Comunidade biótica em rios intermitentes: como as assembleias se estruturam nas fases aquáticas do ciclo hidrológico de intermitência?

Abstract

Intermittent rivers are dynamic systems that present three different phases along their hydrological cycle: lotic (flowing water), lentic (isolated pools) and terrestrial (dry riverbed). Despite the unique dynamic of intermittent rivers, only in recent years, the number of ecological studies that aim investigate your hydrological cycle dynamic increased. These systems have expanded their coverage area around the world due to climate change and extensive use and that can explain the increase in ecological studies in these rivers. It is necessary to understanding the dynamics of intermittent rivers for effective management and inclusion in biomonitoring and recovery programs. In natural intermittency cycle, the drying of the water flow and the formation of isolated pools is a natural disturbance experienced periodically by the community. Thus, resilient communities must reorganize in an alternative structure, after water flow drying and riverbed become a mosaic of isolated pools and terrestrial habitat. That is the result from the effects of the disturbance on the different species of a community. In addition, persistent pools could serve as an essential refuge during the dry period. The aim of this study was to answer the question: How is the biotic community structured between the aquatic phases of the intermittent hydrological cycle? The hypothesis was: the formation of a mosaic of terrestrial and aquatic patches in the riverbed alters differently the structures of fish and macroinvertebrates assemblages. This study was conducted in five streams among Acaraú river catchment, located in Ceará, Northeast, Brazil. The streams were sampled in lotic and pool phases of the intermittent hydrological cycle in 2019. In lotic phase, the streams were divided into ten sections, with transect in the beginning and middle of sections. The physicochemical parameters and depth were measured in five equidistant points in each transect. In pool phase, five transects were established and the same protocol of the lotic phase was followed. In each section of lotic phase and each pool, riparian shading, shelter, macrophytes and flow velocity were estimated. Fish were collected with ten trails and 20 sieve by stream in lotic phase, while for macroinvertebrates, 10 samples (each sample consisting of 3 subsamples in different micro-habitats) were collected in each stream. The number of pool per stretch and sampling of the assemblages in this phase varied according to the availability and size of the pools. The composition of the fish assemblage did not differ between the phases of hydrological cycle, Astyanax fasciatus, Serrapinnus heterodon and Phenacogaster calverti were the dominant species in lotic phase. In pool phase, dominant species were Hemigrammus rodwayi, Poecilia reticulata and Astyanax fasciatus. However, no one species of fish was considered an indicator of any of the aquatic phases of the hydrological cycle. Thus, this assemblage proved to be resistant to the disturbances of the cycle. Macroinvertebrate assemblage presented different compositions between the phases of the hydrological cycle, in addition, the dominant taxa also differed between the phases. Chironomidae, a dominant taxon in lotic phase and Chaoboridae, dominant in pools, belong to the order of aquatic insects Diptera, characterized by presenting larval aquatic phase and aerial adults. Baetidae, Belostomatidae, Dytiscidae and Libellulidae were considered indicator taxa of pool phase. ProTest associated with PROCRUSTES analysis demonstrate there was no significant concordance between the assemblages in the lotic and pool phases. The habitat structure differed between the aquatic phases of the hydrological cycle. Lotic phase demonstrates greater environmental heterogeneity than the pool phase. In lotic phase, 11 species of fish and 13 taxa of macroinvertebrates were significantly correlated to 13 and eight environmental variables, respectively. The substrate silt was the most important variable and was significantly correlated to three species of fish and six taxa of macroinvertebrates. In pool phase, six species of fish and 19 taxa of macroinvertebrates were significantly correlated to six and 10 environmental variables, respectively. Most important variables for fish assemblage were depth, presence of parts of live trees in the stream bed and the dissolved oxygen rate. For macroinvertebrates, the substrates sand, silt and boulders were more relevant. The presence of aquatic refuges (pools) was important for remain of fish species, an assemblage more closely linked to the presence of water. However, the dominance of aquatic insects and the differentiation in the composition of macroinvertebrate assemblage between the phases of the hydrological cycle show that in addition to the presence of refuges, dispersion is a fundamental mechanism of resilience in dynamic systems such as intermittent rivers.