Delaware Research Articles

Predictive Understanding of Stream Salinization in a Developed Watershed Using Machine Learning.

Stream salinization is a global issue, yet few models can provide reliable salinity estimates for unmonitored locations at the time scales required for ecological exposure assessments. Machine learning approaches are presented that use spatially limited high-frequency monitoring and spatially distributed discrete samples to estimate the daily stream-specific conductance across a watershed. We compare the predictive performance of space- and time-unaware Random Forest models and space- and time-aware Recurrent Graph Convolution Neural Network models (KGE: 0.67 and 0.64, respectively) and use explainable artificial intelligence methods to interpret model predictions and understand salinization drivers. These models are applied to the Delaware River Basin, a developed watershed with diverse land uses that experiences anthropogenic salinization from winter deicer applications. These models capture seasonality for the winter first flush of deicers, and the streams with elevated predictions correspond well with indicators of deicer application. This result suggests that these models can be used to identify potential salinity-impaired streams for winter best management practices. Daily salinity predictions are driven primarily by land cover (urbanization) trends that may represent anthropogenic salinization processes and weather at time scales up to three months. Such modeling approaches are likely transferable to other watersheds and can be applied to further understand salinization risks and drivers.

Evaluating a process‐guided deep learning approach for predicting dissolved oxygen in streams

AbstractDissolved oxygen (DO) is a critical water quality constituent that governs habitat suitability for aquatic biota, biogeochemical reactions and solubility of metals in streams. Recently introduced high‐frequency sensors have increased our ability to measure DO, but we still lack the capacity to understand and predict DO concentrations at high spatial resolutions or in unmonitored locations. Machine learning (ML) has been a commonly used approach for modelling DO, however, conventional ML models have no representation of the limnological processes governing DO dynamics. Here we implement and evaluate two process‐guided deep learning (PGDL) approaches for predicting daily minimum, mean and maximum DO concentrations in rivers from the Delaware River Basin, USA. In both cases, a multi‐task approach was taken in which the PGDL models predicted stream metabolism and gas exchange rates in addition to the DO concentrations themselves. Our results showed that for these sites, the PGDL approaches did not improve upon baseline predictions in temporal and spatially similar holdout experiments. One of the approaches did, however, improve predictions when applied to spatially dissimilar sites. Although this particular PGDL approach did not improve predictive accuracy in most cases, our results suggest that process guidance, perhaps a more constrained approach, could benefit a data‐driven DO model.

Optimizing parameter learning and calibration in an integrated hydrological model: Impact of observation length and information

Integrated hydrological modeling is gaining popularity due to its mechanistic representation of the surface and subsurface processes. However, estimating the parameters of such process-based models can be computationally expensive if careful consideration is not given to the length of streamflow observations used during model calibration. Here we evaluate the influence of the calibration period, the role of streamflow information content, and the gage location in parameter learning and calibration of a fully integrated hydrological model, the Advanced Terrestrial Simulator (ATS). We conducted the study at the Upper Neversink River Watershed within the Delaware River Basin, where streamflow observations are available at 11 gages with varying record lengths. We leveraged a recently proposed knowledge-informed deep learning technique for parameter estimation. To assess the impact of observation period and gage location, model parameters were learned on scenarios using different chunks of streamflow observations, including (1) using only one year or consecutive multiple years of streamflow observations at the watershed outlet and (2) using one shared year of observations at each of the sub-catchment gages, with the period from 1991-10-01 through 1999-09-30 as the overall calibration period. Using the estimated parameters, ATS was rerun for each scenario and evaluated on a subsequent period from 1999-10-01 through 2002-09-30. Results show that the basin outlet discharge prediction is mostly improved when using at least four years of observations for parameter estimation. Further, the performance of the calibrated ATS run correlates with the information content of the observed streamflows, suggesting that the information-theoretic metrics could be indicators for selecting the observation period for parameter estimation. Finally, we find that observations from an informative gage can be used in learning parameters to predict the streamflow at a nearby gage, which would potentially lower the computational expense by reducing the watershed domain used in calibration. Our success underscores the potential of using information theory to achieve robust model parameter estimation on a reduced computational budget.

Microplastics in the Delaware River Estuary: Mapping the Distribution And Modeling Hydrodynamic Transport

Microplastics in the Delaware River Estuary: Mapping the Distribution And Modeling Hydrodynamic Transport

Population genetic structure and demographic history reconstruction of introduced flathead catfish (Pylodictis olivaris) in two US Mid-Atlantic rivers.

