Mongolian herders’ vulnerability to <Emphasis Type="Italic">dzud</Emphasis>: a study of record livestock mortality levels during the severe 2009/2010 winter

The livelihoods of people inhabiting inland Eurasia have long been jeopardized by repeated natural hazards associated with a harsh environment and a cold, arid climate. Dzud is a Mongolian word indicating harsh winter conditions. In the present study, we considered dzud damage (e.g., livestock loss) to result from a combination of climate hazard (e.g., cold surges) and herders’ socioeconomic vulnerability. For this study, we integrated crucial socioeconomic factors accounting for major spatiotemporal variations in Mongolia by applying principal component analysis (PCA) to a comprehensive province-level, multiyear dataset. We subsequently characterized the regionality of herders’ vulnerability to the dzud event that occurred during the 2009/2010 winter by conducting a cluster analysis of the provincial principal component (PC) scores for the pre-dzud year (2009). Our results revealed a distinct geographical pattern of vulnerability. Herding households in the northern and northeastern (relatively wet and plain) areas were found to be well prepared for harsh winters, with shelters against wind and availability of forage, including hay, as well as easy access to major urban markets. By contrast, herding households in the southern and southwestern (arid and mountainous) areas were poorly prepared, with inadequate circumstances that facilitate pursuing of otor (movement of nomadic herders in search of better pastures) and lack of access to markets and dzud relief support because of their remote locations. The time coefficients of PC 2, related to winter preparedness, indicated that vulnerability increased between 2003 and 2009 (the pre-dzud year). This was partly responsible for the record-level mortality observed in 2010 across the southern and southwestern rural region, in conjunction with harsh winter weathers.     

Assessment of drought frequency, duration, and severity and its impact on pasture production in Mongolia

Drought frequency, duration, and severity and its impact on pasture productivity in the four main vegetation zones of Mongolia were analyzed using meteorological, soil moisture, and vegetation data during the growing season (April–August) of 1965–2010. Meteorological and pasture drought characteristics were explored using the Standardized Precipitation Index (SPI), the soil moisture anomalies percentile index (W _p), and Palmer Drought Severity Index (PDSI) on 1-month timescale. Generally, 35–37 (15–16 %) by SPI for meteorological drought while 27–29 (12–13 %) by W _p, and 16–21 (7–9 %) by PDSI for pasture drought with different durations were identified over the four vegetation zones during the study period. Most of these droughts (80 % by SPI and 50–60 % by both W _p and PDSI) observed during the entire events occurred on a 1-month duration with moderate intensity. Drought frequencies were not significantly (p > 0.05) different within the four zones. The frequency of the short-term meteorological droughts was observed relatively greater than pasture droughts; however, pasture droughts were more persistent and severe than meteorological droughts. The three indices show that the frequency and severity of droughts have slightly increased over the 46 years with significant (p < 0.05) dry conditions during the last decade of 2001–2010 in the four zones (except in the high mountain). The results showed the W _p was more highly significantly correlated with the precipitation anomalies (r = 0.68) and pasture production (r = 0.55) than PDSI (r = 0.51, p < 0.05 and r = 0.38, p < 0.10, respectively). A statistical model, based on pasture production and the W _p, suggested that the consecutive drought months contribution during the growing season was 30 % (p < 0.05) and that pasture production was more sensitive to the occurrence of droughts during June–August (R ² = 0.32, p < 0.05) as seen in 2000–2002 and 2007. We concluded that a greater severity and frequency of growing-season droughts, during the last decade of 2001–2010, have driven a reduction in pasture production in Mongolia.     

Contributions of multiple climate hazards and overgrazing to the 2009/2010 winter disaster in Mongolia

Mongolian pastoral husbandry is subject to various climate hazards such as dzud (Mongolian for “severe winter conditions”). Dzud in the 2009/2010 winter affected 80.9% of the country and killed more than 10 million livestock (23.4% of the total). To understand the natural and man-made mechanisms of this dzud, we examined the contributions of dzud-causing factors such as climate hazards (cold temperatures and heavy snow) and winter–spring livestock grazing (measured as overgrazing rate), which created a distinct regional pattern of high livestock mortality using a regression tree method. The regression tree model accounted for 58% of the total spatial variation of the mortality and identified various types of dzud in each region. Results showed that during the 2009/2010 winter, almost all of Mongolia experienced extreme cold temperatures, with abnormally large amounts of snow. In addition, more than half of the territory was overgrazed because of the lower pasture biomass resulting from summer drought and livestock overpopulation. At the regional scale, high livestock mortalities occurred in moderately to heavily overgrazed regions in south-central and western Mongolia, resulting from the combination of these factors. Conversely, areas with lower livestock mortalities (or non-dzud) coincided with sufficient pasture capacity in the north and east, even under extreme cold and snow. This indicates the importance of controlling the number of livestock to below the pasture carrying capacity regardless of an inter-annually varying climate. Moreover, we identified critical thresholds of each factor across which serious disasters occurred. These thresholds are practically useful for future livestock management of pasture land.