The information presented on this page may be outdated. It may refer to situations which have changed or people who are no longer affiliated with the university. It is archived as part of Mississippi State University's history. This content may not meet accessibility standards. To request an accessible version, contact webmaster.
Publications Use the following field to search our publication database:
PFAS Sample Cross-contamination Caused by Sampling?
Year: 2024 Authors: Boone L.
PFAS are prevalent in numerous items used when taking samples for laboratory testing. This, coupled with the fact that laboratories calibrate instruments to detect PFAS in the single digit ppt range, has led to a great deal of concern about the potential for cross-contamination of samples caused by sampling. Numerous PFAS sampling SOPs call for significant measures not required when sampling for other contaminants, as well as more field QC samples. Is all this concern warranted? During this presentation attendees will learn about the incidence of cross-contamination that occurred in 2023 over a six-month period in water samples taken from around the country. The data set is comprised of over 14,000 drinking water analytical results. Practices taken to mitigate cross-contamination will also be discussed.
Private Water Well Basics
Year: 2024 Authors: Pigg J.
Many Texas citizens are moving from urban areas to rural areas and as they make that move they are obtaining property that has a private water well and do not have the background knowledge to know what that means. The Texas Well Owner Network (TWON) program brings water quality screening and educational presentations to those citizens to increase awareness and improve actions of private well managers. The program hopes to raise awareness by educating on aquifers, well components, state well completion standards, and well maintenance tips. Our program offers water quality screening for common drinking water issues and we have held over 200 events with over 10,500 participants participating. We have been in over 180 of the 254 counties in Texas with a goal of doing a program in all counties in the state. We work closely with county extension agents and local groundwater conservation district personnel when planning and implementing our program to get local "buy-in" and participation.
Diffusion of water-conserving irrigation practices in the Mississippi Delta
Year: 2024 Authors: Oku E., Quintana-Ashwell N., Yun S., Lacy C., Krutz J.
The expanding irrigated acreage in the Mississippi Delta has resulted in increased withdrawals that exceed the recharge rate of the Mississippi River Valley Alluvial Aquifer (MRVAA), leading to a decline in water levels. This trend poses a threat of rapid depletion, potentially resulting in deteriorating water quality and increased pumping costs in the region. Water-conserving irrigation practices are a key component to any potential solution, promising enhanced water use efficiency and sustainable agricultural practices. Nonetheless, farmers adopt these practices at different times. This study used a duration model to identify the factors that influence the timing of adoption of computerized hole selection (CHS) and center pivot (CP) practices. We found that farmers who attended extension meetings and held the belief that CHS would lower their input costs adopted it more promptly. Also, participation in conservation programs facilitated the quicker adoption of CP. Conversely, greater farming experience and a higher cumulative number of adopters were associated with slower adoption times for both CHS and CP.
PFAS Method Considerations for Water Professionals
Year: 2024 Authors: Boone L.
The EPA has now finalized Maximum Contaminant Levels (MCLs) for six PFAS compounds. The rule will be highly impactful for water professionals and require monitoring using EPA 537.1 or EPA 533. The EPA and state agencies have also started to leverage NPDES permits to restrict discharge of PFAS into our nation's waterways. These permits have called for monitoring using PFAS method EPA 1633 and at time EPA 1621 (Adsorbable Organic Fluorine). Analytical testing for PFAS can be costly and there are other methods such as D8421/EPA 8327 than can be used to monitor PFAS at a more cost-effective level. This presentation will cover the methods previously cited as well as others such as TOP Assay (Total Oxidizable Precursor Assay) which are all relevant tools for both drinking water and wastewater professionals.
An analysis of microbial source tracking metadata in coastal Alabama
Year: 2024 Authors: Janssen B., Bilbrey D., Kiel Reese B., Carmichael R.
Microbial source tracking (MST) is increasingly in demand to define water quality and associated seafood safety and public health risks. While MST-related data have been collected for decades, the data have not been collated and made readily available to define spatial and temporal or other research gaps, inform policy decisions, and ultimately protect natural resources and public health. To address these needs, we established a publicly accessible MST metadata clearinghouse (https://www.disl.edu/research/wastewaterfootprint/alabama-mst-metadata-clearinghouse/) for coastal Alabama to compile and share information about existing or in-progress MST and other monitoring-related datasets. Additionally, the clearinghouse provides a model for similar historical data collation in other coastal areas where microbial source tracking is of interest. The clearinghouse currently includes 30 metadata entries from a wide diversity of participating organizations and researchers, with an opportunity for those with relevant data to contribute through the website's questionnaire. We have organized the metadata into four major indicator categories: bacterial, viral, genetic, and chemical. Entries related to these indicators and their associated methodologies can be accessed using the clearinghouse, among other search parameters. With data dating back to 1953, the clearinghouse boasts an impressive timescale of indicators and methodologies within southern Alabama. A temporal analysis of the metadata found nutrients as the most consistently available indicator, with classical indicators such as e. coli and fecal coliforms rising to prominence after 1990 and newer indicators such as DNA markers and stable isotope analysis gaining traction in the mid-2000s. A preliminary spatiotemporal analysis of the metadata suggests an apparent shift in monitoring sites along Mobile Bay. Such analyses can hopefully help researchers identify data gaps to fill and assist practitioners in locating applicable data.
Cyanotoxin Testing of Mississippi's Seafood During a Cyanobacteria Bloom
Year: 2024 Authors: Fleming M., Broussard K., Carron A., Anzola N.R., Glover K.
The Bonnet Carre Spillway, situated northwest of New Orleans, Louisiana, was opened twice in 2019, resulting in a combined total of 123 days of discharge. This diversion directed large volumes of freshwater from the Mississippi River to Lake Pontchartrain, ultimately draining into the western Mississippi Sound. The continuous influx of freshwater facilitated the rapid and temporary colonization of the Mississippi Sound by cyanobacteria, previously assumed to be unaffected by these freshwater phytoplankton. Cyanobacteria blooms can produce various cyanotoxins, raising concerns about seafood toxicity managed by the Mississippi Department of Marine Resources (MDMR) within state waters. Following the first sighting of Microcystis spp. on June 12, 2019, water samples were collected by MDMR staff and confirmed for species identification by NOAA's Phytoplankton Monitoring Network. Positive tests for cyanotoxins from water samples by the Dauphin Island Sea Lab (DISL) prompted MDMR to assess seafood for toxicity. The cyanobacteria bloom persisted for several months, during which 1,333 water samples were analyzed for the genus, and 92 seafood samples and 77 water samples were tested for cyanotoxins. Testing was conducted by DISL and Green Water Labs. Seafood species tested included various finfish species, Eastern Oyster, Blue Crab, and two penaeid shrimp species (Brown and White). The results from the tissue samples exhibited highly variable values, likely influenced by tissue type, the animal's position in the water column, trophic level, and the condition and density of the bloom. Further research is necessary to quantify cyanotoxin-induced fish toxicity and establish thresholds for consumption.
On-farm efficacy of cover crop treatments on sediment/nutrient load transport abatement and crop yields
Year: 2024 Authors: Hampshire J., Spencer D., Krutz J., Oakley G.
Conservation cropping systems promote nutrient and sediment load reductions in surface runoff. This study was conducted to determine whether conventional tillage with fallow season cover crops can reduce nutrient loads and sediment runoff from on-farm conventionally tilled row-crop fields. The effects of cover crop implementation, in a raised seedbed production system on sediment and nutrient load/loss and cash crop yield from corn (Zea mays L.) and soybean [Glycine max (L.) Merr.], were investigated at six paired production fields across the Delta region of Mississippi on soil textures ranging from clay to silt loam. The results from this study are yet to be analyzed. However, the plan is to report on crop yields and total solids/nutrients transported off field.
Runoff analysis due to spatially variable rainfall and land use changes at a tributary level watershed
Year: 2024 Authors: Parajuli P.
Guyandotte River Watershed (GRW) is a tributary to the Ohio River at West Virginia. This study was conducted at a tributary level because the tributary can serve as an important contributor of runoff that carries pollutants to the main river which may support algal growth in the main river. This study aims to evaluate runoff from the GRW due to spatially variable inputs of rainfall data and land use patterns of the GRW. Spatial distribution of precipitation will be determined based on weather stations, seasons, and topographical characteristics of the watershed. The land use data in the form of cropland data layer will be utilized from the United States Department of Agriculture (USDA). Preliminary results from the model simulations will be presented using appropriate statistics.
Remote Sensing of Water Quality Parameters over Western Mississippi Sound by Using Sentinel-3 OLCI and Machine Learning
Year: 2024 Authors: Ahmad H., Dash P., Turnage G., Moorhead R.J.
Remote sensing has emerged as a crucial tool for monitoring water quality, offering a cost-effective and spatially comprehensive approach for monitoring water quality and assessing aquatic ecosystem health. The remote sensing sensor, Ocean and Land Color Instrument (OLCI), onboard the Sentinel 3A satellite provides significant opportunities for monitoring coastal waters. With a spatial resolution of 300 m across 21 spectral bands, it enables extraction of detailed information about water quality parameters. In this study, we investigated the feasibility of using OLCI products to monitor an optically complex coastal water body in the western Mississippi Sound (WMS), building on the potential of remote sensing to address water quality issues. We employed machine learning (ML) and deep learning models for developing algorithms for chlorophyll-a (Chl-a), colored dissolved organic matter (CDOM), turbidity, and surface dissolved oxygen using OLCI imagery and in situ measurements by an autonomous surface vessel in 2021, 2022, and 2023 from WMS. Our approach involved applying automatic model selection algorithms to determine the optimal combination and number of spectral bands for training models. Notably, for Chl-a estimation, the random forest (RF) model yielded the highest adjusted (Adj) R² of 0.96 with a root mean squared error (RMSE) of 0.31 μg/L. For CDOM, the RF, extreme gradient boosting (XGBoost), and decision tree models produced high Adj R² values of 0.99, 0.98, and 0.98, with RMSEs of 1.18, 1.16, and 1.18 μg/L, respectively. Similarly, for turbidity, RF, XGBoost, and K-Nearest Neighbors models emerged as top performers, demonstrating high accuracy. For dissolved oxygen estimation, RF and XGBoost models exhibited robust performance across various metrics, both achieving an Adj R² of 0.89, indicating an excellent fit between the estimated and actual values. This study demonstrates the efficacy of utilizing OLCI products coupled with ML techniques for robust monitoring of water quality parameters in optically complex coastal environments, significantly contributing to enhanced environmental monitoring, management, and conservation efforts.
CWA Section 401 Water Quality Certification Improvement Rule
Year: 2024 Authors: Becker J.
The final 2023 Clean Water Act Section 401 Water Quality Certification Improvement Rule (2023 Rule) went into effect on November 27, 2023. The final rule is grounded in the fundamental authority granted by Congress to states, territories, and Tribes to protect water resources that are essential to healthy people and thriving communities over the past 50 years. This presentation provides an overview of the final 2023 rule certification process, summarizes some of the major changes in the certification process as compared to the 2020 Rule, and highlights a recent revision to the federal water quality standards (WQS) regulation, effective June 3, 2024, that describes how EPA and states must consider applicable Tribal reserved rights when establishing WQS.
Updates on "waters of the United States"
Year: 2024 Authors: Becker J.
"Waters of the United States" is a threshold term in the Clean Water Act that establishes the geographic scope of federal jurisdiction under the Act. This presentation provides some background on "waters of the United States," including the recent regulatory changes, litigation, and court cases; summarizes key aspects of the pre-2015 regulatory regime and the January 2023 rule, consistent with Sackett; and covers implementation tools and additional resources.
Storm surge predictions for tide gauges along the Gulf of Mexico and U.S. East Coast with deep learning and explainable AI through a unified predictor domain.
Year: 2024 Authors: Zafor M.A., Rashid M.M.
Storm surges cause coastal floods, which are expected to become more frequent with climate change. Understanding, predicting, and mitigating coastal flood risks requires accurate storm surge modeling. Typically, storm surges are modeled at daily time scales at each tide gauge, extracting predictor features from separate predictor domains (around the tide gauge) that correspond to each gauge. Thus, they have multiple limitations: (1) daily surges can be superimposed with sub-daily tide to derive the total water level, (2) redundant predictor data when the predictor domains overlap for adjacent gauges, (3) unnecessary reputation of the same analysis for each tide gauge, necessarily training separate models for each tide gauges. To overcome these challenges, this study models hourly surges simultaneously at multiple tide gauges using a unified predictor domain that covers all selected tide gauges. We train three deep learning (DL) algorithms, including Convolutional Neural Networks (CNN), Long Short-Term Memory networks (LSTM), and a hybrid CNN-LSTM (ConvLSTM), to model hourly surges at tide gauges along the Gulf of Mexico (GOM) and the U.S. East Coast. In short, we develop models to predict hourly surges at 28 tide gauges using predictors (i.e., atmospheric variables) extracted from a single domain. We identify the best-performing models using efficiency statistics such as R2, RSME, and MSE. Results show that the ConvLSTM is the most efficient. The best model is further tested for its ability to reproduce the spatial variability of mean surges, extreme surge statistics, probability distribution, and the characteristics of surge hydrographs (e.g., peak, duration, severity, and intensity). Additionally, we employ explainable DL techniques, such as Gradient-weighted Class Activation Mapping (Grad-CAM), to investigate how the model leverages plausible physical relationships between surge and predictor variables during training and prediction of surges. Results demonstrate that deep learning techniques can effectively address the limitations of current storm surge modeling practices and provide robust predictive performance. These models are valuable for historical reconstruction and future projections of surges under various climate change scenarios.
Rapid Flood Extent Mapping from Satellite Images in a Heavily Clouded Region
Year: 2024 Authors: Ahmed R.U., Rashid M.M.
