Productivity declines threaten East African soda lakes and the iconic Lesser Flamingo.

Soda lakes are some of the most productive aquatic ecosystems.1 Their alkaline-saline waters sustain unique phytoplankton communities2,3 and provide vital habitats for highly specialized biodiversity including invertebrates, endemic fish species, and Lesser Flamingos (Phoeniconaias minor).1,4 More than three-quarters of Lesser Flamingos inhabit the soda lakes of East Africa5; however, populations are in decline.6 Declines could be attributed to their highly specialized diet of cyanobacteria7 and dependence on a network of soda lake feeding habitats that are highly sensitive to climate fluctuations and catchment degradation.8,9,10,11,12 However, changing habitat availability has not been assessed due to a lack of in situ water quality and hydrology data and the irregular monitoring of these waterbodies.13 Here, we combine satellite Earth observations and Lesser Flamingo abundance observations to quantify spatial and temporal trends in productivity and ecosystem health over multiple decades at 22 soda lakes across East Africa. We found that Lesser Flamingo distributions are best explained by phytoplankton biomass, an indicator of food availability. However, timeseries analyses revealed significant declines in phytoplankton biomass from 1999 to 2022, most likely driven by substantial rises in lake water levels. Declining productivity has reduced the availability of healthy soda lake ecosystems, most notably in equatorial Kenya and northern Tanzania. Our results highlight the increasing vulnerability of Lesser Flamingos and other soda lake biodiversity in East Africa, particularly with increased rainfall predicted under climate change.14,15,16 Without improved lake monitoring and catchment management practices, soda lake ecosystems could be pushed beyond their environmental tolerances. VIDEO ABSTRACT.

Low-Cost Approach to an Instream Water Depth Sensor Construction Using Differential Pressure Sensors and Arduino Microcontrollers.

Accurate hydrological data with high spatial resolution is important for flood risk and water resource management, particularly under the context of climate change. The cost of monitoring networks, as well as the characteristics of the hydrological environment itself, can be a barrier to meeting these data requirements, however. This study covers the design and testing of a low-cost, "build-it-yourself", instream water depth sensor providing an assessment of its potential in future hydrological monitoring projects. The low-cost sensor was built using an Arduino microcontroller, a differential pressure sensor and a thermistor, a real-time clock, and an SD card module. The low-cost logger was deployed in tandem with a factory-calibrated Solinst®LevelLogger® 5 Junior for 6 months in the River Wissey, UK. We found the mean absolute error of the Arduino-based logger relative to the commercial setup to be ±0.69 cm for water depth and ±0.415 °C for water temperature. Economically, the Arduino-based logger offers an advantage, costing a total of £133.35 (USD 168.26 at time of publication) comparative to the industrial comparison's cost of £408 (USD 514.83 at time of publication). This study concludes that the low cost of the Arduino-based logger gives a strong advantage to its incorporation in hydrological data collection, if the trade-offs (i.e., time investment and accuracy) are considered acceptable and appropriate for a project.

DNA barcoding indicates multiple invasions of the freshwater snail Melanoides tuberculata sensu lato in Florida.

Melanoides tuberculata sensu lato (Thiaridae) are polymorphic female-clonal snails of Asian and African origins that have invaded freshwaters worldwide, including those in Florida. Although the snails have been documented in Florida for at least 70 years, no studies have investigated whether the observed distribution is due to a single introduction or multiple independent invasions. Here, cytochrome oxidase I was used to measure genetic diversity within and among sites in Florida and compare genetic diversity between Florida and other regions of the world. We also examined the relationship between shell morphology and haplotype diversity to determine if shell morphs can serve as a proxy for haplotypes. In total, we recovered 8 haplotypes randomly distributed across populations in Florida. Phylogenetic reconstruction supported the hypothesis of multiple invasions by diverse representatives of the M. tuberculata species complex. In contrast, shell morphology was not found to be a useful phylogeographic character, with divergent haplotypes represented by similar shell forms. These results suggest that the observed invasion patterns in Florida are best explained by serial introductions, and that shell morphology cannot be used to predict haplotypes or reconstruct invasion history of Melanoides tuberculata s.l. and that extensive taxonomic revisions are needed to investigate invasion dynamics.

A function-based typology for Earth's ecosystems.

