Vulnerability and risk management of Agave species in the Tehuacan Valley, Mexico

Background: Our study analysed the vulnerability of the useful Agave species of the Tehuacan Valley, Mexico, considering ecological, cultural and economic aspects, and management types. We hypothesized that management intensity is proportional to the degree of risk of a species in order to decrease its vulnerability. Methods: Distribution of Agave species was monitored in 36 types of plant associations. Ethnobotanical studies were conducted in 13 villages and six markets. The vulnerability of each species was calculated by assigning risk values to the variables analysed. The vulnerability and management intensity indexes were estimated through the scores of the first principal component of PCA. Variation of management data explained by ecological, cultural and economic information were analysed through canonical correspondence analyses (CCA). A linear regression analysis identified the relation between vulnerability and management intensity. Results: We recorded presence of agave species in 20 of 36 vegetation types. Out of 34 Agave species, 28 were recorded to have one to 16 use types; 16 species are used as food, 13 for live fences, 13 for producing 'pulque', 11 for fibre and ornamental, 9 for construction. Seven species are used for preparing mescal, activity representing the highest risk. Seven Agave species are exclusively extracted from the wild and the others receive some management type. Incipient cultivation was identified in A. potatorum whose seedlings are grown in nurseries. Intensive cultivation through vegetative propagation occurs with domesticated species of wide distribution in Mexico. The highest management intensity values were recorded in widely distributed, cultivated and domesticated species, but the regionally native species more intensively managed were those with higher demand and economic value, protected by collective regulations because of their scarcity. The regression analysis indicated significant relation (R-2=0.677, P<0.001) between vulnerability and management indexes. CCA explained 61.0% of variation of management intensity, mainly by socio-cultural factors (30.32%), whereas ecological data explained 7.6% and the intersection of all factors 21.36%. Conclusions: The highest vulnerability was identified in wild species restrictedly distributed and/or highly extracted. Social pressures may increase the natural vulnerability of some species and these species are particularly those native species receiving some management form.