USGS Science⌕ Search

Geology topics

B. Kenneth Williams

Publications and source records attributed to B. Kenneth Williams.

33 records · Page 2Linked to original sources
USGS1989

Behaviour of captive canvasbacks Aythya valisineria fed different diets during winter

Time activity budget studies were conducted on captive Canvasbacks manintained on ad libitum diets with varying levels of protein and energy during the winter of 1978-79 and 1979-80. No differences could be detected in the behaviour of the ducks as a result of the diets they received. Differences due to season and sex were observed for some behaviours. Activity decreased (P<0.05) during the winter apparently as a mechanism to conserve energy. This decrease occurred in mid-winter irrespective of diet quality and appeared to be an endogenous component of the Canvasbacks' annual cycle. This behaviour pattern has been observed in the wild and seems to persist in captive Canvasbacks.

Wildfowl↗
USGS1989

Procedures for the Analysis of Band-recovery Data and User Instructions for Program MULT

We briefly review methods for inference from band-recovery data and introduce a new, flexible procedure (MULT) for analysis of data from bird-banding studies. We compare our computing method to program SURIV and discuss the relative advanatages of each. We present several basic model structures that can be analyzed using program MULT and for each model structure describe estimation and hypothesis testing and give a data example. We provide a complete description of porgram MULT, which is IBM-PC compatible and may be run as either an interactive or a batch-mode program.

Resource Publication↗
USGS1988

Sex specificity of behavioral dominance and fasting endurance in wintering canvasbacks: Experimental results

Hand-reared canvasbacks (Aythya valisineria) of varying sex ratios were maintained in pens during winter 1980-81 (3M-3F, 6M-0F, 0M-6F) and winter 1981-82 (4M-2F, 2M-4F) and fed two diets (control and stress). They were observed during feeding trials to determine intrasexual and intersexual aggressive activity. There was little evidence that either diet or sex ratio affected the total number of aggressive encounters. Females fed both control and stress diets were more aggressive and spent more time in the small feeding areas than males in pens with 3M-3F, 4M-2F, and 2M-4F sex ratios. Stressed ducks tended to weigh less than controls throughout the study. Females in the 3M-3F and 4M-2F pens weighed less than those in the 0M-6F and 2M-4F pens, respectively. However, relative weight changes throughout the winter were similar for males and females. Thus, results of these experiments do not lead to conclusive rejection of either the behavioral dominance hypothesis or the fasting endurance hypothesis.

Book chapter↗
USGS1987

Frequency sampling in microhistological studies: An alternative model

Frequency sampling in microhistological studies is discussed in terms of sampling procedures, statistical properties, and biological inferences. Two sampling approaches are described and contrasted, and some standard methods for improving the stability of density estimators are discussed. Possible sources of difficulty are highlighted in terms of sampling design and statistical analysis. An alternative model is proposed that accounts for 2-stage sampling, and yields reasonable, we!!-behaved estimates of relative densities.

Journal of Range Management↗
USGS1987

Relationships between nesting populations of wading birds and habitat features along the Atlantic Coast

Using previously published atlas data for 122 mixed-species wading bird colonies on islands along the Atlantic coast (Maine to Florida, 1976-77), we examined relationships between population sizes of 11 species of egrets, herons, ibises, and wood storks (Mycteria americana) and nine habitat variables. On nautical charts, we measured four island characteristics (area, length, width, shape), three isolation factors (distances to nearest island, mainland, and a water barrier),, and two variables related to potential feeding habitat within 5 km of the center of the colony (wetland area and land-water interface, i.e., the linear distance between the marsh/upland and all water bodies within the same 5-km radius). One univariable and five multivariable .procedures were used to determine which habitat features were best related to population size .(all species combined). Multicollinearity problems among the variables limited interpretation for most procedures. Both univariable and the multivariable procedures indicated that land-water interface was the most important of the nine variables, but for all models, less than 10% of the total variance was explained (rz is less than 0.10). The size of the colony was not related to the amount of wetland area (within 5-km).per se. Colony data showed better 'structure' when examined on the basis of geographic and disturbance gradients. Population sizes of colonies near man-altered habitats were compared with those surrounded by relatively natural habitats in three geographic zones: north, middle, and south. Significant differences were found in colony size among the three zones (south largest) and between disturbance types. Surprisingly, in all three zones, colonies near man-altered areas were larger on average than those near more natural habitats in this region. A possible reason for this difference is suggested.

