University of Tasmania
Browse
Botwe_whole_ thesis.pdf (3.92 MB)

Taxa vs traits: analysing stream communities in South Australian dry environments

Download (3.92 MB)
thesis
posted on 2023-05-27, 12:03 authored by Botwe, PK
The availability of water is critical to ecosystem functioning and human health, but it is impacted by anthropogenic activities, such as those associated with intensification of surrounding land use. These can result in a range of aquatic impacts, including: changes to water quality, alteration of aquatic communities and modification of flow regimes. Consequently, stream ecosystems are monitored or assessed using a range of techniques to inform management strategies that attempt to minimise deleterious impacts and maintain the ecological integrity of stream ecosystems. This thesis evaluates two important approaches to assessing biological conditions in stream ecosystems: taxonomic and trait-based approaches. Many studies have assessed stream condition using taxonomic characteristics of macroinvertebrates (i.e., their identities and abundance). While valuable, limitations of this approach are that taxonomic studies are restricted in their extendibility across studies, owing to biogeographic variation in species identities across different regions, and low proportion of variance explained when relating taxonomic composition to environmental factors. Further, taxonomic approaches often do not improve our understanding of the causal mechanisms by which impacts occur. In response to these limitations, functional traits of macroinvertebrates have been proposed as alternative measures of the biological condition of streams. This proposal is based on the potential for trait quantification to overcome biogeographic effects, potentially explain greater variation in aquatic community composition, and improve our understanding of the casual mechanisms by which environmental disturbances affect ecological communities. Using 13 years of data sampled biannually across multiple sites in South Australia, I first undertook a traditional, taxonomic-based study. I related land-use, geographic and environmental variables to community composition in order to determine the most influential physical variables structuring these communities (Chapter 2); in addition to the known importance of flow. I found that physical variables explained 23.7% and 27.3% of the variation in taxonomic space in autumn and spring respectively, with salinity being the most strongly associated variable to community structure in both seasons. Geographic location variables (latitude and longitude) were also predictive of community structure in both seasons, suggesting a strong biogeographic effect on communities. Subsequently, I focussed on salinity and flow as predictors of macroinvertebrate community structure. Secondly, I systematically reviewed the effects of flow and salinity on macroinvertebrate traits (Chapter 3) to better understand the temporal and geographic extent of previous studies, the methodological and analytical approaches used to relate flow and salinity to macroinvertebrate traits, and examine the extent to which there have been consistent outcomes across studies. I found that studies were accumulating at a slow and steady rate (zero to two papers published per year for flow-traits and salinity-traits over the last decade). The geographic coverage of studies was strongly biased towards North America and Western Europe. There were broad trends in which traits appear to be associated with flow and salinity, but there were also inconsistencies in some trait responses to effects of these stressors. These inconsistencies may be due to differences in analytical and methodological approaches among studies. I did not find any studies that examined interactions between effects of flow and salinity on macroinvertebrate traits, even though low flows and high salinities often co-occur and may interact, especially in temporary streams. Thus, I hypothesized that interactions between flow and salinity may underlie some inconsistencies in results across the trait literature. As a result, I investigated the simultaneous effects of flow and salinity on macroinvertebrate traits to examine the extent to which their individual and interactive effects were important (Chapter 4). I found that traits showing inconsistent behaviour in the literature were indeed associated with interactions between flow and salinity, which appeared to be driven by the differential responses of taxa with the same trait. Further, flow and salinity variables explained more of the variation in trait space (27.3% in autumn, 36.7% in spring) than taxonomic space (20.9% in autumn, 27.7% in spring), but there was still strong site-specific effects which suggested that biogeographic effects remain, even for trait composition. To compare traits with taxonomic approaches in more detail independent of biogeographic effects, I analysed data separately within four sites with the most complete flow and environmental records from my South Australian dataset (Chapter 5). I evaluated two perennial and two intermittent flow regimes. I predicted that trait composition would be more similar between seasons than taxa, because there is more turnover with taxa over time than with traits. I also predicted that for both taxonomic and trait composition, intermittent streams would be less similar between seasons than perennial streams because intermittent streams are more likely to lose species by chance during dry periods. These expectations were generally supported by the data. Moreover, trait structure was better explained by flow and salinity than taxonomic structure, although the differences were heterogeneous, ranging from 2% to 40% of additional variation explained. Differences in variation explained were associated with the number of taxa and their pattern of dominance sharing the same trait, and not associated with stream flow regime. This thesis compared and contrasted taxonomic and trait based approaches to evaluating stream ecosystems, with a focus on the effects of flow and salinity. I showed that flow and salinity were important predictors of aquatic macroinvertebrate taxonomic and trait composition in South Australia, and that the interactive effect of these variables was also important, explaining some inconsistences among studies in the literature. Consistent with theory, greater variation was predicted for trait-space than taxonomic. However, in contrast to theory, I did not find that trait-based approaches overcame biogeographic effects. Phylogenetic relatedness among traits and differences in methodological approaches used to relate traits to environmental conditions pose challenges in trait-based analysis, and need to be addressed in future studies. To foster the operative use of species traits for stream bioassessment, there is the need for comprehensive regional or national trait databases resolved at species level, and standardized in terms of trait classification and nomenclature.

History

Publication status

  • Unpublished

Rights statement

Copyright 2017 the author Chapter 2 appears to be the equivalent of a post-print version of an article published as: Botwe, P. K., Barmuta, L. A., Magierowski, R., McEvoy, P., Goonan, P., Carver, S., 2015. Temporal patterns and environmental correlates of macroinvertebrate communities in temporary streams. Plos One, 10(11), e0142370. Copyright 2015 the authors. Licensed under Creative Commons Attribution 4.0 International (CC BY 4.0) https://creativecommons.org/licenses/by/4.0/ Chapter 4 appears to be the equivalent of a post-print version of an article published as: Botwe, P. K., Carver, S., Magierowski, R., McEvoy, P., Goonan, P., Madden, C., Barmuta, L. A. 2018. Effects of salinity and flow interactions on macroinvertebrate traits in temporary streams, 2018. Effects of salinity and flow interactions on macroinvertebrate traits in temporary streams, Ecological indicators, 89, 74-83

Repository Status

  • Open

Usage metrics

    Thesis collection

    Categories

    No categories selected

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC