4 Conclusions
The results of this study clearly demonstrate seasonal re-assembly of floodplain aquatic communities across seasons. Although the aquatic communities were largely shaped by the landscape hydrological connectivity gradient after the transition, the effect weakened over time, and the communities were reassembled based more on the local environments. Animals observed in the floodplain responded differently but well to the seasonally varying environment based on their mode of dispersal and life history, and their responses led to the overall seasonal re-assembly of the community across seasons.
Taxa level analysis showed large contributions of amphibious animals such as amphibians and aquatic insects to the seasonal re-assembly of the floodplain aquatic communities. In spring, Rana pirica (frog) and Hynobius retardatus (salamander), and Limnephilidae (caddisfly) and diverse aquatic insect taxa contributed to the β diversity along the hydrological connectivity gradient. We have previously shown that their habitat selection during the flood recession period aligns their spatial distribution along a hydrological connectivity gradient after the flood (Uno et al., 2022). In summer, as oxygen at many lentic habitats drops to the threatening level for many benthic aquatic animals (Gaufin et al., 1974), many amphibians and aquatic insect taxa disappeared from the ponds. For many amphibian and aquatic insect taxa, their disappearance from the water indicates their emergence to land as adults (Kishida et al., 2013; Kawai & Tanida 2005). Yet, some taxa such as Apataniidae, Nemouridae, Leptophlebidae and Ameletiidae disappeared only in ponds but not in the flowing sites, implying their selective mortality in ponds due to low oxygen level (if not other factors such as predation). Some Hynobius retardatus were left in the waterbody in summer and autumn after most individuals left the water, which previous study indicates are overwinter populations exhibiting phenotypic plasticity response to species interactions (Kishida et al., 2013). In autumn, other aquatic insects such as Ephemeridae, Ephemerellidae, Leptophlebiidae and Nemouridae appeared in the waterbodies. Their habitat selection, based on the local environment in autumn and environmental filtering for the remaining assemblages, has likely re-structured the aquatic assemblages in floodplain ponds based on a stronger association with local environmental conditions than with the connectivity gradient.
Seasonal re-assembly of planktons and fishes were not detected in the taxa-level analysis. Planktons were instead quantitatively responding to the local environmental change and the effects of hydrological connectivity were not detected in this study. On the other hand, many fishes were ontogenetically shifting their habitat based on the body size of individuals occupying different habitats as described well elsewhere (Bellmore et al., 2013; Jeffres et al., 2008). Among 11 species observed in our study, Rhynchocypris percnurus andMisgurnus anguilicaudata seemed to complete their life history in floodplain ponds as individuals with all life stages, including juvenile to mature adults, found in floodplain ponds. In contrast, Cottus nozawae was only in sites with constant flow. Eight other taxa, including Rhinogobius sp., Oncorhynchus masou ,Gymnogobius sp., Pseudaspius spp., Lethenteron reissneri , Salvelinus leucomaenis , Parahucho perryi , andBarbatula oreas temporarily occupied the floodplain ponds as juveniles. In floodplain ponds, we observed their juveniles including young of the year individuals and older cohorts, but not mature adults. Many of their mature adults were observed in the mainstem river and/or downstream lake. Our observations indicate these fish species are using the floodplain waterbodies as nurseries for the juvenile fishes (Endo et al., 2023). Because juveniles of these fishes occupy the floodplain waterbodies for more than one year, and some cohorts of respective species stayed in the waterbodies all year, their movements did not contribute to the taxa level re-assembly across season. Seasonal movements of fishes between floodplain waterbodies and rivers need to be determined with more careful observation of their cohorts or direct observations of their movements (Kanno et al., 2022).
Natural floodplains are characterized by their seasonal and/or temporal transitions between aquatic-terrestrial and lotic-lentic environments (Junk et al., 1989; Tockner et al., 2000). Temporal occupation of such dynamic and ephemeral floodplain waterbodies by amphibious or mobile animals observed in this study has been commonly and widely observed (Whiles & Goldowitz 2001; Jeffres et al., 2008; Holgerson et al., 2019). Therefore, the patterns described in this study of seasonal re-assembly of aquatic communities in floodplain waterbodies from spatial structures aligned with landscape level hydrological connectivity gradients during environmental transition periods to distributions corresponding to the local environment in stable periods, maybe widespread. Taking into account such community structures, maintenance of the diverse aquatic organisms in floodplain waterbodies requires co-existence of waterbodies of various hydrological connectivity to the river, as well as waterbodies of various local environments as characterized by size of waterbodies, temperature, and dissolved oxygen level.
Our study highlighted seasonal dynamics of community spatial structures. Although community structures of a given ecosystem are commonly measured and diagnosed by one-time spatial surveys, this study shows that the spatial structure of communities may change temporally, particularly in such dynamic systems as ecotones. Spatial heterogeneities of the environment at different times of a year can interact, and the spatial connectivity can also restrict the transition of the seasonal re-assembly of animals across seasons. The temporal and spatial dynamics of the natural environment structure the communities in combination, and their holistic understanding would be required for the understanding of such dynamic ecosystems. Furthermore, while this study only examined change in the spatial patterns, the communities at one season would interact with those in another season as ecological legacies, through various factors such as population dynamics of each member of the community, species interactions, nutrient dynamics, and so on (Richardson 1991; Power et al., 2008; Yang 2008). Further studies would show spatially and temporally interacting natural community structures in dynamic environments.
Author contribution
Hiromi Uno: lead all the research activity including Conceptualization, Data Curation, Formal Analysis, Funding Acquision, Investigation, Methodology, Project Administration, Resources, Supervision, Validation, Visualization, Writing-original draft, and review and editing. Shunsuke Utsumi: equally Conceptualization, Formal analysis, Investigation and review and editing. Kentaro Morita equally contributed to the conceptualization, investigation, and review and editing. Osamu Kishida equally contributed to the conceptualization, investigation, and review and editing. Md. Khorshed Alam equally contributed to investigation, and review and editing. Junjiro Negishi equally contributed to the conceptualization, investigation, supervision, and review and editing
Data availability statement
Authors confirm that we will deposit our data in a public repository and indicate the repository of our choice upon acceptance.
Funding statement
KAKENHI Grant Number 22K14915 and 22KJ0004 to H.U. from the Japan Society for the Promotion of Science.
Conflict of interest disclosure
Authors claim no conflict of interest for this manuscript
Ethics approval statement
Our work conforms to the guidelines for the proper conduct of animal experiments in Japan. Field survey of aquatic animals was approved by the fishery and forestry department of Hokkaido (ID 610).
Patient consent statement: nothing to declare
Permission to reproduce material from other sources: nothing to declare
Clinical trial registration: nothing to declare