Introduction
The high rates of change in natural environments direct and indirectly caused by human activities may exert negative effects on terrestrial and aquatic ecosystems (Secretariat of the Convention on Biological Diversity 2020). In lotic environments, human modifications at local (e.g., removal of riparian vegetation), regional (e.g., river dam) and global scales (e.g., climate change) can change water flow, and regularity of flood and flash flood events (Allan 2004, Dudgeon et al. 2006, Poff et al. 2007, IPCC 2014). Changes made to the surroundings of lotic systems, such as the removal of riparian vegetation and the linearization of water courses, reduces the area available for water infiltration, increasing the occurrence of flood and flash flood events (Resh et al. 1988). Furthermore, global climate changes tend to alter the precipitation regime on a global scale (IPCC 2014, Fischer and Knutti 2015, Donat et al. 2016). In tropical regions, precipitation tends to become more concentrated in shorter periods of time, and, consequently, the amounts of heavy precipitation events is expected to increase in the coming years (IPCC 2014, Fischer and Knutti 2015). Then, the flow regime of lotic systems will change due to an expected increase in disturbance events caused by altered precipitation. Despite this general expectation, few studies have experimentally addressed the impact of such changes in the biodiversity of tropical streams. Manipulative experiments are key for understanding the mechanisms behind biodiversity response to environmental changes because, they can separate the effects of different drivers related with the disturbance.
For a long time aquatic environments have been undergoing hydrological disturbances, and studies addressing this topic are essential to understand ecosystem dynamics and biodiversity patterns (Townsend et al. 1997b, Lake 2000, Melo et al. 2003, Sutherland and Freckleton 2013, Shabarova et al. 2021). Historically, knowledge on the effects of hydrological disturbances on biodiversity (e.g., water withdrawal or flash floods) usually focus on taxonomic measures such as species richness and beta diversity indices (Death 1996, Lake 2000, Melo and Froehlich 2004). However, environmental disturbances can affect species differently depending on their functional traits, which still needs to be further studied for aquatic systems (Menezes et al. 2010). Community functionality is an important component of biodiversity, once function considers not only the identity of organisms, but the physiological, behavioral and/or ecological responses of species to the environment (Verberk et al. 2013).
Traits related to permanence and colonization are important for aquatic communities when facing hydrological disturbances. Permanence and colonization traits can be inserted in the so-called resistance/resilience axis, which is relative to the organism occurrence (Townsend et al. 1997a, Verberk et al. 2013, Sarremejane et al. 2020, Wilkes et al. 2020). Resistance traits, such as morphological adaptations for fixation on substrates and body shape, allow organisms to maintain persistent population sizes when resisting disturbances (Mazzucco et al. 2015). Resilience traits are related to life cycle length such that short life cycles enable faster colonization in newly impacted environments. The relationship between resistance and resilience traits is a trade-off, once organisms that have a high resistance to disturbance tend to present reduced resilience characteristics, whereas organisms with low resistance to disturbance tend to have increased resilience (Kappes et al. 2014, Ferzoco 2019).
Functional traits related to the use of the environment (Wilkes et al. 2020), such as for food acquisition and to avoid competition, also affect community functionality. Hydrological disturbances can affect the relative role that functional traits within the resistance/resilience and environment use axes have on organism distribution. An increased environmental disturbance tends to reduce the importance that environmental drivers and competitive interactions have on the functional diversity of communities in a given environment (Lopez et al. 2016). If environmental requirements become less relevant when environmental disturbance is high, aquatic communities will tend to be structured according to traits related to resistance and resilience. In such circumstances, we can expect organisms with different attributes to have different responses to hydrological disturbances. On the other hand, differences in the intensity and frequency of disturbances would select different traits in the communities, leading to different trajectories in the structure of the communities over time.
Under this perspective, we conducted a manipulative hydrological experiment in streams of the Cerrado hotspot to test whether modifying stream flow alter the rate of occurrence and colonization of aquatic insects with different functional traits. We tested two general hypotheses: i) an increased disturbance intensity would reduce the persistence of organisms in the substrates; and ii) higher disturbance frequency would negatively affect organisms that maximize the permanence in the environments. To test these two hypotheses, we categorized the aquatic insects into nine groups with similar functional traits, both morphological and ecological. For each group we established hypotheses related to the permanence and colonization parameters. We designed an experiment in a space-state design (Pedersen et al. 2011), changing both intensity and frequency of water flow, and evaluated the occurrence of the genera. We also estimated how these changes influenced the probabilities of permanence and colonization of each functional group using a Bayesian model. We showed that the frequency of the disturbance was more relevant in changing the probabilities of persistence and colonization, and that this effect is dependent on the functional group.