Population genetic analysis of invasive populations can provide valuable insights into the source of introductions, pathways for expansion, and their demographic histories. Flathead catfish (Pylodictis olivaris) are a prolific invasive species with high fecundity, long-distance dispersal, and piscivorous feeding habits that can lead to declines in native fish populations. In this study, we analyse the genetics of invasive P. olivaris in the Mid-Atlantic region to assess their connectivity and attempt to reconstruct the history of introduced populations. Based on an assessment across 13 microsatellite loci, P. olivaris from the Susquehanna River system (N = 537), Schuylkill River (N = 33), and Delaware River (N = 1) have low genetic diversity (global Hobs = 0.504), although we detected no evidence of substantial inbreeding (FIS = -0.083 to 0.022). P. olivaris from these different river systems were genetically distinct, suggesting separate introductions. However, population structure was much weaker within each river system and exhibited a pattern of high connectivity, with some evidence of isolation by distance. P. olivaris from the Susquehanna and Schuylkill rivers showed evidence for recent genetic bottlenecks, and demographic models were consistent with historical records, which suggest that populations were established by recent founder events consisting of a small number of individuals. Our results show the risk posed by small introductions of P. olivaris, which can spread widely once a population is established, and highlight the importance of prevention and sensitive early detection methods to prevent the spread of P. olivaris in the future.

Pywr-DRB: An open-source Python model for water availability and drought risk assessment in the Delaware River Basin

The Delaware River Basin (DRB) in the Mid-Atlantic region of the United States is an institutionally complex water resources system that provides drinking water for 13.5 million people, plus water for energy, industry, recreation, and ecosystems. This paper introduces Pywr-DRB, an open-source Python model exploring the impacts of reservoir operations, transbasin diversions, and minimum flow targets on water availability and drought risk in the DRB. Pywr-DRB draws on streamflow estimates from emerging data resources, bridging advances in large-scale hydrologic modeling with an improved representation of the basin's evolving water infrastructure and management institutions. Our detailed model diagnostic assessment demonstrates that Pywr-DRB provides substantial improvements over sole use of hydrologic models in capturing the DRB's dynamics. We also explore how water management alters model-derived risk estimates for low flows and water demand shortfalls. Our approach to diagnostic benchmarking and water systems modeling is broadly applicable to other major basins.

Optimisation and improvement of police officers’ special rank types

The article examines the process of reforming the law enforcement system in Ukraine, in particular the creation of the National Police in 2015, and its impact on the rank system of police officers. Special attention is paid to the main changes that occurred as a result of this reform. In particular, such positive aspects as bringing police standards closer to the European level, as well as creating a new police uniform and changing the rank system are highlighted. It is important to note that the reform is aimed at improving the efficiency of law enforcement agencies and their interaction with the public. The article also highlights the problem of the lack of additional ranks and limited career opportunities for junior police officers, which can lead to professional burnout and reduced staff motivation. An analysis of the positive experience of using the extended rank of junior and sergeant ranks in countries such as Canada, Germany, Greece, Spain, as well as the states of Virginia and Delaware in the USA is conducted. The proposed ways of solving the shortcomings in the rank system are aimed at improving the working conditions and motivation of the police personnel. In addition, it is proposed to introduce new ranks for junior and non-commissioned officers of the police in order to stimulate further self-development of professional skills and career growth prospects. In addition, each new rank will have separate duties and functions that will facilitate mutual assistance and cooperation with other police officers. It is also important to emphasise the need to update the design for cadets of higher education institutions with specific learning conditions and lyceums.

A spatial machine learning model developed from noisy data requires multiscale performance evaluation: Predicting depth to bedrock in the Delaware river basin, USA

Spatial machine learning models can be developed from observations with substantial unexplainable variability, sometimes called ‘noise’. Traditional point-scale metrics (e.g., R2) alone can be misleading when evaluating these models. We present a multi-scale performance evaluation (MPE) using two additional scales (distributional and geostatistical). We apply the MPE framework to predictions of depth to bedrock (DTB) in the Delaware River Basin. Geostatistical analysis shows that approximately one third of the DTB variance is at spatial scale smaller than 2 km. Hence, we interpret our point-scale R2 of 0.3 (testing data) to be sufficient for regional-scale modelling. Bias-correction methods improve performance at two of the three MPE scales: point-scale change is negligible, while distributional and geostatistical performance improves. In contrast, bias correction applied to a global DTB model does not improve MPE performance. This work encourages scale-appropriate performance evaluations to enable effective model intercomparison.