Book chapter↗
USGS1986

Analysis of utilization of desert habitats with dynamic simulation

The effects of climate and herbivores on cool desert shrubs in north-western Utah were investigated with a dynamic simulation model. Cool desert shrublands are extensively managed as grazing lands, and are defoliated annually by domestic livestock. A primary production model was used to simulate harvest yields and shrub responses under a variety of climatic regimes and defoliation patterns. The model consists of six plant components, and it is based on equations of growth analysis. Plant responses were simulated under various combinations of 20 annual weather patterns and 14 defoliation strategies. Results of the simulations exhibit some unexpected linearities in model behavior, and emphasize the importance of both the pattern of climate and the level of plant vigor in determining optimal harvest strategies. Model behaviors are interpreted in terms of shrub morphology, physiology and ecology.

Journal of Environmental Management↗
USGS1985

Optimal management strategies in variable environments: Stochastic optimal control methods

Dynamic optimization was used to investigate the optimal defoliation of salt desert shrubs in north-western Utah. Management was formulated in the context of optimal stochastic control theory, with objective functions composed of discounted or time-averaged biomass yields. Climatic variability and community patterns of salt desert shrublands make the application of stochastic optimal control both feasible and necessary. A primary production model was used to simulate shrub responses and harvest yields under a variety of climatic regimes and defoliation patterns. The simulation results then were used in an optimization model to determine optimal defoliation strategies. The latter model encodes an algorithm for finite state, finite action, infinite discrete time horizon Markov decision processes. Three questions were addressed: (i) What effect do changes in weather patterns have on optimal management strategies? (ii) What effect does the discounting of future returns have? (iii) How do the optimal strategies perform relative to certain fixed defoliation strategies? An analysis was performed for the three shrub species, winterfat (Ceratoides lanata), shadscale (Atriplex confertifolia) and big sagebrush (Artemisia tridentata). In general, the results indicate substantial differences among species in optimal control strategies, which are associated with differences in physiological and morphological characteristics. Optimal policies for big sagebrush varied less with variation in climate, reserve levels and discount rates than did either shadscale or winterfat. This was attributed primarily to the overwintering of photosynthetically active tissue and to metabolic activity early in the growing season. Optimal defoliation of shadscale and winterfat generally was more responsive to differences in plant vigor and climate, reflecting the sensitivity of these species to utilization and replenishment of carbohydrate reserves. Similarities could be seen in the influence of both the discount rate and the climatic patterns on optimal harvest strategics. In general, decreases in either the discount rate or in the frequency of favorable weather patterns lcd to a more conservative defoliation policy. This did not hold, however, for plants in states of low vigor. Optimal control for shadscale and winterfat tended to stabilize on a policy of heavy defoliation stress, followed by one or more seasons of rest. Big sagebrush required a policy of heavy summer defoliation when sufficient active shoot material is present at the beginning of the growing season. The comparison of fixed and optimal strategies indicated considerable improvement in defoliation yields when optimal strategies are followed. The superior performance was attributable to increased defoliation of plants in states of high vigor. Improvements were found for both discounted and undiscounted yields.

Journal of Environmental Management↗
USGS1984

A general methodology for maximum likelihood inference from band-recovery data

A numerical procedure is described for obtaining maximum likelihood estimates and associated maximum likelihood inference from band- recovery data. The method is used to illustrate previously developed one-age-class band-recovery models, and is extended to new models, including the analysis with a covariate for survival rates and variable-time-period recovery models. Extensions to R-age-class band- recovery, mark-recapture models, and twice-yearly marking are discussed. A FORTRAN program provides computations for these models.

Biometrics↗
USGS1982

Optimal stochastic control in natural resource management: Framework and examples

A framework is presented for the application of optimal control methods to natural resource problems. An expression of the optimal control problem appropriate for renewable natural resources is given and its application to Markovian systems is presented in some detail. Three general approaches are outlined for determining optimal control of infinite time horizon systems and three examples from the natural resource literature are used for illustration.

