Effects of salinization on tropical freshwater wetland primary producers and aquatic invertebrates

Sea level rise is expected to transform coastal aquatic ecosystems worldwide. The freshwater wetlands of tropical northern Australia are among the most biodiverse and productive ecosystems on the continent, but owing to high regional rates of sea level rise coupled with low-lying land and large tides, they are increasingly affected by saltwater intrusion. The propagule bank, including seeds of aquatic primary producers and eggs of aquatic invertebrates, stored in wetland sediment, is vital for the establishment of ecological communities, and ultimately for primary and secondary production. This experimental study examined the impact of increasing salinity on the emergence of primary producers and invertebrates from sediment cores collected from tropical freshwater wetlands between Darwin and Kakadu National Park in Northern Australia. Sediment cores (n=216) were placed in microcosms and inundated with one of four salinity treatments: 0 ppt (freshwater/control); 7 ppt; 16 ppt, and 35 ppt (seawater) and decanted on approximately days 10, 20, and 90. In comparison to freshwater controls, median chlorophyll a of phytoplankton and benthic algae declined sharply with increasing salinity, and in the highest salinity treatment, were reduced by 93% and 60% respectively. Macrophyte biomass, aquatic invertebrate density, and richness were reduced to near zero in all salinity treatments. Increasing salinity decreases the abundance of primary producers and aquatic invertebrates that emerge from sediment cores in seasonally inundated tropical freshwater wetlands. Macrophyte emergence exhibits no tolerance to salinities above freshwater (0 ppt), and aquatic invertebrate emergence declines rapidly in both abundance and diversity as salinity increases. Our results suggest that freshwater aquatic primary producers and aquatic invertebrates will be severely impacted by sea level rise and saltwater intrusion. Reduced freshwater aquatic primary production and aquatic invertebrates may have important implications for food webs, and other social and biodiversity values of tropical freshwater wetlands.

Methods
Nine wetlands within three coastal floodplains in the Northern Territory, Australia, were sampled in October 2021. Twenty-four sediment cores were collected from each of the nine wetlands. The samples were collected from around the edge of the wetland’s remaining water while avoiding the water’s edge. The cores were collected using an 80 mm PVC pipe and trowel to remove the top 30 mm of sediment. Each core was placed into an individual 1 L plastic container and stored at room temperature at Charles Darwin University and left to dry completely for at least 26 days before the commencement of the experiment. The cores from each wetland were stored at room temperature 25 °C and exposed to 12 hr light/dark cycles of full spectrum double fluorescent lights. The 1 L containers were randomly allocated to one of four salinity microcosm treatment groups: control/0 ppt (freshwater), 5-10 ppt (approx. 7 ppt), 15-20 ppt (approx. 16 ppt), and 30-35 ppt (approx. 35ppt) (sea water), resulting in n= 216 cores broken down into six replicates per treatment per wetland. In addition to the treatments, three empty containers without a sediment core were included for each of the four treatments (totaling 12 containers), to provide a control for potential primary and aquatic invertebrate production arising from the saline bore water used to create treatments.Treatments were created using reverse osmosis freshwater and purified saline bore water which had been modified to represent the composition of seawater. The bore water was modified by removing excess iron and manganese and supplemented with silicate and potassium. Salinity was adjusted to match sea water by addition of KCl and Na₂SiO ₃.

Each of the 1 L containers with a dry sediment core (n= 216) were inundated with the allocated salinity treatment in November 2021 and topped up to maintain a water level above the sediment at all times. The samples were completely decanted on three occasions at approximately days 10, 20 and 90 post-inundation. During each decant, all zooplankton and other aquatic invertebrates were collected using a 125 µm sieve and stored in 50% ethanol for subsequent identification. To measure phytoplankton chlorophyll a concentration a 250 ml water column sample was filtered onto a filter (47 mm, 0.7 µm pore size Whatman binder free glass microfiber filter). Filters were stored in a freezer until analysed with a fluorometer. A sample of the benthic algae (chlorophyll a) was collected during the final decant using a 32 mm circular felt scrubbing pad attached to plywood (Davies and Gee 1993). The felt scrubbing pad was used to scrape algae off a 32 mm (diameter) circle of the benthos (Davies and Gee 1993) and placed into a glass sample jar and stored frozen prior to analysis. Phytoplankton and benthic algal chlorophyll a were measured using the nonacidification technique for acetone-extracted chlorophyll with a Triology®laboratory fluorometer (Turner Designs, Sunnyvale, California) (Welschmeyer 1994). All aquatic macrophytes were collected on the final decant and were qualitatively separated based on visual characteristics at the end of the experiment, but since they did not flower or reach mature adult stages it was not possible to identify species. The macrophytes were then combined by sample and dried in an oven at 60° C for 48 hours, and total dry weight biomass (g) was measured. Zooplankton and other invertebrates were counted and identified to family or genus level, except ostracods which where were sorted into two morphological groups (Ostracoda 1 and Ostracoda 2). This was done, due to the lack of a key to juvenile ostracods, to prevent the mixing of juveniles and adults of the same taxa into different groups.