36, no. 5 (1983), pp. 553- 557.
Senft, R.L., M.B. Coughenour, D.W. Bailey, L.R. Rittenhouse, O.E. Sala, and D.M. Swift. "Large Herbivore Foraging and Ecological Hierarchies," Bioscience vol 37, no.11 (1987), pp. 789- 799.
Squires, V.R. "Behavior of Free-Ranging Livestock on Native Grasslands and Shrublands," Tropical Grasslands vol 16, no. 4 (1982), pp. 161- 170.
Stafford Smith, M.S. Modeling: Three Approaches to Predicting How Herbivore Impact is Distributed in Rangelands. New Mexico State University Agricultural Experiment Station Research Report 628 (1988).
Tazik, D.J., S.D. Warren, V.E. Diersing, R.B. Shaw, R.J. Brozka, C.A. Bagley, and W.R. Whitworth, U.S. Army Land Condition- Trend Analysis (LCTA) Plot Inventory Field Methods, U.S. Army Construction Engineering Research Laboratories (USACERL) Technical Report N-92/03/ADA247931, (USACERL, February 1992).
Turner, M.G., G.J. Arthaud, R.T. Engstrom, S.J. Hejl, J. Liu, S. Loeb, and K. McKelvey, "Usefulness of Spatially Explicit Population Models in Land Management," Ecological Applications, vol 5, no. 1 (1995), pp. 12-16.
Trame, A., S.J. Harper, J. Aycrigg, and J. Westervelt. The Fort Hood Avian Simulation Model: A Dynamic Model of Eco
*****(Need to find the rest)******