Patterns in Fish-Assemblage Structure Over a Decade of Monitoring in the Upper Delaware River Watershed

Patterns in Fish-Assemblage Structure Over a Decade of Monitoring in the Upper Delaware River Watershed

Microplastic concentration, characterization, and size distribution in the Delaware Bay estuary

Microplastic (MP) pollution has been widely reported across water matrices including in estuaries, which are important for the understanding of oceanic MPs. Estuaries can greatly alter the fate, transport, size distribution, and abundance of plastic pollution. The aim of this study was to quantify and characterize MP pollution in the Delaware Bay estuary USA, including the size distribution. Samples (N = 31) were collected from the mouth of the Delaware River to the coastal ocean including multiple frontal zones across two sampling campaigns (2019 and 2022). MP were extracted from the collected particles using wet peroxide oxidation and density separation with saturated sodium chloride. Particles collected on 500 μm mesh sieves were analyzed via Fourier transform infrared (FTIR) spectroscopy. Across all samples, 324 of the 1015 particles analyzed were MP, and 11 macroplastics were observed. MP concentrations ranged from below detection to 4.12 MP/m3 (mean 0.34 ± 0.80 MP/m3). No significant differences were observed between sampling sites; nonetheless, the two highest MP concentrations were observed when sampling along frontal zones with visible debris including macroplastics. Polyethylene (53%) and polypropylene (43%) were the most abundant polymers observed. The majority of the non-plastic particles were classified as particulate natural organic matter (82% of non-plastics). Particles from samples collected during 2022 (N = 864) also had color, morphology, and two size dimensions recorded. MP particle size was significantly associated with sampling site, with the coastal ocean sampling site generally having the smallest MPs. A correlation between total post-extraction particles and total plastic particles was observed. Aspect ratios for the plastics ranged from one to 40.7, with larger ratios for fibers, with a mean (±standard deviation) of 3.39 ± 4.72 (unitless). These aspect ratios can be used to select shape factors used to estimate the total volume of MP in the studied size range. Overall, these results can help inform fate, transport, and risk assessments related to estuarine plastic pollution.

Future Changes in Regional Tropical Cyclone Wind, Precipitation, and Flooding Using Event‐Based Downscaling

AbstractUnderstanding changes in the hazard component of climate risk is important to inform societal resilience planning in a changing climate. Here, we examine local changes in wind speed, rainfall, and flooding related to tropical cyclones (TCs) and compare them across statistical and dynamical modeling approaches. Our focus region is the Delaware River Basin, located in the northeastern United States. We pair event‐based downscaling with large ensemble climate model information to capture the details of extreme TC wind, rain, and flooding, and their likelihood, in a changing climate. We identify local TCs in the Community Earth System Model 2 Large Ensemble (CESM2‐LENS). We find fewer TCs in the future, but these future storms have higher wind speeds and are wetter. We also find that TCs produce heavier 3‐day precipitation distributions than all other summertime weather events, with TCs constituting a larger percentage of the upper tail of the full precipitation distribution. With this information, we identify a small collection of 200‐year return events and compare the resulting TC rain and wind across dynamical and statistical downscaling methods. We find that dynamical downscaling produces peak rain rates far higher than CESM or the statistical downscaling method. It can also produce quite different future changes in precipitation totals for the small set of events considered here. This leads to vastly different flood responses. Overall, our results highlight the need to interpret future changes of event‐based simulations in the context of downscaling method limitations.

Discovery and Genomic Characterization of a Novel Hepadnavirus from Asymptomatic Anadromous Alewife (Alosa pseudoharengus).

The alewife (Alosa pseudoharengus) is an anadromous herring that inhabits waters of northeastern North America. This prey species is a critical forage for piscivorous birds, mammals, and fishes in estuarine and oceanic ecosystems. During a discovery project tailored to identify potentially emerging pathogens of this species, we obtained the full genome of a novel hepadnavirus (ApHBV) from clinically normal alewives collected from the Maurice River, Great Egg Harbor River, and Delaware River in New Jersey, USA during 2015-2018. This previously undescribed hepadnavirus contained a circular DNA genome of 3146 nucleotides. Phylogenetic analysis of the polymerase protein placed this virus in the clade of metahepadnaviruses (family: Hepadnaviridae; genus: Metahepadnavirus). There was no evidence of pathology in the internal organs of infected fish and virions were not observed in liver tissues by electron microscopy. We developed a Taqman-based quantitative (qPCR) assay and screened 182 individuals collected between 2015 and 2018 and detected additional qPCR positives (n = 6). An additional complete genome was obtained in 2018 and it has 99.4% genome nucleotide identity to the first virus. Single-nucleotide polymorphisms were observed between the two genomes, including 7/9 and 12/8 synonymous vs nonsynonymous mutations across the polymerase and surface proteins, respectively. While there was no evidence that this virus was associated with disease in this species, alewives are migratory interjurisdictional fishes of management concern. Identification of microbial agents using de novo sequencing and other advanced technologies is a critical aspect of understanding disease ecology for informed population management.

