The analysis of plant-herbivore interactions between ungulates and vegetation on alpine grasslands in the Himalayan region of Nepal

Abstract

During FAO Project NEP/85/011 a methodology was developed for His Majesties Government ‘Department of National Parks and Wildlife Conservation’ (HMG DNPWC) to determine the impact of lifestock and wild ungulates on condition and physiognomy of alpine grasslands and to derive parametrical relationships between livestock/wildlife distribution and grassland condition.

Between Nov. 1986 and March 1988 eight surveys, lasting from 2–9 weeks were undertaken in various regions of central and eastern Nepal. The comparison of three climatically similar areas (Hinku-Hongu, Sagarmatha, Langtang) showed significant differences in the condition (e.g. grass cover, grass height, bare topsoil, biomass) and physiognomy (vegetation composition) of alpine grasslands. These differences seemed to be not so much a function of ungulate density, but were mainly determined by grazing duration, i.e. winter grazing as well as topographical features.

Overgrazed areas with a decrease of graminees and an increase of unpalatable shrubs (e.g. Rhododendron anthopogon, Berberis sp.) and herbs (Euphorbia wallichii and E. longifolia, Iris sp.) were particularly abundant in the vicinity of villages. Productivity of grasslands seemed to be highest at altitudes between 3400 and 4000 m ab. S.L. Here grass species showed highest coverage, shrubs and herbs only moderate values. Habitat utilisation of lifestock (cattle, water buffalo, yak, hybrids, goats and sheep) as well as of wild ungulates, such as thar (Hemitragus jemlahicus), bharal (Pseudois nayaur), goral (Nemorhaedus goral goral) and serau (Capricornis sumatraensis tahr) were compared. Productivity (biomass and grass coverage) of grasslands could be correlated with thar density.

Differences of utilisation patterns between wild ungulates e.g. bharal and thar, but also between livestock and thar suggested, that some form of competitive exclusion is operating. With the increase of wild ungulates in Sagarmatha N.P., due to protection, conflicts with livestock grazing were evident.

In tourist areas, changes in the structure of lifestock populations-mainly as a response to tourism-, have lead to decreasing utilisation of high alpine pastures (>4000 m a S.L.). This pattern also reflects human emigration rates of up to 12% annually. Whereas this does not necessarily benefit the upper regions, it does increase grazing problems in lower regions, in particular the grazing of forests.

Two examples of stepwise multiple regression analysis are given to detect parametrical relationships between different numbers and combinations of independent variables. Biomass as well as lifestock numbers could be predicted accurately (r=0.70 (0.84), 0.001<p<0.005 (p<0.0001) with three (four) independent variables.