Opportunities to integrate herders' indicators into formal rangeland monitoring: an example from Mongolia

摘要

Despite increasing calls for knowledge integration around the world, traditional knowledge is rarely used in formal, Western-science-based monitoring and resource management. To better understand indicators herders use and their relationship to researcher-measured indicators, we conducted in-depth field interviews with 26 herders in three ecological zones of Mongolia. We asked each herder to (1) assess the overall condition of three different sites located along a livestock-use gradient from their winter camp using a numeric scale, (2) describe the indicators they used in their assessment, and (3) explain what caused their pastures to remain healthy or become degraded. At each site, we collected field data on vegetation variables and compared these with herders' ratings and indicators using linear regression. We used classification and ordination to understand how herders' assessment scores related to plant community composition, and determine how well multivariate analysis of factors determining plant community composition aligned with herders' observations of factors causing rangeland change. Across all ecological zones, herders use indicators similar to those used in formal monitoring. Herders' assessment scores correlated significantly and positively with measured total foliar cover in all three ecological zones, and with additional measured variables in the steppe and desert steppe. Ordination revealed that herder assessment scores were correlated with the primary ordination axis in each zone, and the main factors driving plant community composition in each zone were the same as those identified by herders as the primary causes of rangeland change in that zone. These results show promise for developing integrated indicators and monitoring protocols and highlight the importance of developing a common language of monitoring terminology shared by herders, government monitoring agencies, and researchers. We propose a new model for integrating herder knowledge and p