ABSTRACT
Inundation and fire can affect the structure of riparian vegetation in wetlands. Our aim was to verify if there are differences in richness, abundance, basal area, composition and topographic preference of woody species in riparian forests related to the fire history, flooding duration, or the interaction between both. The study was conducted in the riparian forests of the Paraguay River some of which were burned three times between 2001 and 2011. We sampled trees with a girth of at least 5 cm at breast height in 150 5 × 10 m plots (79 burned and 71 unburned). We also measured height of the flood mark and estimated the flooding duration of each plot. We performed Generalized Linear Mixed Models to verify differences in richness, basal area, and abundance of individuals associated to interaction of fire and inundation. We used an analysis of similarity (ANOSIM) and indicator species analysis to identify differences in composition of species and the association with burned and unburned area according to different levels of inundation. Finally, we used a hierarchical set of Generalized Linear Models (GLM), the so-called HOF models, to analyse each species' specific response to inundation based on topography and to determine their preferred optimal topographic position for both burned as well as unburned areas. Richness was positively associated with elevation only in burned areas while abundance was negatively influenced by inundation only in burned areas. Basal area was negatively associated with time of inundation independent of fire history. There were 15 species which were significant indicators for at least one combination of the studied factors. We found nine species in burned areas and 15 in unburned areas, with response curves in HOF models along the inundation gradient. From these, five species shifted their optimal position along the inundation gradient in burned areas. The interaction of fire and inundation did not appear to affect the basal area, but it did affect the richness, number of individuals, success of some species, and seemed to shape the boundary of these forests as shown by the difference in the positioning of these species along the inundation gradient.
Subject(s)
Disasters , Fires , Floods , Forests , Trees/physiology , Brazil , EcosystemABSTRACT
BACKGROUND: Wild plants are used as food for human populations where people still depend on natural resources to survive. This study aimed at identifying wild plants and edible uses known in four rural communities of the Pantanal-Brazil, estimating the use value and understanding how distance to the urban areas, gender, age and number of different environments available in the vicinity can influence the knowledge and use of these plants by local people. METHODS: Data on edible plants with known uses by communities were obtained through semi-structured interviews. A form with standardized information was used for all communities in order to obtain comparable data for analysis. For the quantitative analysis of the factors that could influence the number of species known by the population, a generalized linear model (GLM) was conducted using a negative binomial distribution as the data consisted of counts (number of citations). RESULTS: A total of 54 wild species were identified with food uses, included in 44 genera and 30 families of angiosperms. Besides food use, the species are also known as medicine, bait, construction, technology and other. The species with the highest use value was Acrocomia aculeata. Older people, aged more than 60 years, and those living in more remote communities farther from cities know more wild edible plants. Statistical analysis showed no difference regarding gender or number of vegetation types available in the vicinity and the number of plants known by locals. CONCLUSION: This study indicated more knowledge retained in communities more distant from the urban area, indifference in distribution of knowledge between genders and the higher cultural competence of elderly people in respect to knowledge of wild edible botanicals.