Abstract
Feeding a growing human population while minimising pressures on land, reducing impacts to freshwater ecosystems, and halting the rapid freshwater biodiversity decline is an enormous challenge. Meeting the demand of a growing population was previously achieved through unsustainable agricultural land-use practices, subsequently increasing in-stream stressor loads. Of those stressors, deposited fine sediment has been unequivocally proven to have widespread negative impacts on running-water ecosystems. Our understanding of fine sediment impacts at a range of spatial scales is substantial, but knowledge of temporal variability in fine sediment remains worryingly incomplete. My overall PhD objective was to quantify temporal variability of in-stream deposited fine sediment levels and evaluate how this variability affects stream ecosystems. The aim was to provide evidence that would aid in the development of biomonitoring programs capable of more efficiently identifying at-risk sites, allow for targeted mitigation efforts, and track the success of restoration efforts.
To address knowledge gaps in the literature, complementary long-term (Chapter 2, published as Davis et al. 2022) and short-term (Chapter 3) field surveys were employed to identify the temporal dynamics of deposited fine sediment and macroinvertebrate responses in the Southland region of New Zealand. A 5-year dataset from 46 river sites was used to investigate the consistency of the relationship between deposited fine sediment (measured as suspendable inorganic sediment, SIS g m-2) and stream macroinvertebrates across three categories of agricultural land-use intensity (low, medium, and high; Chapter 2). Additionally, I compared the performance of four widely-used invertebrate stream health metrics against their recently developed sediment-specific counterparts. Building upon this 5-year dataset, 15 river sites were sampled monthly between November 2019 and October 2020 (Chapter 3) to assess the short-term dynamics of fine sediment and invertebrate community responses. Besides evaluating the performance of sediment-specific invertebrate metrics, a comparison of two common physical measures of fine sediment (Quorer resuspension method and in-stream visual assessment) was made. To further illuminate the impacts of short-term fluctuations in fine sediment observed in Chapter 3, I conducted an 8-week streamside experiment in 128 flow-through mesocosms to investigate the individual and combined impacts of realistic sediment pulses (0-3 pulses) and continuous nutrient enrichment on freshwater ecosystem responses. Five hundred and twelve invertebrate drift and insect emergence samples each were collected throughout the experiment (on four occasions for 48 hours, three of which were after a sediment pulse), plus 128 benthic invertebrate samples collected after 42 days of sediment and nutrient manipulations. Further, organic matter decomposition rates and algal biomass accrual were assessed in all mesocosms. Taken together, my field surveys and manipulative experiment provide a unique perspective into the temporal dynamics of fine sediment and associated freshwater ecosystems responses not yet seen in the literature.
In the 5-year study (Chapter 2), invertebrate stream health metrics were related more frequently to increasing fine sediment levels (in linear and non-linear regressions) at river sites within catchments of high, and even more so, medium land-use intensity. These findings implied greater variability in fine sediment and invertebrate communities at such sites can be expected; therefore, greater biomonitoring effort should be directed towards these sites compared to those within low-intensity catchments. Analysis of monthly fine sediment dynamics and invertebrate community responses (Chapter 3) also identified considerable variation in fine sediment levels that became much more pronounced with increasing catchment land-use intensity. Further, all 15 sites experienced a prolonged period of relatively stable flow from mid-February to June/July, which coincided with the largest short-term increase in SIS at 14 sites. Thus, management actions that are strategically targeted during periods where the greatest risk for sedimentation occurs would likely provide the largest biological gains. Overall, in both field surveys sediment-specific invertebrate metrics performed better than their commonly-used counterparts (based on relationship strength with invertebrate stream health metrics). Further, in the monthly survey (Chapter 3), the Quorer resuspension method (SIS) unequivocally outperformed visual estimates of fine sediment by showing more, and consistently stronger, biologically meaningful relationships (r2≥0.1) with invertebrate response variables. In the mesocosm experiment (Chapter 4), fine sediment frequently (and mostly negatively) influenced all invertebrate responses (benthos, drift, emergence) and altered both the accrual of algal biomass and organic matter decomposition rates. Some nutrient main effects were present but very few interactions with sediment occurred, indicating sediment was the dominant stressor. Drift and benthos invertebrate responses at the community and taxon levels suggested the existence of two influential thresholds (one and three pulses) in response to repeated sediment addition. One pulse caused the loss of most sensitive invertebrates, but after three pulses sediment-sensitive invertebrate taxa disappeared almost entirely. When occurring at a larger spatial scale, this could eliminate valuable source populations, thus threatening successful invertebrate recolonisation following habitat improvements.
Overall, my findings stress the importance of establishing complimentary short- and long-term regional biomonitoring programs focusing on both fine sediment and macroinvertebrates. Such an approach would facilitate management efforts that efficiently prioritised at risk and important river sites and allowed for detailed tracking of biological outcomes that would validate which aspects of restoration efforts are successful. The final section of my thesis (Chapter 5, General Discussion) highlights how successful collaborations between regional authorities and local catchment communities could provide an opportunity to produce complimentary monthly and annual biomonitoring regimes.