Stephen J. Faggotter

Multiple factors determine the effect of anthropogenic barriers to connectivity on riverine fish

Habitat fragmentation is a key anthropogenic factor in biodiversity decline, particularly in aquatic ecosystems. We predicted that differences in fish assemblage composition due to the impact of fragmentation would most strongly affect migratory species, and these effects would be dependent on the interaction between the characteristics of each barrier and the antecedent flow conditions that determine temporal variation in connectivity. These hypotheses were applied to a coastal river network in eastern Australia that is fragmented by multiple weirs and dams, including some with passage facilities. How these facilities interact with flow to mediate hydrological connectivity and hence patterns of community structure is unknown. Five distinct assemblages were identified that were associated with different combinations of environmental factors and barrier characteristics (spatial arrangement, passability), and key differences were due to variation in migration traits. Two spatially distinct assemblages were associated with fragmentation by two impassable barriers. However, the migration traits that accompanied these community changes were inconsistent between these groups, and likely reflected effects of barriers near the estuary and in the middle of the stream network on diadromous and freshwater-migratory species, respectively. Two assemblage groups in the vicinity of passable weirs varied temporally as a function of hydrology and the seasonal upstream movement of juvenile diadromous species. The effect of habitat loss in conjunction with fragmentation was evident, with a further assemblage group occurring in reaches where riparian vegetation and instream habitat have been altered by poor management of agriculture. This study indicates that the impact of habitat fragmentation in rivers depends on the interaction of the migration characteristics of biota, temporal variation in hydrology which mediates connectivity, and the location of anthropogenic barriers. Conservation policies aimed at minimizing human impacts on aquatic biodiversity need to jointly account for the separate impacts of habitat fragmentation and habitat loss.

Factors controlling primary productivity in a wet–dry tropical river

Algal production in rivers fuels foodwebs, and factors controlling this production ultimately affect food availability. Conversely, excessive algal production can have negative effects on higher trophic levels. The present study examined permanent waterholes in a disconnected wet–dry tropical river to determine the controls on algal production. Primary production in this river system was high compared with arid-zone and perennially flowing tropical rivers. Phytoplankton biomass increased over the dry season but this appeared to be because waterhole volume decreased, due to evaporation. Nitrogen (N) was the key limiting nutrient for phytoplankton, with rapid N turnover times, depletion of particulate 15N-nitrogen reflecting increasing N fixationover the dry season, and N stimulation in phytoplankton bioassays. The waterholes were shallow, providing sufficient light for accumulation of benthic algal biomass. Exclosure experiments were also conducted to determine the impact of top–down control on benthic algal biomass, with no evidence that exclusion of fish and crustaceans increased benthic algal biomass. The shallow off-channel waterhole in our study had substantially higher concentrations of nutrients and chlorophyll a than did the on-channel waterholes. This suggests that future anthropogenic changes, such as increased water extraction and increased nutrient inputs, could make the waterholes more vulnerable to deteriorating water quality, such as e.g. algal blooms, low concentrations of dissolved oxygen.

Differences in nitrate and phosphorus export between wooded and grassed riparian zones from farmland to receiving waterways under varying rainfall conditions

Agricultural activities in catchments can cause excessive nutrient loads in waterways. Catchment nitrogen (N) and phosphorus (P) flows may be intercepted and assimilated by riparian vegetation. While prior studies suggest that woody vegetation is preferable for reducing P loads, the question remains: is woody vegetation or grass cover more effective at reducing catchment N and P exports to waterways. To address this we investigated the relative importance of vegetation type, hydrologic and soil microbial processes on N and P losses from soil to a stream. The study involved the analysis of data from two soil microcosm experiments, and a field case study. We found P leaching loss from riparian zones depended significantly on vegetation type (woody vs. grass cover), with lower P exported from wooded riparian zones, irrespective of the scale of rainfall. For N leaching losses, the scale of rainfall had an effect. During high rainfall, vegetation type had a major effect on N leaching loss, with lower N exported from grassed verses wooded riparian zones. However, under low rainfall conditions, soil type and soil C and N stores, potential indicators of soil microbial activity, rather than vegetation cover, affected N leaching. It is hypothesized that soil microbes were reducing N removal under these conditions. We reason that nitrifiers may have played an important role in soil N cycling, as increased soil ammonium had a strong positive effect on nitrate leaching loads, mediated through soil nitrate stores. Whereas, N immobilization, via incorporation into microbial biomass, and denitrification processes appeared to be limited by C availability, with increased C associated with reduced N leaching. Overall, this study identified that N leaching losses from riparian zones appeared to be affected by two different processes, vegetative uptake and soil microbial processes, the relative importance of which was driven by hydrological conditions.

Biotic and abiotic controls on nitrogen leaching losses into waterways during successive bovine urine application to soil

Cattle waste products high in nitrogen (N) that enter waterways via rainfall runoff can contribute to aquatic ecosystem health deterioration. It is well established that N leaching from this source can be reduced by plant assimilation, e.g. pasture grass. Additionally, N leaching can be reduced when there is sufficient carbon (C) in the soil such as plant litterfall to stimulate microbial processes, i.e. denitrification, which off-gas N from the soil profile. However, the relative importance of these two processes is not well understood. A soil microcosm experiment was conducted to determine the role of biotic processes, pasture grass and microbial activity, and abiotic processes such as soil sorption, in reducing N leaching loss, during successive additions of bovine urine. Pasture grass was the most effective soil cover in reducing N leaching losses, which leached 70% less N compared to exposed soil. Successive application of urine to the soil resulted in N accumulation, after which there was a breaking point indicated by high N leaching losses. This is likely to be due to the low C:N ratio within the soil profiles treated with urine (molar ratio 8:1) compared to water treated soils (30:1). In this experiment we examined the role of C addition in reducing N losses and showed that the addition of glucose can temporarily reduce N leaching. Overall, our results demonstrated that plant uptake of N was a more important process in preventing N leaching than microbial processes.

The effect of riparian restoration on channel complexity and soil nutrients

Restoration of riparian vegetation may reduce nutrient and sediment contamination of waterways while potentially enhancing stream channel complexity. Accordingly, the present study used a paired-site approach to investigate the effects of mature regrowth riparian vegetation on river channel morphology and soil nutrients (i.e. nitrogen and phosphorus), comparing four sites of degraded (pasture) and reforested reaches. A revised rapid assessment of riparian condition (RARC) was used to validate the site pairings. Riparian soil nutrient and elemental geochemistry were compared between paired sites, along with two parameters of channel width complexity and two for channel slope complexity. The RARC analysis confirmed the validity of the paired site design. The elemental geochemistry results indicated that underlying geology may affect the paired site analyses. Reaches with mature regrowth vegetation had greater channel width complexity but no difference in their riverbed slope complexity. In addition, degraded reaches had higher soil nutrient (i.e. nitrogen and phosphorus) concentrations, potentially indicative of the greater nutrient retention of pasture grass sites compared with mature regrowth forested reaches with less ground cover. Overall, the present study indicates that restoring mature regrowth riparian vegetation may increase river channel width complexity, although it may require canopy management to optimise the nutrient retention potential necessary to maximise the effect of riparian restoration strategies on freshwater environments.