Ecological Modelling↗
USGS1982

Examination of woodcock nest sites in central Pennsylvania

Evidence is given to support Iow selectivity in choice of nest sites by woodcock (Philohela minor). Habitat characteristics measured at 30 woodcock nest sites were compared with non-nest control sites in Huntingdon County, central Pennsylvania. Mean nest density per year was 1/4.8 ha within the 54-ha study area. Of 14 characteristics measured, higher shrub-stem density at nests was the only variable significantly different (P < 0.05) from controls by univariate t-test. A multivariate t-test showed no significant difference (P > 0.05) between habitat characteristics at nests and controls. A computed discriminant function also indicated little distinction between nests and controls. Spatial distribution of nests for each of the three years did not depart significantly (P :> 0.05) from a random distribution. Although evidence is given for little overall selectivity, greater shrub-stem density at nests and associations related in part to 'edge' habitat may be important in the location of substantial numbers of woodcock nests. The mean distance from nest site to nearest tree (1.0 ? 1.1 m) and to nearest shrub (22.8 ? 17.8 cm) was significantly less (P < 0.01) than for control areas. Additional research on woodcock nest-site selection may lead to enhanced woodcock production in a variety of habitats.

Wildlife Research Report↗
USGS1981

Discriminant analysis in wildlife research: Theory and applications

Discriminant analysis, a method of analyzing grouped multivariate data, is often used in ecological investigations. It has both a predictive and an explanatory function, the former aiming at classification of individuals of unknown group membership. The goal of the latter function is to exhibit group separation by means of linear transforms, and the corresponding method is called canonical analysis. This discussion focuses on the application of canonical analysis in ecology. In order to clarify its meaning, a parametric approach is taken instead of the usual data-based formulation. For certain assumptions the data-based canonical variates are shown to result from maximum likelihood estimation, thus insuring consistency and asymptotic efficiency. The distorting effects of covariance heterogeneity are examined, as are certain difficulties which arise in interpreting the canonical functions. A 'distortion metric' is defined, by means of which distortions resulting from the canonical transformation can be assessed. Several sampling problems which arise in ecological applications are considered. It is concluded that the method may prove valuable for data exploration, but is of limited value as an inferential procedure.

Book chapter↗
USGS1956

Fifty-sixth Christmas Bird Count. 147. Southern Dorchester County, Md

Summary and Recommendations: We suggest that managers are approaching the limits of their ability to improve waterfowl harvest management, primarily because the information needed to make better decisions is being sacrificed by the current approach to setting regulations. We propose an actively adaptive management strategy in which regulatory decisions play a dominant role in reducing uncertainty about population dynamics. The proposed strategy recognizes 'value' in acquiring knowledge only to the extent that it contributes to the objective of optimizing harvests. To implement this strategy, managers will need: (1) a set of regulatory options, with possible constraints on their use; (2) quantifiable harvest management objectives; (3) a set of models that represent an array of meaningful hypotheses about the effects of regulations on populations; and (4) a measure of credibility (or likelihood) for each model, which can be updated regularly using information from waterfowl monitoring programs. Adaptive optimization is an iterative process in which the harvest-management policy converges over time to one that maximizes harvest under the most appropriate model. At each time step, an optimal regulatory decision is identified based on the state of the system and the model likelihoods. In the next time step, predicted population changes from the alternative models are compared with the actual changes provided by the monitoring program, The likelihoods are increased or decreased to the extent that predicted and actual population changes correspond. These updated likelihoods then are used in setting regulations in the next cycle and the process begins again. This iterative process produces the most informative regulations when uncertainty is prevalent and produces maximum sustainable yields as uncertainty is eliminated. We see no major obstacles to implementing this adaptive strategy, although there are a number of practical considerations. First and foremost, managers should assess the 'value' of learning. Only when there is a high degree of uncertainty about the effects of hunting regulations on population dynamics will the merit of our proposed strategy be evident. We suggest that this almost always will be true given our current understanding of the relationship between annual regulations, survival and population growth in waterfowl. Nonetheless, careful consideration should be given to formulating the set of alternative models. There is no value in distinguishing between models which differ in their mathematical formulation or biological realism, but which suggest similar harvest strategies. We suspect that 'mechanistic' models (i.e., those that attempt to capture the essence of biological processes) will make better candidates for model sets than so-called 'phenomenological' models. Assuming that all model sets include a good approximation of reality, learning rates will be dependent on the quality of monitoring programs. Fortunately, a variety of high-quality monitoring plans for many duck and goose populations of North America, when used with our adaptive approach, should provide new knowledge about population dynamics and response to hunting, and, thus, lead to improved management.

Audubon Field Notes↗