Monitoring of Khulans and Goitered Gazelles in the Mongolian Gobi – Potentialand Limitations of Ground Based Line Transects

Central Asian remote rangelands are home to several charismatic, rare and far ranging ungulates which are increasingly becoming under pressure from human encroachment. Population monitoring is challenging due to the vast expanse of the species ranges, tight budgets and limited availability of suitable fixed winged-aircraft. Consequently, many current population estimates are based on pragmatically designed ground-bound transect surveys. Although, ample literature exists on how to design surveys in an ideal world, little effort has been made to demonstrate the potential and limitations of a time-series of ground-bound transect surveys under real world constrains. Since 2003 we have been monitoring the two sympatric steppe ungulates, Asiatic wild ass (“khulan”, Equus hemionus) and goitered gazelles (“gazelle”, Gazella gutturosa), in the Great Gobi B Strictly Protected Area in south-western Mongolia using ground-bound line transects. Both species showed clear species-specific seasonal variation in group sizes which seem related to birthing and mating periods. Data on annual recruitment were impeded by the long flight distances and the difficulty to reliably identify and count young of the year. Distribution of khulans and gazelles showed clear speciesspecific seasonal patterns and highlighted the importance of two oasis complexes. Population estimates of 33 surveys covering 10,383 km2 were highly variable even between consecutive surveys and had huge 95% confidence intervals (khulan: range: 1,707 to 45,040, gazelles: range: 2,564 to 10,766) making them unsuitable to obtain robust baseline population estimates. Although our individual surveys were poor measures of population abundance, they provided important data on group sizes and species distribution and are presently used for Bayesian hierarchical trend modelling and species specific habitat suitability analysis. The ground surveys are relatively inexpensive as compared to aerial surveys and thus can be conducted at short temporal intervals, engaging park staff and researchers with local people thereby helping mutual understanding, information transfer, and detection of illegal activities.