Records of fleas (Siphonaptera) from Delaware.

We present an annotated checklist of fleas (Siphonaptera) known to occur in the state of Delaware based on an examination of Siphonaptera collections at the University of Delaware and the Carnegie Museum of Natural History, as well as new specimens of fleas we collected from wildlife, other hosts, and tick flags. We review published records and compile them herein with our new records, which include 3 species previously unreported from Delaware. With these additions, there are now 18 flea species from 19 avian and mammalian hosts documented from Delaware.

Advancing freshwater science with sensor data collected by community scientists

Autonomous sensor networks providing real‐time data are growing in popularity with community scientists due to instant availability of high‐frequency data. What role does this monitoring play in watershed assessment alongside agency‐run monitoring programs? How accessible, interoperable, and reusable are the data for other researchers? We compared a community science‐led stream monitoring network—EnviroDIY—in the Delaware River Basin, in which more than 50 watershed organizations have deployed more than 100 stations monitoring temperature, electric conductivity, depth, and sometimes turbidity, with the Basin's US Geological Survey (USGS) stream gauge network. The EnviroDIY network (n = 124) complemented the USGS network (n = 102) by monitoring sites with different watershed sizes and land‐use distributions. Although data were accessible and interoperable using a web data portal, community scientists had difficulty sharing metadata that would enable data reuse outside this project and they required support analyzing these large datasets to understand threats to watershed conditions. We address those needs here with a conceptual framework for interpreting data and communicating results.

Spatiotemporal patterns in habitat use of natal and non-natal adult Atlantic sturgeon in two spawning rivers

BackgroundMonitoring movement across an organism’s ontogeny is often challenging, particularly for long-lived or wide-ranging species. When empirical data are unavailable, general knowledge about species’ ecology may be used to make assumptions about habitat use across space or time. However, inferences about habitat use based on population-level ecology may overlook important eco-evolutionary contributions from individuals with heterogenous ethologies and could diminish the efficacy of conservation and management.MethodsWe analyzed over a decade of acoustic telemetry data to understand individual differences in habitat use of federally endangered adult Atlantic sturgeon (Acipenser o. oxyrinchus) in the Delaware and Hudson rivers during spawning season. In particular, we sought to understand whether sex or natal origin could predict patterns in habitat use, as there is a long-held assumption that adult Atlantic sturgeon seldom stray into non-natal rivers.ResultsIn both rivers, migration timing, spawning habitat occupancy, and maximum upstream migration distance were similar between natal and non-natal individuals. While non-natal individuals represented only 13% of fish detected in the Hudson River, nearly half of all tagged fish detected in the Delaware River were non-natal and generally occupied freshwater habitats longer than natal individuals. In both systems males had more heterogenous patterns of habitat use and longer duration of occupancy than did females.ConclusionsThis study demonstrates the importance of non-natal rivers for fulfilling ontogenetic habitat requirements in Atlantic sturgeon. Our results may also highlight an opportunity to improve conservation and management by extending habitat designations to account for more heterogenous patterns in individual habitat use in non-natal freshwater environments.

Deep learning of estuary salinity dynamics is physically accurate at a fraction of hydrodynamic model computational cost

AbstractSalinity dynamics in the Delaware Bay estuary are a critical water quality concern as elevated salinity can damage infrastructure and threaten drinking water supplies. Current state‐of‐the‐art modeling approaches use hydrodynamic models, which can produce accurate results but are limited by significant computational costs. We developed a machine learning (ML) model to predict the 250 mg L−1 Cl− isochlor, also known as the “salt front,” using daily river discharge, meteorological drivers, and tidal water level data. We use the ML model to predict the location of the salt front, measured in river miles (RM) along the Delaware River, during the period 2001–2020, and we compare predictions of the ML model to the hydrodynamic Coupled Ocean–Atmosphere‐Wave‐Sediment Transport (COAWST) model. The ML model predicts the location of the salt front with greater accuracy (root mean squared error [RMSE] = 2.52 RM) than the COAWST model does (RMSE = 5.36); however, the ML model struggles to predict extreme events. Furthermore, we use functional performance and expected gradients, tools from information theory and explainable artificial intelligence, to show that the ML model learns physically realistic relationships between the salt front location and drivers (particularly discharge and tidal water level). These results demonstrate how an ML modeling approach can provide predictive and functional accuracy at a significantly reduced computational cost compared to process‐based models. In addition, these results provide support for using ML models in operational forecasting, scenario testing, management decisions, hindcasting, and resulting opportunities to understand past behavior and develop hypotheses.

