Last Updated on Tuesday, 05 May 2009 17:57 Written by Administrator Tuesday, 12 August 2008 22:33

PREDICTING GRAZER DISTRIBUTION WITH GRASS QUALITY AND QUANTITY PARAMETERS

Predicting species distribution is an important part of ecology. There are numerous factors influencing the distribution of herbivores in an ecosystem. The distribution of food resources within an ecosystem is an important explanatory factor for herbivore distribution. Hence, the distribution of grass quality and quantity parameters (GQQPs) could explain grazer distribution. The goal of this study is to (i) explore the possibility of predicting grazer distribution with grass quality and quantity parameters, making use of new statistical modeling techniques, (ii) explore which parameters have most influence on grazer distribution in the wet and the dry season and (iii) create resource maps for grazer species for practical purposes. The study area is the Greater Makalali Private Game Reserve (GMPGR), South Africa. Maps of GQQPs (grass species abundance, herbaceous biomass, grass species richness and herbaceous coverage) are created by relating field measurements with Landsat ETM+ bands and variables derived from a digital elevation model using Generalized Additive Modeling (GAM). The accuracy of the GQQP maps was not very high, which could be caused by the date the Landsat ETM+ image was taken. The GAM method proved to be a flexible and empirical method. In 2005 and 2006 sighting locations of grazers were recorded throughout the GMPGR. The maps of the GQQPs are used to predict zebra (Equus burchelli) and wildebeest (Connochaetes taurinus) distribution in the wet and the dry season making use of Ecological Niche Factor Analysis (ENFA).

By applying a quantile reclassification to the values of the GQQP maps, so that the GQQP values have a uniform distribution, the grazer distribution was reliably predicted for the wet season. Predicting grazer distribution in the dry season wasn't successful, which could be because other factors than the GQQPs determine the distribution in the dry season, or be a result of the sampling method of the grazer locations, or be caused by a possible change in the values of the GQQPs in the dry season. From the results of the ENFA, 'resource maps' were calculated that mapped the habitat suitability for the grazers. One of the findings is that the only factors positively influencing the distribution of both zebra and wildebeest in the wet season, is a high abundance of Urochloa mossambicensis and in lesser extent a high herbaceous biomass. All other GQQPs had a negative influence on grazer distribution, which could be because the environment a certain grass species grows in is of more influence on grazer distribution than the grazing value of the grass species. This study describes a method that is suited to obtain grazer resource preference in a specific area, but is less suited for studying the universal resource preference of a certain grazer. For wildlife and range management this method provides detailed information on the food selection of grazers within a certain area. The method simultaneously gives insight into grazer food preference and the spatial distribution of the preferred food resources in the area.