Search

The quality of freshwater input from tributaries of the Western Mississippi Sound (WMSS) impacts the quality of coastal water. Hydrological and hydrodynamic models can be coupled to assess the impact of freshwater inflow from coastal watersheds. This study aims to compare the performance of a hydrodynamic model and a hydrological–hydrodynamic coupled model in detecting the effect of freshwater inflow from the coastal watersheds of the state of Mississippi into the WMSS. A hydrological model, the Soil and Water Assessment Tool (SWAT), and a hydrodynamic model, the visual Environmental Fluid Dynamics Code (vEFDC), were coupled to evaluate the difference between the hydrodynamical modelling approach, which employs an area-weighted approach to define flow and nutrient concentrations, and the more recent coupling model approach, which uses a hydrological model to determine the flow and nutrient load of the model. Furthermore, a nutrient load sensitivity analysis of the effect of freshwater inflow on water quality in the WMSS was conducted in addition to assessing the repercussions of tropical depressions. Hydrological assessments of the major tributaries watersheds of Saint Louis Bay (SLB) at the WMSS were performed using the SWAT model. After calibration/validation of the SWAT model, the streamflow output from the SWAT was incorporated into the vEFDC model. Finally, hydrodynamic simulation of the SWAT-vEFDC model was conducted, and water quality output was compared at different SLB locations. The salinity, dissolved oxygen, total nitrogen (TN), and total phosphorus (TP) were assessed by comparing the vEFDC and SWAT-vEFDC outputs. The results indicated that hydrological input from the SWAT alters the flow and nutrient concentration results as compared to an area-weighted approach. In addition, a major impact on the concentration of TN and TP occurred at the location where the freshwater flows into SLB. This impact diminishes further away from the point of freshwater inflow. Moreover, a 25% nutrient load variation did not demonstrate a difference in water quality at the WMSS besides TN and TP in a post-tropical depression scenario. Therefore, the SWAT-vEFDC coupled approach provided insights into evaluation of the area-weighted method, and of hydrological model output to the hydrodynamical model, the effect of freshwater inflow into coastal waters, and nutrient sensitivity analysis, which are important for integrated coastal ecosystems management. [ABSTRACT FROM AUTHOR]

Transformer neural networks (TNNs) have caused a paradigm shift in deep learning domains like natural language processing, gathering immense interest due to their versatility in other fields such as time series forecasting (TSF). Most current TSF applications of TNNs use only historic observations to predict future events, ignoring information available in weather forecasts to inform better predictions, and with little attention given to the interpretability of the model's use of explanatory inputs. This work explores the potential for TNNs to perform TSF across multiple environmental variables (streamflow, stage, water temperature, and salinity) in two ecologically important regions: the Peace River watershed (Florida) and the northern Gulf of Mexico (Louisiana). The TNN was tested and its prediction uncertainty quantified for each response variable from one‐to fourteen‐day‐ahead forecasts using past observations and spatially distributed weather forecasts. A sensitivity analysis (SA) was performed on the trained TNNs' attention weights to identify the relative influence of each input variable on each response variable across prediction windows. Overall model performance ranged from good to very good (0.78 < NSE < 0.99 for all variables and forecast horizons). Through the SA, we found that the TNN was able to learn the physical patterns behind the data, adapt the use of input variables to each forecast, and increasingly use weather forecast information as prediction windows increased. The TNN's excellent performance and flexibility, along with the intuitive interpretability highlighting the logic behind the models' forecasting decision‐making process, provide evidence for the applicability of this architecture to other TSF variables and locations. Key Points: Transformer neural network (TNN) for mid‐range time series forecasting (TSF) across multiple environmental variables using past conditions and weather forecastTNN performance showed very good predictive ability across diverse environmental time series up to 14 days in the futureSensitivity analysis allowed for interpretation of the dynamic role of different model inputs across variables and prediction windows [ABSTRACT FROM AUTHOR]

ABSTRACT Oysters provide a wide variety of ecosystem services including furnishing habitat, supporting an economic industry, and enhancing water quality. To identify suitable areas for oyster restoration and aquaculture, areas [...]

The northern Gulf of Mexico is facing high rates of wetland loss due to subsidence and sea level–rise, which has encouraged the application of various wetland restoration techniques. Marsh terracing is a restoration technique that has been implemented since the early 1990 s in Texas and Louisiana, yet few studies have been conducted to evaluate its effectiveness. Marsh terraces are segmented berms of soil built in coastal ponds that were once vegetated marshes. Marsh terracing is hypothesized to dissipate wind waves, encourage marsh expansion, and reduce shoreline erosion. This study (1) assessed the effectiveness of the most common terrace shapes (linear, chevron, and square) and spacing (100, 110, and 120 m) at reducing significant wave height (Hs), (2) assessed the effectiveness of alternative terrace designs for reducing Hs during different wind conditions, and 3) estimated the construction costs of alternative terrace designs. The Simulating WAves Nearshore (SWAN) model was used to simulate wind–driven waves in ponds with real and hypothetical terrace designs. Results revealed that: (1) The chevron shape provided the greatest reduction in Hs during all wind conditions, reducing Hs by up to 54%. (2) Hs reduction was not affected by terrace spacings. (3) Based on wave attenuation, the chevron design with a 120 m terrace spacing provided the optimal outcome with an estimated construction cost/ha of $6332 in a 250,000 m2 site compared to the terrace shapes and spacings evaluated in this study. This study will help coastal managers design marsh terraces to address wetland erosion in the Gulf of Mexico and other coastal areas facing similar environmental problems. [Display omitted] • This study investigates the optimal design for marsh terraces at reducing waves. • Chevron terraces provided the greatest reduction in significant wave height (Hs). • The chevron terrace shape reduced Hs up to 54%. • Chevron terraces with a 120 m terrace spacing provided the most optimal outcome. • The estimated cost of a chevron terrace with 120 m spacing was $6332/ha. [ABSTRACT FROM AUTHOR]