Assessment of distribution patterns for the grey kangaroo within southern Queensland : an application and appraisal of faecal pellet counts and aerial survey for the study of grey kangaroo density and distribution

Abstract

The study utilizes the wildlife investigational techniques of faecal pellet counts and aerial survey to examine relationships between grey kangaroos (Macropus giganteus) and the habitats they occupy in southern inland Queensland. In accomplishing this aim an assessment is made of the usefulness of the techniques for the study of grey kangaroo density and distribution.Faecal pellet counts are found to provide valid indices of grey kangaroo density because: (a) the shape and size of pellets differ from those of other macropods (within the study areas); (b) defecation rate appears to be reasonably consistent; (c) pellet decay rates and determination of pellet ages may be ascertained satisfactorily; and (d) two separate pellet censuses for an area of constant kangaroo density returned similar results. The use of pellet groups, rather than single pellet counts, is not, however, appropriate for censuses as the number of pellets per pellet group varies significantly between habitats. Surveys based on counts of individual pellets are necessary if the method is to be used in kangaroo research. Defecation is strongly associated with feeding activity. Pellet counts therefore provide reliable estimates of the feeding distribution of the grey kangaroo.Analysis of variance, regression analysis and cluster analysis are employed in order to define habitat preferences, as defined by pellet counts, within the Durikai area. Of these, cluster analysis is found to be the most effective technique. Results identify a preference for habitats displaying an abundance of both shelter and forage reserves. Where habitats meeting this criterion are available climax forest and cleared land, which feature an abundance of one component (shelter or forage) in the absence of the other, receive relatively little use.An analysis is made of the usage grey kangaroos make of cleared country on a seasonal basis in southeastern (Durikai) and southwestern (BolIon) Queensland. Distribution trends defined from pellet counts are related to pasture status as defined by visual assessment, rainfall totals, soil moisture storage and pasture growth models. The study demonstrates that grey kangaroos are sensitive to changes in forage status, particularly pasture phenology. As quality and quantity of forage reserves decline, grey kangaroos make progressively greater use of cleared areas. This results in marked seasonal changes in distribution patterns. Of the variables quantified, rainfall provides the best correlations with kangaroo usage of cleared country, explaining 55 per cent of the variation in distribution at Durikai and 91 per cent at Bollon.For faecal pellet data from cleared country it is established that grey kangaroo density declines in a systematic manner as distance from shelter (wooded country) increases. The rate of decline is adequately described through use of the general negative exponential model. The findings illustrate the reluctance of grey kangaroos to venture far from shelter although the distances involved are well in excess of previously suggested limits. It is established that where the rate of decline is near constant for a series of surveys, data gathered by sampling close to the forest edge may be used to predict densities farther out into the clearing. A similar analysis for forested country establishes that there is no relationship between kangaroo density and the proximity to cleared land.A further assessment verifies that the faecal pellet data adhere to the negative binomial distribution. As the dispersion parameter varies widely between surveys, however, it is not possible to make precise statistical comparisons between separate surveys. The degree of clustering in the pellet data (between surveys) appears to be related to seasonal patterns of habitat usage.Aerial surveys are used to obtain estimates of grey kangaroo density for a 47100km2 strip of southern inland Queensland. Landsat imagery is used to define broadscale habitat zones and results are computed on a zonal basis. Habitat data at this scale form a useful base for the design of aerial surveys. The population estimate of 746 359 ± 79 070 (N ± S.E.) establishes that grey kangaroo density for this part of Queensland is high, being highest in the areas supporting the highest exploitation rates (commercial shooting). A comparison between separate surveys, however, suggests that the currently used correction factors, which adjust for ease of sighting kangaroos in differing habitats, are invalid. The seasonal movement patterns of the grey kangaroo lead to variations in aerial counts.