As the United Nations develops a post-2020 global biodiversity framework for the Convention on Biological Diversity, attention is focusing on how new goals and targets for ecosystem conservation might serve its vision of 'living in harmony with nature'1,2. Advancing dual imperatives to conserve biodiversity and sustain ecosystem services requires reliable and resilient generalizations and predictions about ecosystem responses to environmental change and management3. Ecosystems vary in their biota4, service provision5 and relative exposure to risks6, yet there is no globally consistent classification of ecosystems that reflects functional responses to change and management. This hampers progress on developing conservation targets and sustainability goals. Here we present the International Union for Conservation of Nature (IUCN) Global Ecosystem Typology, a conceptually robust, scalable, spatially explicit approach for generalizations and predictions about functions, biota, risks and management remedies across the entire biosphere. The outcome of a major cross-disciplinary collaboration, this novel framework places all of Earth's ecosystems into a unifying theoretical context to guide the transformation of ecosystem policy and management from global to local scales. This new information infrastructure will support knowledge transfer for ecosystem-specific management and restoration, globally standardized ecosystem risk assessments, natural capital accounting and progress on the post-2020 global biodiversity framework.

Historical Analysis Exposes Catastrophic Seagrass Loss for the United Kingdom.

The spatial extent of seagrass is poorly mapped, and knowledge of historical loss is limited. Here, we collated empirical and qualitative data using systematic review methods to provide unique analysis on seagrass occurrence and loss in the United Kingdom. We document 8,493 ha of recently mapped seagrass in the United Kingdom since 1998. This equates to an estimated 0.9 Mt of carbon, which, in the current carbon market represents about £22 million. Using simple models to estimate seagrass declines triangulated against habitat suitability models, we provide evidence of catastrophic seagrass loss; at least 44% of United Kingdom's seagrasses have been lost since 1936, 39% since the 1980's. However, losses over longer time spans may be as high as 92%. Based on these estimates, historical seagrass meadows could have stored 11.5 Mt of carbon and supported approximately 400 million fish. Our results demonstrate the vast scale of losses and highlight the opportunities to restore seagrass to support a range of ecosystems services.

Influence of urban river restoration on nitrogen dynamics at the sediment-water interface.

River restoration projects focused on altering flow regimes through use of in-channel structures can facilitate ecosystem services, such as promoting nitrogen (N) storage to reduce eutrophication. In this study we use small flux chambers to examine ammonium (NH4+) and nitrate (NO3-) cycling across the sediment-water interface. Paired restored and unrestored study sites in 5 urban tributaries of the River Thames in Greater London were used to examine N dynamics following physical disturbances (0-3 min exposures) and subsequent biogeochemical activity (3-10 min exposures). Average ambient NH4+ concentrations were significantly different amongst all sites and ranged from 28.0 to 731.7 µg L-1, with the highest concentrations measured at restored sites. Average NO3- concentrations ranged from 9.6 to 26.4 mg L-1, but did not significantly differ between restored and unrestored sites. Average NH4+ fluxes at restored sites ranged from -8.9 to 5.0 µg N m-2 sec-1, however restoration did not significantly influence NH4+ uptake or regeneration (i.e., a measure of release to surface water) between 0-3 minutes and 3-10 minutes. Further, average NO3- fluxes amongst sites responded significantly between 0-3 minutes ranging from -33.6 to 97.7 µg N m-2 sec-1. Neither NH4+ nor NO3- fluxes correlated to sediment chlorophyll-a, total organic matter, or grain size. We attributed variations in overall N fluxes to N-specific sediment storage capacity, biogeochemical transformations, potential legacy effects associated with urban pollution, and variations in river-specific restoration actions.

Effects of Parasite Infection and Host Body Size on Habitat Associations of Invasive Aquatic Snails: Implications for Environmental Monitoring.

The Comal River, a spring-fed system in central Texas, was invaded in the 1960s by two Asian aquatic snails (Thiaridae: red-rimmed melania Melanoides tuberculata and quilted melania Tarebia granifera) and subsequently by three of their trematode parasites (the avian eye-fluke Philophthalmus gralli in the 1960s; the gill trematode Centrocestus formosanus in the 1990s; and the intestinal fluke Haplorchis pumilio in the 2000s). Previous snail collections (2001-2002) established that habitat conditions significantly affect the distribution of both snail species. However, the effects of snail size (known to influence infection prevalence) and habitat conditions (known to influence snail size) on trematode infection patterns in this system were not evaluated. In a re-evaluation of this data set, logistic regression analyses with individual snails showed that for both M. tuberculata and T. granifera populations, large snails were more likely to be infected than small snails, and habitat conditions were significantly related to infection in T. granifera. However, only snail size was significant in explaining the probability of infection in M. tuberculata. This result was confirmed by linear regression models, which showed that both infected and noninfected M. tuberculata used similar habitats, as large individuals in both infection categories were found in patches dominated by fine substrates and high levels of aquatic vegetation and detritus. For the large size-class of T. granifera, noninfected individuals were found primarily in habitats with silt/sand substrates and high vegetation and detritus cover, while infected individuals occurred among all available habitats. Using these results, we suggest that targeted sampling of large individuals of M. tuberculata in habitats with high detritus and vegetation and large individuals of T. granifera in any habitat can be used to efficiently ascertain parasite "hot spots" and to evaluate changes in parasite prevalence or detect the invasion of new parasites in these thiarid snails.

Variability of UK seagrass sediment carbon: Implications for blue carbon estimates and marine conservation management.

Seagrass meadows provide a multitude of ecosystem services, including a capacity to sequester carbon dioxide (CO2) within their sediments. Seagrass research in the UK is lacking and there is no published data on sediment carbon (C) within UK seagrass meadows. We sampled 13 Zostera marina meadows along the southwest coast of the UK to assess the variability in their sedimentary organic carbon (OC) stocks. The study sites were considered representative of sub-tidal Z. marina meadows in the UK, spanning a gradient of sheltered to exposed sites, varying in formation, size and density, but found along the same latitudinal gradient. OC stocks (Cstocks) integrated across 100cm depth profiles were similar among all sites (98.01 ± 2.15 to 140.24 ± 10.27 Mg C ha-1), apart from at Drakes Island, which recorded an unusually high Cstock (380.07 ± 17.51 Mg C ha-1) compared to the rest of the region. The total standing stock of C in the top 100cm of the surveyed seagrass meadows was 66,337 t C, or the equivalent of 10,512 individual UK people's CO2 emissions per year. This figure is particularly significant relative to the seagrass area, which totalled 549.79 ha. Using estimates of seagrass cover throughout the UK and recent UK C trading values we approximate that the monetary value of the UK's seagrass standing C stock is between £2.6 million and £5.3 million. The C stock of the UK's seagrass meadows represent one of the largest documented C stocks within Europe and are, therefore, of important ecosystem service value. The research raises questions concerning the reliability of using global or regional data as a proxy for local seagrass C stock estimates and adds to a growing body of literature that is looking to understand the mechanisms of seagrass C storage. When taken with the fact that seagrass meadows are an important habitat for commercially important and endangered species in the UK, along with their declining health and cover, this research supports the need for more robust conservation strategies for UK seagrass habitats.

Eco-hydromorphic Classification for Understanding Stream Macroinvertebrate Biodiversity in Brunei Darussalam, Northern Borneo.

Kate Baker, Michael A. Chadwick, and Zohrah Haji Sulaiman (2016) Linking ecology with river geomor- phology and hydrology (geomorphic and hydraulic template) plays an important role in the study of macroinver- tebrate biodiversity. This understanding and knowledge is crucial in implementing sensible conservation management for ecosystem health monitoring. However, most macroinvertebrate research has been conducted in temperate ecosystems. This study examines the eco-hydrogeomorphology and macroinvertebrate biodiversity of two remote tropical streams in northern Borneo (Bukit Pagon catchment, Brunei Darussalam's highest mountain - 1850 m) using temperate classification models, more specifically, biotopes. Fast flowing biotopes were defined as bedrock runs and cobble riffles whilst the slow flowing biotopes were deposition pools. Macroinvertebrate size structure associated with biotopes, which can influence overall ecological processes, was also investigated. Forty-three macroinvertebrate taxa were recorded during the study; biodiversity was similar between the study streams. There were differences among biotopes with the lowest diversity occurring in fast flowing biotopes (p = 0.05*). Community structure also varied among the biotopes. Cluster analysis of macroinvertebrate abundance revealed an 0.8 dissimilarity between the fast and slow biotopes. Several taxa were found in multiple biotopes, which is likely linked to the occurrence of moss and leaf litter. Macroinvertebrate size structure distribution between the fast and slow biotopes was statistically different. Our findings suggest biotopes may be an appropriate scale to investigate macroinvertebrate biodiversity in tropical streams. Specifically, we found that biotopes had different macroinvertebrate communities and richness. Further research is required to understand the importance of habitat parameters that are not directly related to flow velocities such as moss. These habitats are important as places of refuge, allowing colonisation that would otherwise be inhospitable during flood periods.

