Abstract. Increases in temperature and grazing intensity are believed to promote the relative abundance of C 4 plants in grassland communities in Inner Mongolia. However, there is a lack of understanding as to which factor is the primary driver at the household scale. The relative abundance of C 4 plants in grassland communities within 32 households was monitored over a 5-year period in the typical steppe region of Inner Mongolia. The relationships between the mean annual temperature, grazing intensity and their combinations on the patterns of the relative abundance of C 4 plants across the land managed by these households were analysed. The results showed that (1) the herbage mass of the typical steppe grassland was mainly composed of C 3 plants (87%); (2) the C 4 plants were more sensitive to, and can be used as indicators of, environmental changes. These C 4 species included Cleistogenes squarrosa (Trin.) Keng, Chenopodium glaucum Linn. and Salsola collina Pall.; (3) both increasing temperature and grazing intensity promoted the relative abundance of C 4 plants. Grazing intensity was the primary driver of the change in relative abundance of C 4 plants in this region. Not only did grazing change the micro-environment of the grasslands, but also the C 3 species were preferentially grazed by the livestock. Comparison of the results with previous studies on the temporal variation in the abundance of C 4 plants suggests that the relative importance of grazing and climatic factors depends on the spatial scales of the studies, with climate being of greater importance at the regional rather than the household scale.

Abstract. Understanding the relationship between the aboveground net primary production (ANPP) and annual precipitation in arid and semiarid grasslands is crucial for assessing the effects of climate change on grassland ecosystems. The temporal pattern of ANPP, based on long-term data on a semiarid ecosystem in Inner Mongolia, was investigated. The biomass of perennial grasses, perennial forbs and Stipa grandis P. Smirn., showed a positive relationship with annual precipitation. The amount of annual precipitation also changed the annual biomass of 13 other dominant species and consequently the ANPP. The coefficient of variation of the ANPP of the plant community was lower than the coefficient of variation of annual precipitation. Irrespective of the strong inter-annual variation in annual precipitation, the positive relationship found between ANPP and annual precipitation suggests the dependence of ANPP upon hydrological variations in typical steppe. Our findings highlight the importance of dominant perennial species and functional groups in mediating the responses of ANPP to annual precipitation in the typical steppe in northern China.

Abstract. Herders' perceptions are important in relation to adopting adaptive behavioural approaches to climate change. A survey of 1080 herder households was conducted in six rangeland regions of Inner Mongolia in northern China to investigate relationships between herder's perceptions and actual climate change. Across all six regions, more than 90% of herders perceived a decline in precipitation in the past 30 years, whereas actual data showed no significant change. Many herders also perceived an increase in temperature, which was in agreement with the temperature data, although the proportion of herders who perceived a change in temperature differed among regions. A further survey of 58 households was conducted to better understand the reasons for these differences, by simultaneously surveying their perceptions of the changes in both rainfall and rangeland condition. The results showed that the herders' perceptions of a decrease in rainfall could be influenced by their perceptions of the changes in the condition of their rangeland (reductions in production). An alternative explanation could relate to rainfall being interpreted in terms of the frequency and pattern rather than the absolute amounts. These findings have important implications for understanding how herders' perceptions can improve their adaptive behavioural responses to climate change.