Fair Graph Learning Using Constraint-Aware Priority Adjustment and Graph Masking in River Networks

Accurate prediction of water quality and quantity is crucial for sustainable development and human well-being. However, existing data-driven methods often suffer from spatial biases in model performance due to heterogeneous data, limited observations, and noisy sensor data. To overcome these challenges, we propose Fair-Graph, a novel graph-based recurrent neural network that leverages interrelated knowledge from multiple rivers to predict water flow and temperature within large-scale stream networks. Additionally, we introduce node-specific graph masks for information aggregation and adaptation to enhance prediction over heterogeneous river segments. To reduce performance disparities across river segments, we introduce a centralized coordination strategy that adjusts training priorities for segments. We evaluate the prediction of water temperature within the Delaware River Basin, and the prediction of streamflow using simulated data from U.S. National Water Model in the Houston River network. The results showcase improvements in predictive performance and highlight the proposed model's ability to maintain spatial fairness over different river segments.

“We Are All There to Make Sure the Baby Comes Out Healthy”:

“We Are All There to Make Sure the Baby Comes Out Healthy”:

Amplified Extreme Floods and Shifting Flood Mechanisms in the Delaware River Basin in Future Climates

AbstractHistorical records in the Delaware River Basin reveal complex and spatially diverse flood generating mechanisms influenced by the region's mountains‐to‐plains gradients. This study focuses on predicting future flood hazards and understanding the underlying drivers of changes across the region. Using a process‐based hydrological model, we analyzed the hydrometeorological condition of each historical and future flood event. For each event, at the subbasin scale, we identified the dominant flood generating mechanism, including snowmelt, rain‐on‐snow, short‐duration rain, and long‐duration rain. The rain‐induced floods are further categorized based on the soil's Antecedent Moisture Condition (AMC) before the event, whether dry, normal, or wet. Our historical analysis suggests that rain‐on‐snow is the primary flood mechanism of the Upper Basin. Although most frequent, the magnitude of rain‐on‐snow floods is often less severe than short rain floods. In contrast, historical floods in the Lower Basin are primarily caused by short rain under normal AMC. Given the uncertainties in climate projections, we used an ensemble of future climate scenarios for flood projections. Despite variations in regional climate projections, coherent perspectives emerge: the region will shift toward a warmer, wetter climate, with a projected intensification of extreme floods. The Upper Basin is projected to experience a marked decrease in rain‐on‐snow floods, but a substantial increase in short rain floods with wet AMC. The largest increase in flood magnitude will be driven by short rains with wet AMC in the Upper Basin and by short rains with normal AMC in the Lower Basin.

Navigating Risk:: Understanding Chronic Disease Factors in Delaware's College Population.

This study aimed to assess the prevalence of chronic disease risk factors among college students, particularly within a minority-serving institution in Delaware, to inform targeted prevention efforts. A quantitative cross-sectional research design was employed, administering surveys to 457 students at Delaware State University over a three-year period. Surveys assessed tobacco use, physical activity, and dietary habits. Descriptive statistics were used to analyze the data. The majority of respondents were people of color (83%) and female (74%). Significant findings included 24.1% reporting trying tobacco cigarettes, 4.9% smoking in the last 30 days, and 37.2% trying e-cigarettes, with 11.8% vaping in the last month. Regarding physical activity, 65.6% reported being active for 120+ minutes per week, while only 5.7% consumed four or more servings of vegetables daily. Behavioral disparities were observed, with more males reporting tobacco use but also higher engagement in physical activity compared to females. Nutritional intake was inadequate for both genders. Convenience sampling may limit generalizability, highlighting the need for larger, randomized studies. Colleges are pivotal settings for promoting healthy behaviors and addressing chronic disease burdens. Integrated approaches involving education, policy changes, and community engagement are crucial. Investing in health promotion programs is essential for cultivating a healthier future population. This study underscores the importance of preventive measures and equitable access to resources for addressing chronic disease risk factors among college students.