A primary FIsh Gill Cell System (FIGCS) for environmental monitoring of river waters.

Studies were conducted to assess the feasibility of a primary FIsh Gill Cell culture system (FIGCS) for both laboratory and field based environmental monitoring of rivers known to be affected by metal contamination. FIGCS were exposed in the laboratory and in the field to water from the River Hayle, a metal-contaminated system in Cornwall, United Kingdom. Water chemistry, including transition metal concentrations, changes in transepithelial electrical resistance (TEER), cell viability and the expression of metal responsive genes, metallothionein A and B were measured. FIGCS tolerated river water in the laboratory showing no loss in TEER or cell viability following 24h exposure. The cells also tolerated transport to the field (∼1000 km and 30 h) and exposure to unfiltered and filtered river water. Metallothionein A and B, a measure of intracellular biologically active metals, expression was induced in the laboratory and field on exposure to water from sites with elevated metal concentrations compared to those sites where metal levels were below water metal Environmental Quality Standards. This demonstrates that FIGCS detects bioreactive metals in river waters on exposure in the laboratory or field and can be used for on-site environmental monitoring as well as investigations into bioavailability and toxicity of contaminant mixtures in natural waters.

Stream ecosystem response to chronic deposition of N and acid at the Bear Brook Watershed, Maine.

The Bear Brook Watershed in Maine (BBWM) is a long-term, paired watershed experiment that addresses the effects of acid and nitrogen (N) deposition on whole watersheds. To examine stream response at BBWM, we synthesized data on organic matter dynamics, including leaf breakdown rates, organic matter inputs and standing stocks, macroinvertebrate secondary production, and nutrient uptake in treated and reference streams at the BBWM. While N concentrations in stream water and leaves have increased, the input, standing stocks, and breakdown rates of leaves, as well as macroinvertebrate production, were not responsive to acid and N deposition. Both chronic and acute increases of N availability have saturated uptake of nitrate in the streams. Recent experimental increases in phosphorus (P) availability enhanced stream capacity to take up nitrate and altered the character of N saturation. These results show how the interactive effects of multiple factors, including environmental flow regime, acidification, and P availability, may constrain stream response to chronic N deposition.

Urbanization affects stream ecosystem function by altering hydrology, chemistry, and biotic richness.

Catchment urbanization can alter physical, chemical, and biological attributes of stream ecosystems. In particular, changes in land use may affect the dynamics of organic matter decomposition, a measure of ecosystem function. We examined leaf-litter decomposition in 18 tributaries of the St. Johns River, Florida, USA. Land use in all 18 catchments ranged from 0% to 93% urban which translated to 0% to 66% total impervious area (TIA). Using a litter-bag technique, we measured mass loss, fungal biomass, and macroinvertebrate biomass for two leaf species (red maple [Acer rubrum] and sweetgum [Liquidambar styraciflua]). Rates of litter mass loss, which ranged from 0.01 to 0.05 per day for red maple and 0.006 to 0.018 per day for sweetgum, increased with impervious catchment area to levels of approximately 30-40% TIA and then decreased as impervious catchment area exceeded 40% TIA. Fungal biomass was also highest in streams draining catchments with intermediate levels of TIA. Macroinvertebrate biomass ranged from 17 to 354 mg/bag for red maple and from 15 to 399 mg/bag for sweetgum. Snail biomass and snail and total invertebrate richness were strongly related to breakdown rates among streams regardless of leaf species. Land-use and physical, chemical, and biological variables were highly intercorrelated. Principal-components analysis was therefore used to reduce the variables into several orthogonal axes. Using stepwise regression, we found that flow regime, snail biomass, snail and total invertebrate richness, and metal and nutrient content (which varied in a nonlinear manner with impervious surface area) were likely factors affecting litter breakdown rates in these streams.