Satellite imagery has gained considerable traction for near real-time and rapid flood mapping because it is largely agnostic for flooding mechanisms and drivers of flooding, hence applicable to any floods, whether pluvial, fluvial, or both (i.e., compound flood). However, flood mapping using satellite images is challenging due to cloud cover and shadows, specifically in the regions where floods usually occur during monsoons with huge clouds (e.g., South Asian countries). Typically, optical radar images (e.g., Landsat) alone or combined with Synthetic Aperture Radar (SAR) images (e.g., Sentinel-1) under the multisource flood mapping approach are used for rapid flood detection. These optical images are often affected by clouds and cloud shadows; hence, they are limited to providing flood inundation information. In contrast, this work presents a simple yet effective approach for flood delineation using Sentinel-1 SAR images where the backscatter values of seasonal water bodies are used to identify the flooded pixels. SAR satellite images are insensitive to clouds and thus ideal for mapping floods in densely cloud covered areas. The method is used to map the extent of flooding during the catastrophic flood in the northeastern part of Bangladesh in 2022 leveraging cloud-based computing capabilities of Google Earth Engine (GEE). The results are comparable to the Global Flood Awareness System (GloFAS) and other widely employed methods, such as the multisource flood mapping (MSFM) technique. Besides, for the instances where MSFM miscalculates flood extent due to cloud covers, our approach provides flood maps comparable to GloFAS. Furthermore, the method is applied for mapping the flood extent for the monsoon months (May to October) from 2015 to 2023, thereby generating a probabilistic flood map. This work enhances the utilization of SAR images for rapid flood delineation and probabilistic flood mapping, enabling decision-makers and emergency responders to get precise information during a flooding event as well as future mitigation measures.
Parallel Implicit Solvers for 2D Numerical Models On Structured Meshes
Year: 2024 Authors: Zhang Y., Al-Hamdan M.Z., Chao X.
This paper presents the parallelization of two widely used implicit numerical solvers for the solution of partial differential equations on structured meshes, namely, the ADI (Alternating-Direction Implicit) solver for tridiagonal linear systems and the SIP (Strongly Implicit Procedure) solver for the penta-diagonal systems. Both solvers were parallelized using CUDA (Computer Unified Device Architecture) Fortran on GPGPUs (General-Purpose Graphics Processing Units). The parallel ADI solver (P-ADI) is based on the Parallel Cyclic Reduction (PCR) algorithm, while the parallel SIP solver (P-SIP) uses the wave front method (WF) following a diagonal line calculation strategy. To map the solution schemes onto the hierarchical block-threads framework of CUDA on GPU, the P-ADI solver adopted two mapping methods: one block-thread with-iterations (OBM-it) and multi-block-threads (MBM), while the P-SIP solver also used two mappings: one conventional mapping using effective WF lines (WF-e) with matrix coefficients and solution variables defined on original computational mesh, and a newly proposed mapping using all WF mesh (WF-all) on which matrix coefficients and solution variables are defined. Both the P-ADI and the P-SIP have been integrated into a two-dimensional (2D) hydrodynamic model, CCHE2D (Center of Computational Hydroscinece and Engineering), model, developed by National Center for Computational Hydro-science and Engineering at University of Mississippi. This study for the first time compared these two parallel solvers and their efficiency using example and applications in complex geometries, which can provide valuable guidance for future uses of these two parallel implicit solvers in computational fluids dynamics (CFD). Both parallel solvers demonstrated higher efficiency than their serial counterparts on CPU (Central Processing Unit): 3.73~4.98 speedup ratio for flow simulations; and, 2.166~3.648 speedup ratio for sediment transport simulations. In general, the P-ADI solver is faster than but not as stable as the P-SIP solver; and for the P-SIP solver, the newly developed mapping method WF-all significantly improved the conventional mapping method WF-e.
Integrating Autonomous Surface Vessel Data and UAS Imagery for Accurate Turbidity Estimation over the Oyster Reef in the Western Mississippi Sound: A Machine Learning Approach
Year: 2024 Authors: Nur A.M., Dash P., Wolfe J.S., Turnage G., Hathcock L.
Turbidity is a crucial water quality parameter that impacts aquatic organisms, including fish and oysters, by disrupting food webs and creating low-oxygen zones. While satellite remote sensing has been widely used for monitoring coastal and inland waters, it faces limitations in spatial and temporal resolutions, and cloud cover can impede imagery availability. Recently, Unmanned Aircraft Systems (UAS) based remote sensing has emerged as a valuable alternative to overcome these challenges. This study leverages an Autonomous Surface Vessel (ASV) equipped with water quality sensors to automate traditional water sampling approach, enabling real-time data collection along predefined transects. The objective was to develop a robust turbidity estimation algorithm by integrating UAS-measured water-leaving remote sensing reflectance (Rrs), ASV-measured turbidity data, and machine learning models. Data were collected during four field campaigns across different seasons from over the Merill Shell oyster reef in the western Mississippi Sound using the ASV. Concurrently, UAS flights in July 2021 and September 2022 captured imagery using a Micasense RedEdge-MX multispectral sensor with five spectral bands. Principal Component Analysis (PCA) was applied to reduce redundancy and extract significant information from the highly correlated spectral bands and derived indices. Seven machine learning algorithms were trained using the first five principal components out of 19 variables: Random Forest (RF), Support Vector Machine with Radial Kernel (SVM-RBF), Multiple Linear Regression (MLR), Gradient Boosting Decision Tree (GBDT), Extreme Gradient Boosting (XGBoost), Enhanced Adaptive Regression Through Hinges (EARTH), and Cubist. The SVM-RBF model demonstrated the highest performance with an R² of 0.943, RMSE of 0.454, and MAE of 0.359, followed by Cubist and XGBoost. This study underscores the effectiveness of machine learning algorithms in generalizing turbidity data from ASV and highlights the potential of UAS imagery for continuous monitoring of small-scale water bodies, such as those over an oyster reef. It also demonstrates the feasibility of integrating ASV data with high-resolution imagery for large-scale machine learning algorithm development with extensive ASV datasets, offering a robust water quality monitoring and management approach.
Boosting upland soil health by integrating soil amendments and cover cropping
Year: 2024 Authors: Dai W., Zhang X., Feng G., Huang Y., Adeli A.
Upland soils are often deficient in organic carbon, and prone to water and nutrient losses. We conducted a four-field study to investigate the effects of soil amendments and cover crops on soil physiochemical properties (texture, mean weight diameter of aggregates, cation exchange capacity, pH, organic carbon, total nitrogen, calcium, and magnesium,) in upland top soil (0-15 cm soil layer) under a no-till corn system near Pontotoc, Mississippi. Treatments of winter cover crop and no cover crop were implemented to the main plots, while soil amendments with fertilization treatments included poultry litter and inorganic fertilizer nitrogen alone and in combination with flue gas desulfurization gypsum and lignite, and an unfertilized control were assigned to the sub-plots. After four years of growing corn, the application of poultry litter in combination with flue gas desulfurization gypsum and lignite significantly (p < 0.05) increased cation exchange capacity, pH, organic carbon, total nitrogen, and the content of sand as compared to the other treatments. However, the cover crop treatments did not cause differences in these indicators. Together, the integration of organic and inorganic amendments such as poultry litter, flue gas desulfurization gypsum, and lignite into a no-till corn system appears to enhance soil physical and chemical properties, potentially improving soil health in upland soils. These findings suggest that such integrated practices could be beneficial to sustainable agriculture in upland regions.
The synergy of cover cropping and nutrient management improves soil health in a no-till dryland soybean cropping system in Mississippi
Year: 2024 Authors: Dai W., Zhang X., Feng G., Huang Y., Shankle M.W.
Agronomic practices can affect soil health. A five-field study was conducted in Pontotoc County, Mississippi, to evaluate the effects of cover crops and nutrient management on soil chemical properties (soil organic carbon and pH), and soil physical and hydrological properties (soil bulk density, aggregate stability, saturated hydraulic conductivity, and field capacity) in a no-till, dryland soybean cropping system. The experiment utilized a split-plot design with cover crops [native vegetation (control), cereal rye (Secale cereale), winter wheat (Triticum aestivum), hairy vetch (Vicia villosa), and mustard (Brassica rapa) plus cereal rye] as the main factor and fertilizer source [no fertilizer (control), inorganic commercial fertilizer, and poultry litter] as the secondary factor. After five years of experimentation, relative to native vegetation lands, cereal rye significantly (p < 0.05) reduced soil bulk density by 5%, while mustard plus cereal rye significantly (p < 0.05) increased soil organic carbon content by 15%. Additionally, the use of winter wheat, hairy vetch, and mustard plus cereal rye resulted in the highest values for pH (6.18) and field capacity (29%) across all treatments (p < 0.05). Poultry litter addition substantially (p < 0.05) reduced soil bulk density by 5%, increased soil organic carbon content by 16%, and enhanced field capacity by 14% relative to the control. However, cover crops and poultry litter did not affect soil aggregate stability and saturated hydraulic conductivity. Our study demonstrates that the integration of cover crops such as cereal rye and poultry litter in a no-till, dryland soybean production system improved soil physiochemical and hydraulic properties, further enhancing soil health and agricultural sustainability.
Dynamic monitoring of phycocyanin concentration in Western Mississippi Sound: Integrating Machine Learning Algorithms and Feature Selection Techniques with Uncrewed Aircraft Systems Imagery and Autonomous Surface Vessel Data
Year: 2024 Authors: Islam M.S., Dash P., Liles J.P., Nur A.M., Moorhead R.J.
Phycocyanin (PC) pigment is unique to cyanobacteria. Thus, measuring PC facilitates the monitoring of cyanobacterial blooms in aquatic environments. This paper evaluated ten machine learning algorithms (MLAs) for obtaining spatiotemporal variations of PC concentrations in the Western Mississippi Sound (WMS) using remotely sensed imagery from uncrewed aircraft systems and in situ PC concentrations measured by an autonomous surface vessel. Subsequently, the influence of river discharge and climatic variables on cyanobacterial concentrations were investigated by using a time-series of cyanobacteria maps. To derive the best PC retrieval model, a comprehensive list of 85 features was initially generated, including individual spectral bands, their band ratios, several vegetation indices, and three-band indices. To select the best feature subset for each MLA, the study adopted a combined approach utilizing two innovative feature selection techniques: Sequential Backward Floating Selection (SBFS) and Exhaustive Feature Selection (EFS). SBFS was employed initially to iteratively remove features to optimize model performance. Subsequently, EFS evaluated all combinations of features suggested by SBFS and selected the best subset. Among the ten MLAs, extreme gradient boosting performed the best (R2 = 0.835, root mean square deviation = 0.419 µg/l, unbiased mean absolute relative difference = 0.176 µg/l, and average percentage difference = 18.072%) in retrieving PC concentration. The time-series analysis revealed variations in PC concentration in WMS from 2018 to 2022. The highest average concentration was observed in 2019, attributed to the introduction of diverted Mississippi River water through the Bonnet Carre spillway, leading to an unprecedented cyanobacterial bloom. Additionally, the average PC concentration was consistently higher in summer than any other time of the year, likely due to elevated air temperatures and ample sunshine promoting cyanobacterial growth. The method formulated in this study enhances quantitative monitoring of PC concentrations in coastal waters such as WMS and provides valuable insights for future water quality monitoring initiatives in other regions.
Evaluating Irrigation Scheduling Methods and Telemetry Services on Soybean Production under Sharkey Clay in the Mississippi Delta
Year: 2024 Authors: Russell D., Gholson D.
Irrigation scheduling, the decision of when and how much water to apply to a field, aims to maximize irrigation efficiency by replenishing the soil profile with the exact amount of water needed. Studies indicate producers often over-irrigate, exceeding the water needed for optimal crop yield, as confirmed by soil moisture sensors. Recently, companies have developed hardware and software to aid in irrigation management decisions. This study aimed to evaluate various irrigation scheduling methods and sensor telemetry services on soybeans grown in Sharkey Clay soil. Initiated in spring 2023 at the National Center for Alluvial Aquifer Research (NCAAR), the study utilized a randomized complete block design with three replications. Each replication included eight eight-row plots (26.67 ft x 450 ft) on 40“ row spacing. Evaluated methods included Watermark 200SS soil moisture sensors at a -75 kPa irrigation trigger, Simplot's SmartFarm irrigation service, Goanna Ag's GoField irrigation service, an NCAAR-developed Sentek relative rate of depletion method, an NCAAR-developed soil water balance model, a free soil water balance app (SI Crop Fit) from the University of Georgia and Florida, a weekly calendar schedule, and a no irrigation control. The soybean variety planted was Asgrow 47XF2 at 130,000 seeds/acre. The study site was furrow irrigated and utilized a skip-row irrigation pattern. Irrigation was delayed until the R2 growth stage and triggered by the respective treatments. Irrigation ceased at the R6.5 growth stage, with data on yield and total water use collected. In the first year of the experiment, minimal yield differences were observed among methods, but the NCAAR soil water balance model yielded 5 bu/ac higher than Goanna Ag's GoField service. The SI Crop Fit app required the least irrigation (3 times) with a yield of 70.3 bu/ac, while the NCAAR soil water balance model and Sentek method (4 times) produced the highest yields at 79.9 and 79.0 bu/ac, respectively. Excessive irrigation was noted with Watermark (-75 kPa) (5 times), weekly calendar (6 times), and Goanna Ag (7 times). In its first year, the study found treatments with fewer irrigations most profitable. The study was continued in 2024, and this year's results will be presented at the conference.
Microplastic Pollution in Runoff and Standing Water from Flooded Farms in the Mississippi Delta, and Potential Remediation with Biochar
Year: 2024 Authors: Heinen E., Olubusoye B., Cizdziel J., Moore M., Taylor J.
Microplastics (MPs) in agricultural landscapes are of great concern with the rise of plasticulture, the practice of using plastic materials for agricultural applications. Plastics degrade with exposure to light and water, breaking them into smaller fragments. Agricultural runoff is a source of MP pollution as it transports the particles downstream to water bodies, however the characteristics of MPs in such runoff and efforts to mitigate such pollution is not widely studied. Here we report initial findings from a study of MPs in both agricultural runoff and in standing water from a flooded farm in the Mississippi Delta, as well as results from a column feasibility study investigating the efficacy of biochar in removing MPs from the agricultural runoff. MPs in the field samples were extracted using a chemical digestion to remove natural organic matter and a density separation to float the plastics. The isolated MPs were characterized using Fourier transform infrared microscopy (µ-FTIR). For the runoff, an average of 237 MPs/L (range 27–609) were observed. The most prevalent polymer types were identified as polyethylene, polyamide, polyvinyl chloride, polyurethane, acrylonitrile butadiene styrene, and polyarylamide. Experimental fields in Indianola, MS are flooded in the fall in order to provide habitats for migratory birds as well as improve soil denitrification rates. Standing flood water from two fields was sampled and the preliminary results show MP concentrations of 51 MPs/L in one field. The most commonly identified polymer types are polyethylene terephthalate and polypropylene. The column study tested both pinewood and sugarcane biochar. Different shapes, sizes, and types of MPs were stained with Nile red dye to observe their movement through the column using fluorescence. Between 86.6% and 92.6% of the MPs were retained in the column regardless of the MP shape, size, and type. Overall, these findings show that MPs are prevalent in both farm runoff and standing water, and that migratory birds may be exposed to MPs therein. Further, biochar shows promise in capturing MPs in agricultural runoff and this approach warrants further scrutiny in field experiments. This presentation will include an update on ongoing studies to collect additional data on MPs at these locations.
Remote Sensing and Machine Learning for Monitoring and Mapping Total Suspended Solids in the Mississippi Sound
Year: 2024 Authors: Caballero C.B., Martins V.S., Paulino R.S., Butler E., Sparks E.
The Mississippi Sound (MS), located along the coasts of Mississippi, western Alabama, and eastern Louisiana in the northern Gulf of Mexico, is vital for commercial and ecological activities. Water quality (WQ) issues are prevalent in this region, which has led to WQ being routinely identified by natural resource managers, the tourism and seafood industries, and many other sections as a top concern that should be addressed. Therefore, WQ monitoring in this region is essential for maintaining and conserving water resources, economic activities, and human health. However, traditional point sampling methods are limited to capture spatial variability and are costly. In contrast, remote sensing with orbital sensors offers comprehensive temporal and spatial monitoring of water's optical properties, including Total Suspended Solids (TSS). Mapping TSS distribution in coastal waters is critical for understanding and managing the ecosystem health and coastal processes. In view of that, this study aims to develop an integrated framework that leverages satellite imagery, in-situ data, and machine learning algorithms to quantify and map the distribution of TSS in the Mississippi Sound. We developed an empirical model for monitoring TSS concentration on the Mississippi Coast based on the multispectral sensor OLCI onboard Sentinel-3. Four campaigns were carried out on the Mississippi Coast, and data on reflectance and TSS were collected. This data was then analyzed using two machine learning algorithms: Random Forest (RF) and Support Vector Regression. The results showed that the RF model was the best fit for the data patterns (r-squared = 0.978, bias log = 1.05, and RMSE = 1.57). The model was then applied to three Sentinel-3 OLCI images of the MS and validated with TSS in situ data. The accuracy assessment of the estimated TSS achieved an r-squared of 0.626, a bias log of 1.2, and an RMSE of 7.43 mg/L. This study shows that remote sensing data can accurately estimate TSS on the Mississippi Coast. The empirical model developed can serve as a valuable tool for reducing costs and efforts associated with fieldwork measures and for gaining insights into the dynamics of TSS over the Mississippi Coast, enabling more effective management and conservation strategies for this ecologically and economically important coastal region.
Integrating ET datasets into Watershed Simulations: Insights from Goodwin Creek Experimental Watershed in Mississippi
Year: 2024 Authors: Rébillout L., Shuchana I., Al-Hamdan M., Ozeren Y., Bingner R.
An accurate description of a watershed water budget lies, in a large part, in capturing the amount of water in the form of precipitation that the watershed receives, and the water returned to the atmosphere through evapotranspiration (ET). This study investigates the use of remote sensing data (ECOSTRESS, MODIS) and data from land surface models (NOAH and MOSAIC from NLDAS-2) to predict ET as an alternative to the commonly used Penman-Monteith equation where ET is derived from other climate variables and land use/land cover data. Using the estimated ET from these sources, the Annualized Agricultural Non-Point Source (AnnAGNPS) model was applied to Goodwin Creek Experimental Watershed located in north-central Mississippi. This study uses a baseline simulation based on a one-meter DEM delineation of the watershed, manual compilation of the management practices and land use for the Goodwin Creek Experimental Watershed, and using data from 16 local climate stations. Firstly, the daily ET datasets were compared against one another for the 2020-01-01 to 2020-12-31 time period using the Normalized Nash-Sutcliffe Efficiency coefficient (NNSE) as a comparison metric. The pairwise comparisons between MODIS, NOAH, and MOSAIC yielded a minimum NNSE equal to 0.65 which indicated a good consistency between the datasets. Secondly, a sensitivity analysis was performed to determine the response of the model to the ET datasets. Multiple scenarios derived from the baseline simulation using different climate data sources and ET data forced or computed with the Penman-Monteith equation were ran and the outputs (peak discharge, total streamflow, total runoff, water from sheet and rill, and total sediment) were compared against those of the baseline simulation. Simulations that used both NLDAS-2 climate data and ET data from NOAH and MOSAIC showed a higher level of similarities to the baseline simulation than the other scenarios that did not. Finally, selected model outputs aggregated at a monthly frequency were compared to observations of total runoff, total sediment, total fines, and total sand from 1982 to 1991 obtained from local monitoring stations. The results show that outputs were in better agreement with observations when remotely sensed ET was used for AnnAGNPS simulations of Goodwin Creek experimental watershed.
Spatio-temporal variation in surface water in the Mississippi using machine-learning methods with time-series remote sensing data and driving factors
Year: 2024 Authors: Tariq A., Davis J.B.
Surface water bodies play a critical role in the hydrological cycle, supporting ecosystems, human activities, and agricultural practices. Monitoring and understanding the spatio-temporal variations in surface water are crucial for effective water resource management, particularly in large river systems like the Mississippi River. This study employs machine-learning methods to analyze time-series remote sensing data, aiming to elucidate the driving factors behind the variations in surface water across the Mississippi River Basin.
Exploring the sun- and sky-glint effect correction of Sentine-3A/B images over coastal waters
Year: 2024 Authors: Paulino R.S., Martins V.S., Caballero C.B.
Sentinel-3 A/B satellites are collecting 300m daily multispectral data in sun-synchronous orbits around the Earth and were designed specifically for studying ocean colors. The Ocean and Land Colour Instrument (OLCI) sensor aboard the satellite is well-suited for monitoring various features of coastal water, such as chlorophyll levels, harmful algae blooms, turbidity, and organic matter concentration. Nevertheless, spectral water-leaving radiance measured by satellite sensors is highly affected by atmospheric interference, adjacency effects from land targets, and ocean surface conditions, such as sky- and sun-glint effects. Sky- and sun-glint effects are caused and intensified by water surface waves and image acquisition geometry, denoting the reflection of the sky and sunlight from the water's surface, and these effects must be corrected to derive the spectral watercolor features accurately. Existing empirical methods used for removing water surface effects in satellite images are inappropriate for Sentinel-3 images due to limitations of their spectral range (λ = 400 to 1020 nm). The primary goal of this study is to evaluate the sky- and sun-glint correction method for Sentinel-3 (A/B) images. An adjusted empirical method, employing power functions between the blue and near-infrared bands on a pixel-a-pixel approach to remove the sky- and sun-glint effects from water reflectance spectra, was applied on three Sentinel-3 (A/B) images obtained over the Mississippi coast. To validate the water reflectance spectra, a total of 52 points from in-situ observation were used. The outcomes revealed that our methodology using power functions across multispectral images generated consistent water reflectance spectra compared to in-situ observations. The Sentinel-3 (A/B) image spectra ranging from 490 to 779 nm displayed an average Bias of 18.4% after sky- and sun-glint corrections. Before the correction, the Bias values were higher than 100% for all dates analyzed. In conclusion, this methodology significantly enhanced the quality of water reflectance spectra; however, it is limited to blue and red-edge band wavelengths and may result in underestimations of water reflectance magnitude and, consequently, influencing water quality parameter retrievals.
A water quantity assessment for an established tailwater recovery system in the Mississippi Delta
Year: 2024 Authors: Nelson A., Moore M., Delhom C.
In the Mississippi Delta region, tailwater recovery (TWR) systems are an important best management practice to address both water quality and quantity issues. TWRs are surface water capture and irrigation reuse systems, using a combination of a ditches and reservoirs to capture surface water and pumps to move the collected surface water from the ditch into reservoirs and to irrigate nearby fields. To determine if established TWR systems are an effective way to reduce groundwater use, a 10 year old, ditch-only TWR system in Sunflower County, MS was equipped with velocity and flow meters, water level loggers, and rain gauges. This study found 22% of the total applied irrigated water over two growing seasons was sourced from collected tailwater runoff in this system. During the 2023 growing season, only 15.5% of the input water was lost from the system through the outflow pipe, indicating that 84.5% of the input water (precipitation plus irrigation) during the growing season was retained in the soil, utilized for plant growth, or was recirculated in the tailwater recovery system. After a decade of use, the studied TWR system is meeting its dual purpose of reducing groundwater pumping and retaining potentially nutrient and sediment-rich runoff waters within the system, thereby preventing it from entering downstream waterbodies.
Implementation of Site-Specific Management of Nitrogen, Phosphorus and Seeding Rate to Accelerate Nutrient Reduction
Year: 2024 Authors: Galloway L., Spencer D., Reynolds Z.
Variable rate application of fertilizer and variable seeding rate can reduce input costs without sacrificing yield potential. However, adoption of site-specific management practices for fertilizer use and seeding rate will be minimal if farmers are not convinced that site-specific management practices will lower inputs and improve or maintain yield. The objective of this study was to create in-field management zones with crop yield and soil data and to evaluate variable rate of nitrogen (N), phosphorus (P) and seeding rate to improve nutrient use efficiency, reduce soil runoff and improve water quality. On-farm trials were implemented across Mississippi on corn and soybeans. Methodology, data collection and analysis will be discussed.
Impacts of Biodiversity of Short-Rotation Woody Crops on Water Quality
Year: 2024 Authors: Jones N., Siegert C., Dominici B., Shafqat W., Himes A., Renninger H.
Mitigating agricultural nutrient runoff and improving water quality is a key challenge in meeting food and energy demands. To address this challenge, short-rotation woody bioenergy crops, specifically Populus deltoides (eastern cottonwood) and its hybrids can be planted at the interface of riparian areas and agricultural production fields to alleviate fertilizer runoff into adjacent bodies of water. This research employed an experimental design to evaluate the effects of P. deltoides diversity on tree productivity and nutrient uptake and how it mitigates agricultural runoff. We deployed ion exchange resins 0.5 meters below the soil surface at four different sites in Mississippi that contained either monoculture plantings of a single P. deltoides genotype or a mixture of two genotypes for the entire growing season. Across two years and all sites, nitrate concentrations were reduced by 23%, and ammonia concentrations were reduced by 51% relative to concentrations in agricultural soils. In both years, multi-genotype plots reduced soil ammonia concentrations more than single-genotype plots, while the opposite trend was observed for soil nitrogen. The results of this study display the efficiency of the short-rotation woody crops in reducing water quality degradation that may have positive downstream impacts. This study can be used as an example of mitigation techniques for fertilizer runoff in agricultural fields to limit such degradation of water quality and prevent monetary/economic loss for agricultural producers.
On-farm efficacy of cover crop treatments on sediment/nutrient load transport abatement and crop yields
Year: 2024 Authors: Hampshire J., Spencer D., Krutz J., Oakley G.
Conservation cropping systems promote nutrient and sediment load reductions in surface runoff. This study was conducted to determine whether conventional tillage with fallow season cover crops can reduce nutrient loads and sediment runoff from on-farm conventionally tilled row-crop fields. The effects of cover crop implementation, in a raised seedbed production system on sediment and nutrient load/loss and cash crop yield from corn (Zea mays L.) and soybean [Glycine max (L.) Merr.], were investigated at six paired production fields across the Delta region of Mississippi on soil textures ranging from clay to silt loam. The results from this study are yet to be analyzed. However, the plan is to report on crop yields and total solids/nutrients transported off field.
Feasibility assessment of the Groundwater Transfer and Injection Pilot project through pilot testing and regional hydrogeologic modeling, Shellmound, Mississippi
Year: 2024 Authors: O'Reilly A., Wren D., Locke M., Rossell W., Guira M., Mirecki J.
Groundwater depletion of the Mississippi River Valley alluvial aquifer (MRVAA) threatens sustainability of agroecosystems in the intensively cultivated Mississippi Alluvial Plain. The U.S. Department of Agriculture, Agricultural Research Service, National Sedimentation Laboratory partnered with local stakeholders, U.S. Army Corps of Engineers, and U.S. Geological Survey (USGS) to conduct and assess the performance of the Groundwater Transfer and Injection Pilot (GTIP) project in the Delta region of Mississippi (Delta). The GTIP project uses a riverbank filtration-based managed aquifer recharge (MAR) approach, consisting of extracting groundwater from one well near the Tallahatchie River, transferring water through a 1.8-mile pipeline, and reinjecting the water into the MRVAA via two wells where water levels have substantially declined. The system has a design capacity of 1,500 gal/min and began operation in 2021. Two injection tests were conducted for durations of 89 and 204 days, yielding total injected volumes of 550 and 575 ac-ft, respectively. At the injection site, groundwater levels increased up to 6.7 ft whereas major ion concentrations decreased 21% on average in the MRVAA injection zone. An inset groundwater flow model covering an area of 405 square miles was developed by USGS based on a regional parent model of the entire Delta. The inset model leverages a high-resolution airborne electromagnetic survey to test three alternative layering conceptualizations. The transient (1900–2018) inset model was developed using MODFLOW 6 and Soil Water Balance models. Using the PEST++ Iterative Ensemble Smoother, the three models were calibrated against groundwater levels, streamflows, and stream stage. Results indicate that the most detailed representation of MRVAA layers produced the best calibration. A forecast model indicates that under average irrigation and recharge conditions (2010–15), the GTIP project has the potential to increase groundwater levels as much as 10 ft around the injection site by 2050, but sustained increase would require yearly repetition of water transfer rates. Findings of the pilot testing and hydrogeologic modeling demonstrate the GTIP project can increase the amount of water in the MRVAA under the unique hydrogeologic conditions in the Delta. To help assess the possibility of large-scale implementation of MAR throughout the Delta, a third injection test and further forecast modeling will be conducted.
Application of electrical resistivity tomography (ERT) and electromagnetic induction (EMI) for groundwater site investigation
Year: 2024 Authors: Buskes E., Wodajo L.T., Alim M.S., Hickey C.J., O'Reilly A.M.
The sustainability of groundwater supplies is critical to meeting present and future water demands. Such sustainability is threatened by rapid, unchecked population growth, rising quality of life, increasing demand for food and energy, and impacts of climate change, all of which increase stress levels on our limited groundwater resources. Groundwater depletion increases the need to improve the characterization of subsurface heterogeneity and identify optimal locations and approaches for the placement of groundwater extraction and monitoring wells. In this study, two surface-based geophysical methods, electrical resistivity tomography (ERT) and electromagnetic induction (EMI), were used for subsurface characterization and delineation of aquifers at a USDA hydrologic research site encompassing a creek and a pasture. The study site was in the Goodwin Creek Experimental Watershed in Panola County, Mississippi. A 501-m long ERT survey at 3-m electrode spacing was conducted to characterize the subsurface heterogeneity and delineate the depth and thickness of the aquifer. An EMI survey was conducted using an EM31 instrument over a larger area to provide greater spatial coverage of the site. Disturbed and undisturbed soil samples were collected at different locations to determine soil properties and help interpret geophysical results. A comparison of ERT and EMI results consistently showed where high resistivity values on the ERT are indicated with low conductivity on the EMI map. ERT results indicate that the aquifer thickness varies between <10m near the creek and >30m in the pasture, with variable depths to the top of the aquifer. Results from this study show the advantages and capabilities of electrical resistivity tomography (ERT) and electromagnetic induction (EMI) to delineate aquifers and identify optimal well locations and recharge areas, leading to improved groundwater management and minimizing well failures due to poor placement.
Coupled Numerical Surface and Groundwater Flow Model for Complex Eroded Topography with Gullies using a Conservative, Meshfree, Mimetic Method
Year: 2024 Authors: Fang J., Al-Hamdan M., Vieira D.
For soil conservation practices, it is imperative to precisely quantify the vulnerability of soils to erosion so that necessary and proper measures can be taken to protect farmland. Many studies have shown that soil erodibility is highly affected by soil saturation, which in turn is majorly related to the frequency and intensity of rainfall events as well as land topography. However, current soil erosion models generally ignore the impact from the spatial and temporal variations of soil saturation on erodibility, mostly due to the lack of high-quality field data, especially data with high spatial and temporal resolutions. This gap could be filled by determining dynamic soil saturation patterns using process-based numerical models because such models capture the general hydrodynamic processes, thus providing a set of high-resolution data. The soil saturation in the region near the land surface is mostly controlled by rainfall-induced infiltration and runoff, which involve the interaction of both surface water and groundwater hydrological processes. As a result, a coupled numerical surface and groundwater flow model was developed where the surface water flow is described by the diffusive wave approximation, and the groundwater is modeled with the mixed-form of the Richards equation. The van Genuchten model is used for the soil-water characteristic curves. In addition, as the topography of eroded farmlands is usually complex due to the presence of irregularly shaped gullies, a finer spatial discretization is typically needed in the area near gullies to obtain an accurate modelling result for surface and groundwater exchanges. With traditional mesh-based models, local refinement is particularly difficult because of restrictions on mesh geometry to maintain numerical accuracy. On the other hand, meshfree models offer great flexibly when adding or moving points, making it more suitable for this type of application. Thus, a meshfree mimetic method, which can conserve the mass both locally and globally, was adopted. This newly developed model was first verified with multiple benchmark cases. It was then applied to an agricultural field in Jasper County, Iowa where past erosion had created networks of ephemeral gullies. Both the spatial and temporal variations of soil saturation were simulated, demonstrating its applicability of the model to real problems.
Efficient Bayesian Experimental Design for Estimating Stream and Aquifer Hydraulic Conductivity using Conservative Meshfree Mimetic Numerical Groundwater Model
Year: 2024 Authors: Fang J., Al-Hamdan M., O'Reilly A.
Riverbank filtration system is commonly used in managed aquifer recharge (MAR) projects to sustainably extract clean and safe water resources. It takes advantage of the supply from the infiltrated stream water as well as the filtration function of riverbed and aquifer sediments. To evaluate the efficacy of this system, it is essential to accurately quantify the contributions from the surface and subsurface water. To this end, streambed conductivity and aquifer hydraulic conductivity need to be measured because they crucially determine the intensity of surface-groundwater exchanges. Although a direct measurement is generally difficult, these two parameters can be estimated through the inverse calculation of a pumping test near a river, and the locations of the monitoring wells are important for the accuracy of the estimation result. Thus, before field campaigns it is necessary to conduct an optimal design of the monitoring wells where a process-based numerical model is typically employed. To model a pumping test numerically, fine spatial discretization is often needed for the region near the pumping wells owing to the sharp local hydraulic gradient. On the other hand, much coarser spatial discretization is preferred for the remaining model area to reduce the computational cost. To efficiently tackle this multi-scale problem, the finite volume method using nonmatching meshes, in which grids can be refined locally, was proposed. However, the simulation accuracy of mesh-based numerical models is overall restrained by the quality of meshes. Unstable numerical solutions are frequently encountered when a set of overly irregular meshes is employed, while nonmatching meshes are typically in an irregular shape. As a result, a new meshfree mimetic method, which can robustly handle complex topography, facilitate local discretization refinement and possess mass conservation properties both locally and globally, was developed in this study. This new model was then applied to a Bayesian experimental design of the monitoring wells for estimating streambed and aquifer hydraulic conductivity of a MAR site using riverbank filtration at Shellmound, Mississippi, USA, to provide suggestions for future field campaigns.
The Mississippi Sound Estuary Program Comprehensive Conservation & Management Plan
Year: 2024 Authors: Martin S., Zapfe C., McQueen E., Sparks E.
The Mississippi Sound Estuary Program (MSEP) was created in 2023 with the purpose of connecting the restoration and investment efforts across Mississippi state agencies, federal agencies, academic institutions, non-governmental organizations, and community members to ensure that restoration and investment is done in a collaborate, cohesive, and science-based way. To achieve this, the MSEP has been working across these groups to develop a Comprehensive Conservation and Management Plan (CCMP) to guide MSEP activities for the next five years. Through many discussions, surveys, and workshops, we have identified an array of action items covering education and engagement, water quality, wildlife, and habitat. This presentation will describe both the process by which we have developed the CCMP as well as highlight projects with broad support in the region, such as creating a comprehensive project inventory.
Riparian Hardwood Restoration
Year: 2024 Authors: Nettles J., Hawks B., Tyson C., Hanks D.
SMZs are a foundational forest management practice, are critical for protecting water quality and minimizing sediment and chemical input. SMZs provide additional ecosystem services such as streambank stabilization, inputs of leaf litter and woody debris for food webs, habitat for terrestrial and aquatic species, aesthetics, thermal protection, and timber value. In the Southeastern U.S., state forestry agencies provide harvesting guidelines for intermittent and perennial streams. BMP guidelines often include the ability to remove a portion of canopy trees (or basal area) in an SMZ, and research indicates both partially harvested and unharvested SMZs adjacent to harvests mitigate sediment and protect water quality. Riparian areas in the Piedmont and Coastal Plain of the Southeastern U.S. historically were dominated by diverse mixed pine/hardwood stands that included sweetgum (Liquidambar styraciflua), red maple (Acer rubrum), loblolly pine (Pinus taeda), yellow poplar (Liriodenron tulipifera), oak (Quercus spp.), and several other tree species. Before implementation of modern Best Management Practices (BMPs), Streamside Management Zones (SMZs) along intermittent and perennial streams were harvested during routine clearcuts and replanted with pine. A large portion of Weyerhaeuser Southern Timberlands SMZs are now dominated by mature pine, closed canopy forests. Removing pine to regenerate hardwoods in riparian areas has the potential to improve in-stream organic matter input, positively affecting biodiversity, and would add other forest cover types to the landscape mosaic including mixed-hardwood riparian areas and early successional riparian habitat with high likelihood of wetland species. However, this potential practice needs to be evaluated for short-term water quality effects as well as long term benefits. This study, Riparian Hardwood Restoration, was initiated in early 2023. The goal is to quantify effects of harvesting pine, potentially above current BMP limits, from SMZs on water quality, water quantity, and hardwood regeneration to determine if, when and where it is appropriate in the Southeastern US. Across thirty-two perennial and intermittent streams, we will quantify the effects on hardwood regeneration, tree species diversity, water quality, and aquatic biodiversity compared to more typical SMZ treatments. Guidance will be provided for foresters and agencies to outline conditions where this practice is appropriate.
Utilization of the Agricultural Water Inspections to Mitigate Risks on Farms
Year: 2024 Authors: Bond R., Abdallah-Ruiz A., Silva J.
Microbial water quality risk assessments are essential tools for identifying and mitigating areas susceptible to contamination on farms, particularly in states like Mississippi and Alabama where agricultural practices (ready to eat produce) are prevalent. We will discuss the framework in the agricultural water quality requirements in the Food Safety Modernization Act (FSMA) which mandates routine inspections meant in reducing microbial contamination. This framework requires systematic evaluation of irrigation water sources, delivery systems, and on farm water usage, and identifies areas on farms that are susceptible to microbial contamination. The risk assessments (inspections) incorporate various ways of mitigation including pathogen testing, environmental monitoring, and farmer driven data analysis to pinpoint reasonably foreseeable hazards as defined under FSMA. We will highlight key factors contributing to microbial contamination risks, such as water source type, proximity to livestock, and farm management practices. We will underscore the importance of implementing specific FSMA-compliant water management and mitigation strategies to ensure microbial safety to protect public health. We will also give real world examples of the “dos and don'ts“ from research conducted throughout Mississippi and Alabama.
Integrating Watershed and Surface Water Models for Simulating Flow, Sediment, and Nutrient Dynamics in Channel Networks
Year: 2024 Authors: Chao X., Zhang Y., Al-Hamdan M., Bingner R., Rébillout L.
This paper presents a technical approach that integrates watershed and surface water models to simulate flow velocity, sediment transport, and nutrient distributions within channel networks. The AnnAGNPS watershed model, developed at the ARS National Sedimentation Laboratory, is applied to simulate runoff, sediment, and nutrient loads from upland watersheds. Outputs from AnnAGNPS provide inlet boundary conditions for the CCHE1D channel network model, developed at the National Center for Computational Hydroscience and Engineering (NCCHE), University of Mississippi, to simulate flow, sediment, and nutrients in downstream channels. The integrated modeling system was validated and applied to the Goodwin Creek Experimental Watershed in Mississippi. This watershed was delineated using TOPAGNPS based on a 1-meter DEM, with cells and reaches of the watershed, as well as channel networks, obtained for model simulation. To improve the effectiveness of model simulation, lower-order channels were removed from the system. However, contributions of the correlated sub-watersheds were considered at the nearby channel nodes. This integrated model system provides a useful tool for evaluating water quality in both the upland watershed and the receiving water bodies. Additionally, the effects of watershed practices on the channel water quality can also be assessed.
Investigating the effects of algal content, dissolved oxygen, and temperature on a high-frequency acoustic attenuation system in a controlled laboratory environment
Year: 2024 Authors: Carpenter W., Goodwiller B., Taylor J., Hom E.
Scientific monitoring and research of natural watersheds allows for a greater understanding of changes taking place in the natural infrastructure society is built upon. With the effects of climate change and increased use of water and other natural resources, erosion and other phenomenon can negatively impact not only land and water, but also the agriculture and society on and around it. In an effort to preserve the world's natural resources, the accurate monitoring of sediment transport in streams is being used as a tool in understanding changes in the agroecosystem. The Single Frequency Acoustic Attenuation System (SFAAS) has been developed at the National Center for Physical Acoustics (NCPA) at The University of Mississippi to measure fine suspended sediment concentrations in ephemeral streams. This acoustic system uses two 20 MHz transducers in a transmit-receive configuration at a fixed distance to measure acoustic amplitude. The acoustic system is calibrated using site-specific, physical measurements. A relationship between the experimental volume and a clear water reference value is used to calibrate the loss in acoustic signal over the signal path. In post-processing, the strength of the received signal in the test environment is compared to the strength of the received signal in clear water to create a relative acoustic measurement. This relationship of acoustic amplitude difference can then be compared to pump samples to establish a calibration to estimate suspended sediment concentration. Previous research has indicated that other phenomena may affect the acoustic signal of this system. It was hypothesized that these phenomena could include dissolved oxygen, water temperature and algal activity. Laboratory experiments were designed and conducted to monitor water temperature, dissolved oxygen concentration, and fluorescence in the controlled laboratory during the acoustic system's use. Calibrations were derived to account for the effect of each of these variables on the SFAAS acoustic signal. Dissolved oxygen concentration has no discernable effect. It has been shown that water temperature has a significant effect on attenuation due to the physical deformation of the transducer piezoelectric element. Therefore, a water temperature correction is applied to the acoustic data to correct the received signals. The overall effect of algal concentration is still under investigation. A dilution experiment in the laboratory using Spirulina algae shows that the acoustic system can monitor algae concentrations over a range of fluorescence values to a near saturation volume of algae.
Evaluation of Sector Control Variable Rate Irrigation (VRI) on a Production Field
Year: 2024 Authors: Theobald S., Tagert M.L., Paz J., Lo H.
In Northeast Mississippi, access to groundwater is limited due to drilling depths, and only 37% of the annual precipitation in the region occurs during the growing season. As a result, on-farm water storage (OFWS) systems have been built throughout the region in recent years. These systems capture and store precipitation and runoff that can be used for irrigation during the growing season. Due to the limited amount of rainfall received during the growing season, producers in Northeast Mississippi have a finite amount of water to use for irrigation. This study evaluates the benefits of sector control variable rate irrigation (VRI) on an 18-hectare production field under center pivot sprinkler irrigation in Noxubee County, MS. During the 2023 growing season, cotton was planted on May 9th and May 10th and harvested on October 2nd - 4th. Elevation, yield, and soil moisture data collected from 2018-2021 were analyzed and used to create two distinct irrigation management zones within the field. A soils layer was not included in this geospatial analysis because previous gridded soil sampling confirmed a homogenous soil type of silty clay loam with small areas of silt loam. Two irrigation treatments were applied both to a “dry” irrigation management zone in the southern section of the field and a wet irrigation management zone in the northwestern section of the field. The conventional irrigation treatments received 1.9 cm of water, which was the conventional amount of irrigation applied by the producer. The reduced irrigation treatments received 1.5 cm of water, which was a 20% reduction from the conventional amount of water applied. Each zone was sub-divided into six different pie-shaped sectors, and both irrigation treatments were replicated three times in each zone. The center pivot is equipped with a Lindsay Growsmart IM3000 magnetic flow meter to measure water use. Two sets of Watermark 200SS granular matrix sensors were placed in the centroid of the outermost span of each sector at depths of 30 and 61 cm to measure soil water tension in the rooting zone throughout the growing season, and sensors were removed just before harvest. Soil tension data is being analyzed with yield data to determine if water savings were realized without a yield loss. This presentation will include preliminary results from this multi-year study.
Assessment of century long groundwater exchange between cropland and forestland in Mississippi
Year: 2024 Authors: Ouyang Y., Feng G., Wan Y.
Groundwater resource depletion, exacerbated by extensive utilization and climate change, has emerged as a critical concern in the US and around the world. Employing the US Geological Survey's MERAS (Mississippi Embayment Regional Aquifer Study) groundwater model, we assess spatial distribution, flow exchange, and storage of groundwater between the adjacent cropland and forestland in Upper Yazoo River Watershed, Mississippi over a 115-year simulation period (1900 to 2014). Under normal climate and over the 115 years, the average groundwater head declined 2.7m in cropland but only 1 m in forestland. We attribute the faster groundwater head decline in cropland primarily to intensive groundwater pumping for crop irrigation. The average groundwater flow from forestland to cropland surpassed the reverse flow by a factor of 238, owing to a higher surface elevation in forestland and intensive groundwater pumping in cropland. Over the 115 years, cropland proved to be a net receiver of groundwater, while forestland emerged as a net deliverer. A comparative analysis revealed that, under a very wet climate scenario (i.e., 20% increase in precipitation), cropland groundwater storage-in (i.e., receiving groundwater) increased by 25%, while forestland storage-out (i.e., delivering groundwater) surged by 30%. Conversely, in a very dry climate (i.e., 20% decrease in precipitation), cropland storage-in decreased by 33%, and forestland storage-out diminished by 22%. Results underscore the discernible impacts of climate changes on groundwater storage and suggest that afforestation in the region would be an alternative for saving groundwater resources and supplying more waters to croplands. Our findings offer valuable insights into groundwater supply planning not only in Mississippi but also in comparable conditions worldwide.
Water footprint of cotton and sorghum production under conservational management practices
Year: 2024 Authors: Dhakal M., Locke M., Reddy K., Moore M., Krutz L.J.
Intensive tillage and little to no surface residue management are linked to the deterioration of soil hydraulic properties and increased water footprint of row crop productions in the Mississippi Delta. Cover cropping and minimum tillage are the alternate management practices that can sustain soil health and water storage and improve crop water use efficiency. This study determined whether soil-plant-water relations could be manipulated through conservation production systems that affect soil water dynamics and the water footprint of cotton (Gossypium hirsutum) and sorghum (Sorghum bicolor) production. Effects of tillage [conventional tillage (CT) vs. no-tillage (NT)] and cover cropping [no cover crop (NC) and Austrian pea (Pisum sativum) cover crop (CC)] on soil water balance during crop growth periods were modeled using Root Zone Water Quality Model (RZWQM2) for a site near Stoneville, MS to estimate seasonal evapotranspiration (ET), surface runoff, and deep percolation. Capacitance sensors measured soil volumetric water content to 120 cm in 2020 and 2021 on a half-hourly basis. Pooled across years, NT increased VWC in the 0-120 soil profile by 7.1% (35.4 to 38.0 cm) and 8.1% (35.3 to 38.2 cm) during sorghum and cotton periods, respectively, compared to CT. No-tillage reduced surface runoff by 12.3 and 17.8% over its conventional counterpart in sorghum and cotton, resulting in an 11 and 14% increase in ET. Cover crop treatments neither impacted soil water storage nor crop water use in both years. Averaged across years, no-tillage reduced the water footprint from 0.51 to 0.45 m3 kg−1 grain of sorghum and 0.78 to 0.73 m3 kg−1 lint of cotton. Results demonstrate NT systems can affect the water footprint of cotton and sorghum production by reducing runoff and increasing soil water storage in the humid subtropical region of the U.S.
How Random are Extreme Streamflow Events?
Year: 2024 Authors: Raczynski K., Grala K., Cartwright J., Dyer J.
Like most natural processes, river flows are characterized by variability over time, which affects regional water resources, associated water quality, and availability. While most natural and man-made systems are able to function well under a normal range in hydrologic conditions, the inherent variability includes the presence of extremes, such as floods or droughts. Extreme events are often considered “random” phenomena in that they cannot be accurately predicted using statistical measures; however, studies conducted on streamflow patterns indicate the presence of certain periodic dependencies affecting the repeatability of extreme occurrences. Factors influencing this behavior include the seasonality of precipitation, the onset and melting of snow cover, or long-term atmospheric circulation. The repeatability is gradually being recognized and partially adapted to the definition of extreme phenomena, like in the case of drought. The aim of this study is to quantitatively assess the extent to which streamflow variability depends on non-random processes manifesting at various scales. To address this objective, daily flow data from 3,135 US Geological Survey (USGS) stream gauges for the period 1970–2023, distributed across the United States, were utilized for analysis. Flows were aggregated into various temporal scales from weekly to annual, after which time series analysis methods such as the Hurst exponent, harmonic analysis, and seasonal and trend decomposition using Loess (STL) were applied to define patterns within the data. The results of this study show a high temporal dependency and spatial correlations within the data, indicating that extreme hydrologic events exhibit tangible non-random patterns associated with physical forcing mechanisms. For annual data, over 70% of all examined gauges exhibit pattern persistence, while on a monthly scale only about 8% show either no clear patterns or anti-persistence behavior. Long-term patterns suggest the presence of processes influencing repeatability from 2 to 7 years, while seasonal patterns exhibit high stability in all studied gauges. Harmonic analysis showed that on a monthly scale, 50 harmonic functions can explain over 80% of the variance, and with an increase in aggregation step, fewer functions are needed to reach high variance percentages (e.g., only eight functions are needed for the annual time series). Among the examined cases, low flows are characterized by a higher influence of recurring processes compared to high flows. Overall results show the existence of non-random and repeatable patterns within extreme hydrologic conditions, indicating a high potential to support long-term streamflow forecasting capabilities. Further work building on these identified patterns could lead to a significant enhancement of the accuracy and scope of predictive models, which in turn can support improved decision-making and water resource management.
Developing machine learning-based pedotransfer function to produce high-resolution soil saturated hydraulic conductivity map
Year: 2024 Authors: Mahmud M.I., Holt R.M., Wodajo L., Hickey C.J., O’Reilly A.M., Davidson G.R.
Saturated hydraulic conductivity (Ks) is one of the most critical soil characteristics that controls the partitioning of precipitation and irrigation water into surface runoff and soil water and regulates water transport in the vadose zone. Determination of Ks is not only challenging but also expensive and impractical for large-scale applications. Pedotransfer functions (PTFs) address this issue by estimating this otherwise difficult-to-obtain parameter using readily available soil information. However, conventional PTFs have shown limited success in accurately predicting Ks. Machine learning-based pedotransfer functions (ML-PTFs) offer a promising solution, consistently outperforming traditional PTFs in Ks prediction. The increasing availability of large soil databases and advancements in machine learning algorithms enhance the potential to improve the robustness of ML-PTFs. This study employed machine learning techniques, such as artificial neural networks, boosted regression trees (BRT), and random forests, with data from over 20,000 soil samples to develop new PTF models for Ks estimation. We evaluated various models and assessed the relative importance of different predictor variables in estimating Ks. Among the evaluated models, decision tree-based ensemble models, particularly BRT, demonstrated superior performance, achieving a root mean square error of 0.336 (log10 Ks[cm/h]) and an R² value of 0.873. Clay content emerged as the most significant predictor, followed by nearly equal contributions from sand content and bulk density. The best-performing ML-PTF model, based on the BRT algorithm, was applied to the surface soil of the 250m gridded SoilGrids dataset, an ML-driven global soil information system based on soil profile data and remote sensing-based covariates, to produce a high-resolution map of Ks across the Mississippi Alluvial Plain (MAP). The regional map of Ks will be helpful in optimal irrigation design, groundwater recharge estimation, and regional hydrological and climate models.
Investigating the correlation of EM38 and Veris apparent electrical conductivity to soil properties of agricultural fields
Year: 2024 Authors: Wodajo L., Locke M., Steinriede, Jr R., Samad M.A., Hickey C.
Soil sample collection from agricultural fields is time-consuming and conducted on coarse grids, possibly missing variabilities between sampling locations. On the other hand, surface-based and non-invasive geophysical measurements, such as electromagnetic induction (EMI) surveys, offer high-resolution spatial maps with detailed field variability and can be conducted at a much higher sampling rate. Although processing soil samples provides the most accurate data on soil properties, establishing correlations between soil's physical and geophysical properties can help reduce the number of soil samples needed and provide a spatial distribution of soil properties on vast agricultural fields. This study involved collecting apparent electrical conductivity (ECa) measurements using the Geonics EM38 and the Veris U series survey instruments on an agricultural field in Sunflower, MS. The farm is approximately 186 acres and part of USDA's Conservation Effects Assessment Project (CEAP) network. The study performed correlation analysis between ECa data from both survey methods and USDA's soil properties data (clay, water content, and soil EC). A comparison of EM38 and Veris ECa maps revealed differences in the magnitude of the measured ECa values, but both methods showed similar spatial distributions in the field. While ECa measurements from both methods exhibited good correlations with measured clay content, the correlation with water content was lower, likely due to variations in the timing and field conditions of the soil water content and the survey assessments. This study emphasizes that applying rapid geophysical methods and establishing correlations with soil properties could significantly enhance soil and crop management decisions. This includes identifying farm field variability, improving soil sampling strategies, and generating spatial variability maps of parameters relevant to plant growth, such as clay content, soil salinity, soil texture, and soil moisture.
Predict the effect of subsurface drainage systems on soil workability in wet spring across the eastern Mississippi
Year: 2024 Authors: Peng R., Feng G., Bi G.
High soil moisture levels frequently occurred across Mississippi during each planting season. In wet springs, timely field operations were often delayed due to poor soil workability, which could negatively impact soil physical properties and structure. Additionally, growers had to apply fertilizers in wet conditions, despite the risk of nutrient loss through leaching and runoff. The objective of this research was to evaluate the effects of various drainage systems on soil workability during wet spring. We simulated subsurface drainage and surface runoff in spring for various combinations of 2 drain depths and drain spacing (5-50 m) in DRAINMOD. Additionally, we calculated the number of work days that can support agricultural vehicle traffic and timely planting based on the moisture in the upper 5 cm soil. The results denoted that shallow 0.75-m drain depth typically provided one extra work days compared to the deep drains, particularly in wet condition. It was recommended to install shallow drains with 10-m spacing to ensure 2 work days in nearly 80% of March and April, and to use shallow drains with narrow 20-m spacing for timely planting in 82% of May. This study offers valuable guidance for designing optimal subsurface drainage systems to ensure timely spring field operations in Mississippi.
Optimizing UAS Bathymetric Sonar Data Collection and Interpolation Methods for Accurate Mapping of Mississippi Waterbodies
Year: 2024 Authors: Bashit M.S., Pricope N.
Rapid developments in bathymetric data collection technology and capabilities, especially miniaturization and mounting on unoccupied aerial systems (UAS)/drone platforms, have revolutionized the collection of 3-D data using active remote sensing sensors. Depth measurements of resources below water surfaces involve significant time, cost, labor, and technology-intensive tasks over various spatial scales that have variable success across the remote, dangerous, turbid, and tannic waters of the state. With a UAS platform, data can be collected quickly and safely, without the need for or with limited need for expensive and time-consuming, ground-based surveys or the use of specialized boats or aircraft. The main objective of our study was to test the relative capabilities of collecting UAS sonar/echo sounder-integrated bathymetric data in different water environments with variable water quality parameters and depths, validated using a suite of ground-based techniques. A secondary objective was to determine the best-performing interpolation method for deriving Digital Elevation Models (DEMs) from UAS echo sounder data collected at varying data survey methods. Finally, our efforts have yielded extremely promising results. We've successfully obtained precise bathymetry data with centimeter-level accuracy as much as 2-meter resolution. These results will lead to time and cost savings through increased inspection capabilities, improved mapping and models of water-logged and water-covered areas, and more robust measurements of drainage system capacities for a variety of monitoring and planning applications. Moreover, UAS echo sounder data can be a useful tool to provide insights for 3-D mapping to bridge the gap between terrestrial and marine surveys as well as between the mapping of specific morphological elements and entire landscapes.
Modeling Farm-Scale Watersheds to Study Impacts of Winter Cover Crops on Water Quantity and Quality of Farms in the Mississippi Delta using Agricultural Policy Environmental Extender
Year: 2024 Authors: Maskey M., Gholson D., Delhom C., Delhom G.D., Nelson A.
Sustainable farming practices and environmental stewardship are receiving increasing attention in the agricultural community, especially in the Mississippi Delta, due to the significant impact of winter cover crops and no-tillage practices on water quantity and quality. Cover crops may have a substantial effect on water quantity and quality by reducing surface runoff, lowering nutrient requirements, and improving infiltration rates. However, the scientific community has faced challenges in evaluating the effects and benefits of cover crops on a farm scale due to limited information on management practices. To address these limitations, we built multiple hydrologic models for paired soybean/corn farms across the Mississippi Delta based on three years of management and yield data, measured surface runoff, and weather data. The primary goals of this study were to a) set up small-scale watershed models (Agricultural Policy Environmental Extender, APEX) for paired fields with available records, b) develop an event-based calibration procedure, and c) investigate changes in agrohydrological attributes such as surface runoff quantity and quality, sediment/erosion, crop yield, and biomass due to cover crop mixes on irrigated and rainfed farms. We also aimed to study the impact of cover crops on deep percolation, which is crucial for sustainable groundwater resources in the Delta. Additionally, we discussed further research on upscaling farm-scale analysis to the watershed level in the Lower Mississippi River Basin to improve best management practices in the Delta farms.
Weaving a web of phosphorus research in the Mississippi Alluvial Plain
Year: 2024 Authors: Witthaus L., Pawlowski E., Locke M., Moore M., Taylor J.
Several recent studies using county-level agronomic datasets have highlighted that the Mississippi Alluvial Plain, i.e. the Delta, is phosphorus (P) deficient – more P is removed through crop harvest than added in reported fertilizer use. Yet, regional environmental research in the Delta has demonstrated that edge-of-field runoff is high in P and the region has streams and lakes not limited by phosphorus availability. Considering more P is exported from Delta systems through harvested crops and runoff, one may hypothesize that Delta soil is rich in P from legacy sources. We tested this hypothesis through various sampling campaigns and several experiments. Our results indicate that P patterns are similar in cropland and forest habitats and are therefore of natural sources due to the depositional history in the alluvial plain. Additional work is underway to understand patterns in groundwater P concentrations, and stream sediment P dynamics across the Delta. Understanding P dynamics and relationships in the Delta will require many threads of research to determine the patterns in soil, sediment, aquatic, and agronomic P pools. This presentation weaves together these research threads from work at the National Sedimentation Laboratory – pulling together data from lake sediment cores, edge-of-field water samples, soils, sediments, and irrigation water samples.
Develop sustainable and resilient management practices and cropping systems to deal with waterlogging in rain season and drought in dry season
Year: 2024 Authors: Feng G., Ouyang Y.
The long-term analysis since 1900 indicated that annual rainfall exceeded 1125 mm in 3 out of 4 yr. The mean monthly precipitation in the crop non-growing season from November to the following April was 127 mm, while only 82 mm on average was received during the crop growing season. Flooding and waterlogging potentially prevent timely tilling, fertilizing and planting in spring. While erratic and inadequate rainfall often result in seasonal drought for agricultural production. Continuous application of commercial inorganic fertilizer has acidified soils, reduced soil organic matter and degraded soil health. In order to deal with those challenges that Mississippian faces, our effort was taken to explore all the opportunities that could provide solutions to those problems. Soil amendments of both agricultural and industry by-products such as poultry litter, biochar, biosolid, FGD, lignite, gypsum and biochar were tested across the state under cover crops, no-till or minimum till in various cropping systems. The purpose was to develop effective management practices for improving soil water holding capacity and enhancing water infiltration, ultimately, for mitigating waterlogging in wet season and water stress in dry season. The soil organic matter (SOM) ranged from 0.9 to 5 %, with a mean value of 3 % for the 167 soil samples we measured. Our results revealed a pronounced effect for silt loam soils with 1 to 28% clay as SOM was larger than 2%. The soils amended with poultry litter had higher SOM (3%) and soil field capacity FC (35%), while the soils without amendment of poultry litter had lower SOM (2%) and FC (30%), a positive effect of SOM on FC was observed. We found that a critical range of SOC for plant available water content from 14 to 18 g kg-1 could significantly improve FC, FC started increasing as SOM was increased over 2%, it is the threshold level of SOM for improving FC. The highest level of SOM that amendment of poultry litter can increase was not greater than 6.0 %. There is large room for SOM improvement in subtropical humid regions. Investigation on no-tilled farm fields with silt loam soil that continuously received 6.7 Mg ha-1 biochar for 10 years in the region found that biochar addition increased FC (26%) and permanent wilting point (43%). Cover crop (CC) can increase soil organic matter by 15% and storage of rain water in soils by 13% during the crop growing season. The combination of manure and gypsum can improve soil FC and infiltration under CC rather than no CC conditions. We found cover crops and soil organic amendments poultry litter and biochar are effective agronomic management practices in mitigating waterlogging and drought in the humid environment.
Exploring the potential incorporation of NHDPlus data in AIMS for hydrological terrain attribute extraction and modeling
Year: 2024 Authors: Sahin A., Rébillout L., Pophet N., Ozeren Y., Al-Hamdan M.
Developed by the University of Mississippi, National Center of Computational Hydroscience and Engineering (NCCHE), and the U.S. Department of Agriculture, Agricultural Research Service (USDA-ARS), The Agricultural Integrated Management System (AIMS) is a web-based decision support tool with automating input data preparation, which utilizes the Annualized Agricultural Nonpoint Source (AnnAGNPS) model for watershed simulations. AnnAGNPS relies on the topographic analysis tool TOPAGNPS to extract terrain attributes. This study explores the potential of incorporating NHDPlus datasets into AIMS and the assimilation of NHDPlus data layers, such as flowlines, catchments, and elevation derivatives as AnnAGNPS input data. This would require the development of algorithms and tools to align and harmonize NHDPlus data with the existing AIMS infrastructure. Developed by the USGS and partners, NHDPlus provides a nationally seamless hydrography framework with resolution and attribute richness compared to other available datasets. Watersheds are delineated for each NHDPlus stream reach, including both left and right banks and the source watersheds for the most upstream reaches. This process involves the manipulation of raster data to split watersheds accurately. Moreover, the NHDPlus dataset lacks LS Factor values crucial for the analysis. These values were computed independently for each sub-watershed. The topological relationships inherent in the NHDPlus reach attributes were then utilized to collect and integrate relevant data for AnnAGNPS simulations. The Upper Pearl River Watershed in Mississippi was used as a regional test case to demonstrate the applicability and performance of this integrated approach.
Utilizing Soil Spectra and Machine Learning techniques to Identify USDA and USCS Soil Classification System
Year: 2024 Authors: Kasaragod A.K., Thomas J., Oommen T., Cole M., Jayakumar P.
Soil texture, a fundamental soil physical property, is a vital input for various scientific and engineering applications, including agriculture. The conventional methods to analyze soil texture, especially for large spatial coverage, heavily rely on field and/or laboratory measurements that are often time-consuming and lead to destructive use of the collected soil samples. Remote sensing methods, however, are a reliable and quick alternative to estimate and predict various soil physical and chemical properties. Remote sensing data can be collected from various platforms focusing on a wide range of electromagnetic spectrums that reflect variability of soil properties at different wavelengths at different spatial, spectral, and temporal resolutions. Advanced machine learning and deep learning techniques enable the extraction of complex, interrelated information from high-dimension remote sensing data. The wavelength range from 400 to 2500 nm is often utilized for soil texture identification tasks as these ranges best reflect the variations in soil physical and chemical properties. The soil spectral libraries (SSL) consist of high spectral resolution soil spectra that can be utilized to estimate soil texture. The openly available SSLs, such as the Kellog Soil Spectra Library (KSSL) and the Open Soil Spectra Library (OSSL), include multiple SSLs across the globe and offer many opportunities to estimate soil texture classes with high accuracy. This study uses the openly available SSLs and three deep learning architectures (VGG-16, ResNet-16, and Swin transformers) to predict USDA (United States Department of Agriculture) and USCS (Unified Soil Classification System) soil texture classes. The SSLs consist of data from 400 to 2500 nm wavelength range with 2nm spectral resolution along with clay, sand, and silt fractions calculated for each soil spectra sample. Different pre-processing and data transformation methods are employed to highlight better the various spectral features reflecting soil texture variability. The results demonstrate promising usability in employing the SSL data and deep learning techniques for soil texture prediction tasks. Additionally, results of this study have potential to be used with satellite/UAV remote sensing hyperspectral data aiming at predicting soil texture classes with better spatial coverage and resolution.
Corn Grain Yield and Nitrogen Uptake Efficiency as Affected by Nitrogen Application Methods
Year: 2024 Authors: Hutton M., Reynolds Z., Spencer D., Krutz J.
Nitrogen (N) fertilizer inputs are annually applied to 98% of the 291,500 Mississippi corn hectares, comprise 36% of farm production costs, and are a significant source of environmental pollution. This research was conducted to determine whether manipulating nitrogen application could maintain or improve productivity and profitability while mitigating adverse environmental outcomes. The effects of N placement, depth, rate, and closing the coulter injection trench on corn [Zea mays (L.) Moench] productivity, profitability, and runoff N losses were investigated at Tchula, MS on a Dubbs silt loam (fine-silty, mixed, active, thermic Typic Hapludalfs) and Brooksville, MS on a Brooksville silty clay (fine, smectitic, thermic Aquic Hapluderts). After the first year, closing the injection trench increased corn grain yield 377 kg ha-1 relative to leaving the injection trench open (p = 0.0623). N rate, application method, lateral knife placement, and injection depth appear to have no effect on grain yield (p ≥ 0.1861). This research will be continued in 2024 and additional investigations will be conducted to study whether altering N application methods can reduce off-site loss of N or improve profitability. Preliminary results suggest that protecting N by closing the application trench may reduce N loss enough to increase plant uptake and subsequent yield.
Delineation of Groundwater Stress Zones in Coastal Lowland Aquifers Using Downscaled GRACE Satellite Data
Year: 2024 Authors: Jahan M.N., Yarbrough L., Easson G., Ghaffari Z., Yasarer H.
The Coastal Lowland Aquifers of Texas, Louisiana, Mississippi, Alabama and Florida is a vital water source, supplying approximately one billion gallons daily for public and private use. However, extensive withdrawals have caused significant issues, including saltwater intrusion and land subsidence in certain areas. This study applied downscaled GRACE (Gravity Recovery and Climate Experiment) satellite data of 2003 and 2023 to evaluate Groundwater Storage (GWS) and identify the potential groundwater stress zones over this period. A Random Forest Model (RFM) was used to generate monthly 4-km Groundwater Storage Anomaly (GWSA) maps. The model incorporated variables such as monthly anomaly of total precipitation, mean temperature, normalized differences of vegetation Index (NDVI), evapotranspiration for 2003 and 2023; Shuttle Radar Topography Mission (SRTM) DEM; slope angle; dominant soil type, and lithology to downscale the monthly Total Water Storage Anomaly (TWSA) by GRACE/GRACE-FO of 2003 and 2023 from 3 degrees (~333 km) to 4 km resolution. The RFM was then reapplied with all the parameters, including the downscaled monthly TWSA, to further downscale the monthly GWSA for 2003 and 2023. This GWSA was derived by subtracting the sum of the monthly anomalies of Root Zone Soil Moisture (RZSM), Plant Canopy Surface Water (PCSW), and Snow Depth Water Equivalent (SWE) from the monthly TWSA using data from Global Land Data Assimilation System (GLDAS). Validation of the downscaled monthly GWSA showed high R2 values (0.85 to 0.98) and low RMSE values, indicating accurate predictions. Spatial analysis of the downscaled GWSA maps for 2003 and 2023 revealed that the western part of the area is more depleted in GWS compared to the eastern part. These findings are consistent with the groundwater level data from USGS for 2003 and 2023, which also showed excessive groundwater depletion in most of the western part of the area, especially in Baton Rouge, Louisiana, and Houston, Texas. This study provides valuable insights for stakeholders to develop effective groundwater management plans before undertaking any regional groundwater development activities.
Rural Hazard Resilience Tools: Bridging Hydrology, Geospatial Technology and Citizen Science to Enhance Flood Disaster Resilience of Rural Communities in Data-Scarce Regions of the US
Year: 2024 Authors: Thomas J., Mohan S., Oommen T., Xue P., Meadows G.
Flooding associated with extreme rainfall events and wind-driven storm surges has devastating consequences on societies due to their disaster potential in terms of human casualties, as well as environmental and economic impacts. Among various natural disasters, coastal and fluvial flooding presents a formidable challenge to rural livelihood across the Great Lakes region, including Indigenous and post-industrial communities. While the Federal Emergency Management Agency (FEMA) flood map is the primary source of flood risk information in the United States, many rural counties, including those in the Upper Peninsula (in Michigan), are still unmapped. Typically, flood risk models incorporate a model chain that includes hydrological models or frequency analysis for estimating flood discharge magnitude, hydrodynamic models for mapping flood inundation, and damage (vulnerability) functions for risk assessment. Despite the low disaster resilience of rural communities in the Great Lakes region to flooding, flood mitigation efforts have been hindered by the lack of data and appropriate tools for understanding the flood risks. This study introduces the development of the Rural Hazard Resilience Tools (RHRT), a suite of tools to help rural communities in the western Upper Peninsula improve their resilience and adaptation to flooding and coastal disasters. The RHRT includes a tool for assessing flood risk, a web application for collecting crowdsourced data, including photographs of flood events, and a geospatial platform to visualize the flood risk with critical infrastructure and community resilience indicators. The flood risk assessment tool utilizes a simplified modeling approach with the Height Above the Nearest Drainage (HAND) model and synthetic rating curves (SRC) for approximate flood inundation mapping with minimal input data (i.e., DEM) and computational resources, the Simulating WAves Nearshore (SWAN) model for coastal flood inundation modeling, USGS regional regression equations for estimating peak discharge, and depth-damage functions of the HAZUS-MH flood model for estimating losses due to building-level impacts. The crowdsourced data collection application is an example of citizen science engagement that allows the community members and public to upload photos of flood inundation during an event to address current data gaps, improve flood hazard modeling by validating the HAND-based flood models, and enhance automated flood information extraction using machine learning algorithms. The geospatial visualization platform is designed to disseminate flood risk information to rural communities and to help the decision-making authorities, emergency service professionals, and other community leaders make more informed decisions for flood risk mitigation. The methodological framework of the RHRT is cost-effective, less resource-intensive, and easy to implement, making these tools transferable across multiple spatial scales.
Effect of evapotranspiration data sources on runoff and erosion simulation results for the Pelahatchie Bay watershed
Year: 2024 Authors: Shuchana I., Rébillout L., Al-Hamdan M., Ozeren Y., Bingner R.
The impact of utilizing various evapotranspiration (ET) sources for simulations of runoff and erosion from Pelahatchie Bay watershed in Mississippi was studied using the Annualized Agricultural Non-Point Source (AnnAGNPS) watershed model for the period between 2010 and 2019. The ET datasets considered were from the remote sensing Terra MODIS product and the NOAH Land Surface Model from the North American Land Data Assimilation System (NLDAS-2). The Terra MODIS product, derived from the Moderate Resolution Imaging Spectroradiometer (MODIS) on the Terra satellite, provides global ET data every eight days based on satellite observations. The NOAH model uses hourly ET data from NLDAS-2, which also relies on remotely sensed data for ET estimation. Post-processing was performed on both datasets to obtain daily ET values for the simulations. When ET data was not directly provided, the Penman-Monteith equation within AnnAGNPS was utilized to determine ET, resulting in values 15% greater than NOAH and 28% greater than the MODIS remotely sensed ET products. The results showed that when NOAH or MODIS daily time-series ET data were used instead of AnnAGNPS-computed daily ET values, simulations of total streamflow, direct runoff, and sediment yield were reduced. Comparisons with the AnnAGNPS streamflow simulation results using each ET source compared to measured data from an USGS station near the watershed outlet will highlight the impact of using the best source of ET data available for watershed management planning. The findings emphasize the impact of ET data choice on watershed modeling outcomes and the detailed results of these analyses will be presented in this conference.
Tracking groundwater changes with GRACE-FO during irrigation season
Year: 2024 Authors: Ghaffari Z., Awawdeh A.R., Yarbrough L.D., Easson G.
Groundwater, the largest accessible source of fresh water globally, supplies 63% of U.S. freshwater for irrigation. In the U.S., 40% of the water supply and over 40% of irrigation water rely on groundwater. This resource is more reliable than precipitation and surface water, but its increased use has led to significant depletion. Arkansas, which accounted for 49% of U.S. rice production and 49.6% of planted acres in 2023, ranks third in irrigated land area and second in irrigation water volume. Adequate water supply is crucial for sustaining agricultural production, necessitating in-situ groundwater monitoring. High-resolution, continuous hydrological products are vital for sustainable water management and predicting water-related patterns under climate change and human impacts. The Gravity Recovery and Climate Experiment (GRACE) mission, launched in 2002 and succeeded by GRACE-FO (Follow-on), revolutionized hydrological monitoring by measuring Earth's gravity field variations and providing critical data on terrestrial water storage (TWS). Despite its coarse spatial (3°) and temporal (monthly) resolutions, GRACE has been essential for studying groundwater storage changes and supporting long-term water management strategies. However, there is a growing need for finer resolution data at local scales. This research uses a Random Forest model to downscale GRACE mascon data from a 0.5° grid to a 1 km spatial resolution. Our goal is to track groundwater storage (GWS) changes with GRACE-FO during irrigation seasons to determine: 1) Is GRACE capable of tracking water level changes during these seasons? 2) If so, what is the time lag in GRACE data reflecting these changes? The findings could significantly benefit water resources management, particularly in agriculture.
The development of long-term mean annual total nitrogen and total phosphorus load models for Mississippi, U.S. using RSPARROW
Year: 2024 Authors: Gain E., Roland V.
The effective management of nutrient pollution in rivers and streams often requires not only a robust water-quality monitoring program, but also an understanding of the sources of nutrients and the factors that influence their delivery to water bodies across the landscape. Because it is only feasible to monitor a small percentage of streams, model-based approaches are often utilized for the management of unmonitored watersheds. SPARROW (SPAtially Referenced Regression On Watershed attributes), a hybrid statistical and process-based mass balance model, uses nonlinear least-squares regression to correlate water quality observations with sources and transport-related properties of the watershed. SPARROW models have often been used at regional and national scales, and across broad spatial, geographic, and climate scales. A large domain affords the inclusion of more calibration sites which allows for the inclusion of more explanatory variables in the final model. However, larger calibration domains may result in predictions that are biased for a particular subregion. Models calibrated for a specific state may offer state resource managers more accurate estimates of nutrient loads for the watersheds that they manage. In cooperation with the Mississippi Department of Environmental Quality (MDEQ), we used load estimates for 46 nitrogen calibration sites and 51 phosphorus calibration sites to develop 2018 base year nitrogen and phosphorus models for the state of Mississippi. These models provide source apportioned estimates of nitrogen and phosphorus loads, yields, and concentrations for all catchments within the state.
A water quality study of the upper Pearl River Watershed in Mississippi using a web-based management system
Year: 2024 Authors: Pophet N., Rébillout L., Ozeren Y., Chao X., Al-Hamdan M.
Covering 7,588 square kilometers, the Upper Pearl River Watershed feeds into the Ross-Barnett Reservoir, one of Mississippi's largest surface water storage areas and the primary water source for approximately 200,000 people in Jackson and the surrounding areas. The Upper Pearl River Watershed faces significant water quality issues due to agricultural runoff, urbanization, and other nonpoint source pollutants, requiring a comprehensive study to understand and mitigate these impacts. However, watershed modeling study presents several challenges, including extensive input data preparation and computational resource limitations, which can hinder effective model execution for large watersheds. This study utilizes input data prepared by the Agricultural Integrated Management System (AIMS)—a web-based system developed by The National Center for Computational Hydroscience and Engineering (NCCHE) and the USDA-ARS-National Sedimentation Laboratory—to run watershed simulations with the Annualized Agricultural Nonpoint Source Model (AnnAGNPS) for the Upper Pearl River Watershed from 2010 to 2022. The objectives of the study are to (i) assess the water quality discharge from the Upper Pearl River Watershed to the Ross-Barnett Reservoir, (ii) demonstrate the capability of AIMS to automatically prepare input data such as channel networks, soil, climate, and management data for watershed modeling, and (iii) test the concept of parallelization in running AnnAGNPS for large-scale watersheds like the Upper Pearl River Watershed. Preliminary results are presented, including total streamflow, sediment, and nutrient discharged from the watershed outlet to the reservoir. Simulated total streamflow results are compared with data from seven USGS stream gauge stations within the watershed. Additionally, the speedup achieved through the parallelization of the AnnAGNPS model is highlighted, demonstrating the potential for more efficient watershed modeling.
Electromagnetic Induction for Rapid Estimation of Infiltration Rates
Year: 2024 Authors: Samad M.A., Hickey C.J., Wodajo L., Surbeck C.Q., D’Alessio M.
Understanding the ground surface infiltration rate is crucial for managing and optimizing the interaction between water and land, contributing to sustainable environmental development, management, and protection. However, conventional infiltration rate measurement techniques, such as double-ring infiltrometer (DRI), well permeameter, and infiltration basin percolation tests, are time-intensive and limited to spatial resolution. This research introduces a simplified and time-efficient infiltration estimating technique by incorporating high-resolution electromagnetic induction (EMI) measurements and hydrogeological methods. An experiment was conducted on a 622.43 m² infiltration-detention basin at the University of Mississippi instrumented with two inlet and one outlet pipes. An empirical relation between electrical conductivity and infiltration measurements from the in-situ DRI tests is established for a limited number of points. The high spatial resolution of infiltration rates is then estimated using the established empirical relation, and leveraging the Green-Ampt model with the modified Newton-Raphson method from high-resolution EMI measurements. The geophysics-aided estimated infiltration result was evaluated using a water balance model. A comparison of infiltration rate maps generated by interpolating the DRI and geophysics-aided measurement showed that the geophysics-aided infiltration rate map has a more robust prediction. At steady state conditions, the geophysics-aided mean infiltration rate of ~2.85 cm/h closely conforms with the water balance model mean infiltration rate of ~2.90 cm/h. This outcome demonstrates that geophysics-aided measurements can be implemented to expedite estimation of infiltration properties with high spatial resolution.
Effects of Off-Season Fall-Winter Crop Field Flooding on Nitrous Oxide Production
Year: 2024 Authors: Rosson A., Hoeksema J., Taylor J., Blocker V., Counce E.
Increased nitrogen (N) fertilizer use and cultivation of N-fixing crops has dramatically increased agricultural N inputs to the global N cycle. Nitrous oxide (N2O) has a warming potential ~300x greater than carbon dioxide (CO2) and enters the atmosphere primarily through denitrification which converts bioavailable nitrate (NO3-) to inert nitrogen gas (N2) with N2O as an intermediate step in the process. Wetlands are important habitats for denitrification and N regulation, being efficient at N removal and uptake with their wet anoxic soils, but a large proportion of natural wetlands have been lost worldwide. In the Mississippi Alluvial Valley (MAV), shallow habitat management on fallow crop fields may provide some nutrient regulation and wildlife habitat services previously provided by natural wetlands. Crop fields in the Mississippi Delta are being flooded post harvest, primarily for migratory shorebird habitat, and work has been done to characterize influences on denitrification; however, little is known about N2O production relative to N2 associated with this practice. We measured N2 and N2O production from flooded fields throughout fall and winter of 2023. We observed significant interactions between temperature and flood duration, showing increasing N2O flux with decreasing temperature when accounting for flood duration. Additionally we conducted an experiment once in both fall and winter to estimate the influence of inundation gradients on N2O emissions from laboratory flooded sediment cores. In the fall, we observed a significant interaction between field soil flood status and time flooded in the laboratory on N2O flux. Cores gathered from the submerged zone (negative flux) were different from cores gathered from the dry and intermediary zones (positive flux) in the first ~8-12 hours after laboratory flooding. All zones N2O fluxes converged around zero in the ~24 hour and final ~32 hour sampling times. N2 flux was only affected by depth, being higher in the dry zone than the submerged zone, but not the intermediate. In the winter N2O flux was significantly more negative for the ~24 hour sampling than the other two time points across all depths . N2 flux had more positive flux rates in both unsubmerged zones compared to the submerged zone. Further results on soil and dissolved nutrients will be presented. These results will help determine ecosystem services and disservices associated with shallow water management for migratory shorebird habitat in MAV agroecosystems.
Drought Impacts on Hydrologic Annual Residence Time and Sediment Nitrogen Concentrations in Reservoirs: Findings from Sediment Records
Year: 2024 Authors: Webster B., Waters M., Golladay S.
Reservoirs as the receivers of suspended materials from rivers and watersheds are hotspots for sediment and nutrient deposition. The phosphorus dynamics, including deposition, has been widely considered but sediment nitrogen deposition has received less attention at the watershed scale. In this study, we compared average annual hydrologic residence time along with other physical parameters to the sediment nitrogen concentrations for eight reservoirs in the Southeastern USA. Residence time is determined by water storage in a reservoir and dam release rate and can be linked to dynamic processes like trap efficiency. The eight reservoirs varied regarding residence time, surface area, water volume, land use, catchment area, primary usage (hydropower and storage) and other parameters. Using paleolimnological techniques, long-term reservoir flow data, federal long-term drought indices, and GIS tools, average annual residence time was found to have the strongest relationship with sediment nitrogen concentrations (R2 = 0.79) when compared to all other reservoir parameters. Residence time drives nitrogen deposition by allowing longer periods of algal growth followed by deposition of nitrogen in particulate organic form. Photosynthetic pigments supported this inference and identified cyanobacteria being the primary producer group most related to residence time (R2 = 0.73) followed by diatoms (R2 = 0.56). In periods of drought, basin managers are forced to abide by critical yields and base flow regulations to maintain reservoir water storage thus increasing water retention times. Following our sediment and residence time based model, during these extreme long-term drought periods residence time increased by 45 – 60% increasing nitrogen delivery to the sediments by roughly 2.5 – 4%.
Design subsurface drainage systems to optimize crop production in RAINFED agricultural fields across Mississippi
Year: 2024 Authors: Peng R., Feng G., Bi G.
Soybean is the most important crop in MS, covering 2.13 million harvested acres valued at $1.60 billion in 2023. The majority of the precipitation in MS occurred in fallow season, particularly in wet March and April. Excessive wetness and waterlogging during this period prevented timely tilling, planting, and caused serious excess water stress in the early stage of crop growth, potentially impacting the yield. The objective of this research was to evaluate the effect of drainage system on crop yield and determine optimum planting window under various drainage systems for the dominant soil and local weather conditions. We conducted various simulations for combinations of 2 drain depths and drain spacing (5-50 m) in DRAINMOD. These simulations calculated daily soil moisture at different soil depths in wet spring. Moreover, we identified appropriate drainage systems to ensure planting based on required workable days. Furthermore, the relative crop yield was simulated by setting up planting dates to determine the optimal date under various drainage systems. This study will provide valuable guidance for farmers to determine windows of spring fieldwork in MS.
Impact of cover crops and field conservation on water runoff quantity and quality in cotton production
Year: 2024 Authors: Pilgram W., Gholson D., Nelson A., Locke M., Simpson A.
Agricultural production in the mid-southern area of the United States can generate large amounts of water runoff from furrow irrigation. A field study was established in 2019 to examine the effects of cover crops and conservation practices on runoff water quality and quantity in cotton (Gossypium hirsutum L.). A randomized complete block design with three blocks and seven treatments was arranged on 2.5 hectares. No-tillage (NT); NT with cover crop (CC); NT with CC and subsoil (SS); Strip tillage (ST); ST with CC; ST with CC and SS; and reduced tillage (RT, SS) were evaluated on a Dubb silt loam (fine-silty, mixed, active, thermic Typic Hapludalfs) and Bosket very fine sandy loam (fine-loamy, mixed, active, thermic Mollic Hapludalfs). The site consists of 21 plots that are 150 m long by eight rows spaced 1 m apart with a 1.8 m -wide earthen levee to separate each plot. Plots were equipped with automated surface runoff sampling equipment. Composite runoff samples were collected during runoff events to determine sediment and nutrient (N and P) losses. The objective of this study is to examine the effects of different tillage methods and cover crops on runoff water quantity and quality in cotton. Preliminary results indicate that total nitrogen and phosphorus concentrations were highest in the NT-CC and ST-CC treatments, especially earlier in the growing season, but total phosphorus was higher later in the summer months in RT-SS treatments. Total solids also peaked in the later summer months and were highest in NT-CC and ST-CC plots. Additional concentration and load results will be presented.
Investigating the tree bark microbiome of bald cypress (Taxodium distichum) in the Yazoo-Mississippi Delta as a function of hydrologic and water quality dynamics
Year: 2024 Authors: Barrett D., Heintzman L., Davidson G., Jackson C., Moore M.
Bald cypress (Taxodium distichum) is a deciduous conifer tree, endemic to the southeastern United States, where it grows in continuously inundated wetlands, swamps, and oxbow lakes. Within the Yazoo-Mississippi Delta (YMD), bald cypress often occurs near agricultural lands, where they experience variable hydrology via natural and anthropogenic causes (precipitation, irrigation, drainage, etc.). Consequently, these trees are subject to dynamic agrochemical inputs and exposed to eutrophication events via runoff. Ecosystem services, such as nutrient cycling, are driven by microbial communities present in aquatic systems. Within YMD and other water bodies, bald cypress barks provide a large, and seasonally stable, surface area for microbial communities to form. While ecological effects of hydrology and water quality are well documented with respect to microbial communities within the water column, few studies examine the effects of these factors on bald cypress microbial communities. As such, we sought to understand how the microbiome of bald cypress bark responds to altered hydrology and water quality. We collected bald cypress bark samples (n=282) over a six-month timespan (Nov 2023 – May 2024) from 18 trees located in three oxbow lakes (six trees per site) in the YMD, representing a continuum of regional production and management. From each tree, bark was sampled at: a) ~80 cm. above the water line; b) in the “splash zone” (a zone subject to waves) which we defined as the ~20 cm. above the water line; and c) ~20 cm. below the water line. Preliminary results indicate a vertical gradient in the tree bark microbiome. The highest amounts of DNA recovered were from below the water line, with decreasing amounts moving upward. After Illumina 16S amplicon sequencing, community composition will be correlated with hydrology and water quality metrics (e.g. DO, nitrogen content, phosphorus content, and turbidity), to determine primary abiotic drivers of the bald cypress bark microbiome within the YMD. Results are also expected to enhance nutrient cycling models and may provide insight into potential biodegradative properties.
Improving Water Availability Modeling Capacity Through the Development of a Data Management System and Automated Workflow to Develop Geologic Frameworks
Year: 2024 Authors: Bolton W., Kymes M., Hoogenboom B., Duncan L., Kress W.
An accurate geologic framework is essential for groundwater flow modeling, as it defines the structures that influence recharge, groundwater-surface water interactions, and subsurface water movement. Traditionally, groundwater flow models have relied on geologic surfaces derived from borehole data to create a layered geologic framework. In these models, grid cells with similar hydraulic properties are grouped into zones and assigned specific parameter values. Initial parameter values are typically based on literature and adjusted during model calibration. However, these frameworks and the associated data often lack a centralized database for future use, leading to redundant efforts when developing new frameworks, especially as new data emerges. Growing demand for water availability studies and complex engineering scenarios necessitates a flexible, integrated system capable of managing extensive datasets collected over decades from various mapping campaigns and from diverse geophysical methods and instruments. To address these needs, the U.S. Geological Survey (USGS) and the Mississippi Department of Environmental Quality (MDEQ) are collaborating to improve the geologic framework of the Mississippi Embayment (ME). Currently, the project has compiled 26,000 picks of geologic formation tops from 3,355 boreholes using 22 data sources from various USGS and state publications and log archives into a comprehensive database. These geologic formation picks were used to guide the interpretation of 35,000 line-kilometers of Airborne Electromagnetic (AEM) data to further refine the geologic surfaces. An integrated 3D geologic framework was created through automated, reproducible workflows. The workflow is also capable of producing 2D and 3D visualizations, detecting anomalies, and producing rasterized surfaces of geologic formation tops. These automated outputs can be readily utilized as input for a groundwater flow model and are scalable to various resolutions, supporting regional, sub-regional, and local-scale applications. Furthermore, this workflow simplifies the updating process of the geologic framework as new data are acquired. The long-term plan is to develop a web-enabled platform that would allow government agencies and academic institutions to upload, cross-check, and publicly share critical geologic datasets.
Improving Flood Delineation using Sentinel-1 SAR by Combining Histogram Thresholding and Digital Elevation Model in Earth Engine
Year: 2024 Authors: Thalakkottukara N.T., Thomas J., Oommen T.
A warming climate leading to intensification of hydrological cycle may increase extreme rainfall events, worsening riverine flooding with loss of life and significant economic damage. Accurately delineating these floods' extent and depth is essential for effectively mitigating their impacts. Synthetic Aperture Radar (SAR) imagery is often favored over optical images for flood inundation mapping, as microwaves can penetrate clouds and function day and night. Sentinel-1 is a commonly used SAR sensor due to the availability of historical data in Google Earth Engine (GEE). However, SAR images often struggle to differentiate between flooded and non-flooded areas, leading to higher misclassification rates, particularly in urban and vegetated areas. Traditional methods like Otsu's histogram thresholding can effectively discriminate open water with low backscatter values from other land cover types. However, they struggle with flooded urban areas and smooth surfaces like roads and runways. Machine learning algorithms offer better classification accuracy at the cost of extensive training data and ground truth, which can sometimes be challenging and labor-intensive. The Height Above Nearest Drainage (HAND) technique has proven effective for flood delineation and risk mapping. Still, it requires hydro-proofing of Digital Elevation Models (DEMs) and cannot be calculated on the fly in GEE due to restrictions on iterative processes. To address this, we developed a Height Above Lowest Elevation (HALE) model in GEE, which dynamically calculates the HALE of each pixel in the DEM. This is achieved with the help of the Global Surface Water dataset available in GEE to determine the lowest elevation points. We can delineate the flood extents and estimate flood depths by integrating HALE with binary classification results from Otsu thresholding. A manual threshold for HALE provides the flood extent, and subtracting the maximum HALE value from this image generates the flood depth. This method can be applied to other SAR sensors, though further testing is needed. The current algorithm is most effective for river channels with low bed slopes and may not perform as well in steep terrains. Future work includes automating the HALE thresholding process and enhancing the algorithm for better performance in steep channels. This approach provides a cost-effective solution for improving flood delineation and mapping, leveraging freely available SAR data and advanced DEM processing in GEE.
Nutrient and Sediment Reductions Associated With Cover Crop - Minimum Tillage Best Management Practices
Year: 2024 Authors: Hill M., Baker B., Badon T., Ramirez-Avila J., Evans K.
Nutrient enrichment from agricultural landscapes to receiving waterbodies is a contributing factor to habitat degradation worldwide. Excess fertilizer runoff in the upper and lower Mississippi River Basin is a primary driver of the large seasonal hypoxic zone in the Gulf of Mexico. This study examined the impact of combined agricultural best management practices, cover crop and reduced tillage, on water quality indicators, specifically nutrient and sediment loading from row-crop fields in western Mississippi. A paired-field study was established with edge-of-field water monitoring stations to measure discharge and collect runoff samples. The treatment was compared to a conventional farmer management control. Experimental sites were established at eleven working farms and operated over the course of three years (n=134). Assessed water quality indicators were total suspended solids, turbidity, nitrate-nitrite nitrogen, total Kjeldahl nitrogen, total nitrogen, orthophosphate, and total phosphorus. Relative reductions were calculated and analysis of paired water quality samples was performed using the Wilcoxon signed-rank test. Increased runoff volume was observed (p=0.024), which when coupled with modest decreases in total nitrogen, suspended solids, and phosphorus concentrations, resulted in no net change in transport (p>0.1). Nitrate-nitrite N concentrations and loads were significantly reduced (p<0.01, p=0.03). Use of cover crop-reduced tillage practices may help assuage nutrient enrichment concerns to aquatic ecosystems, particularly when considering exogenous N fertilizers. However, said practices may be detrimental to P limited aquatic ecosystems through increases to orthophosphate transport (p<0.01). On-farm management practices can have a definite impact on non-point source water pollution from fields in the study region. Further investigation is warranted to delve into the mechanisms underlying the observed increase in discharge associated with the treatment.
New Capabilities of DSS-WISE Web, A Web-Based, Automated Flood Inundation Modeling Decision Support System
Year: 2024 Authors: McGrath M., Al-Hamdan M., Pophet N., Smith P.
The National Center for Computational Hydroscience and Engineering (NCCHE) at the University of Mississippi has been developing and operating the Decision Support System for Water Infrastructural Security (DSS-WISE Web) since November 2016. With funding from the FEMA National Dam Safety Program, DHS S&T, and others, this simple and easy-to-use tool has been relied upon by over 2,100 users from the dam safety community to submit over 80,000 dam-break flood inundation simulations. Most simulations can be set up in under 5 minutes, and detailed, GIS-compatible results are available within 30 minutes for 85% of cases. This tool has been designed to allow users to get a first-level analysis of flood inundation, flood arrival time, and Human Consequences (HCOM) with Population At Risk (PAR) in various hazard categories for prioritization and screening. It needs only a minimum set of inputs without requiring users to obtain expensive software, servers, or numerical modeling expertise. The system uses an automated data preparation procedure to input into a verified and validated numerical model running behind a secure web-based portal for setup and results access and download. The web-based interface for the DSS-WISE Web model was recently updated to version 3.0, which brought several new and improved capabilities. These features include the ability to model dams in series, the ability to model levees for terrain modification, a breach parameter calculator, and improved terrain elevation visualization and querying, among others, which will be showcased as a poster presentation.
Detecting Water-Quality Improvements from Nutrient Reduction Strategies in Delta Streams
Year: 2024 Authors: DeVilbiss S., Hicks M., Gain E.
Conversion of large alluvial plain landscapes to intensive row crop agriculture has caused considerable habitat and water-quality alterations in Mississippi Delta streams. Numerous nutrient reduction best management practices have been implemented in two pilot watersheds in the Mississippi Delta, Harris and Porter Bayou, in efforts to minimize the impact of nutrient enrichment due to agricultural practices and improve water quality. However, given the long history of stream alteration, nutrient enrichment, and high spatial and temporal variability of weather patterns, detecting water-quality improvements from management efforts has proven challenging. To overcome these challenges, we analyzed long-term (10 year) post-restoration water-quality data in the two pilot watersheds using a combination of generalized additive modeling (GAMs) and weighted regression on time, discharge, and season (WRTDS). By accounting for non-linear relationships between nutrient concentrations, discharge, year, and season, GAMs detected significant reductions in nutrient concentrations over time, but reductions were highly seasonal. Nutrient load estimates using WRTDS indicate that nutrient loading in both study watersheds have also decreased substantially since restoration practices were implemented. Collectively, our results demonstrate that nutrient reduction strategies are effective in Delta watersheds. Further, flexible analytical approaches that account for the high degree of nutrient and discharge variability typical of agricultural watersheds can provide a deeper understanding of how nutrient enrichment responds to best management practices.
Investigating Overtopping Failures in Earthen Levees Using Anura3D
Year: 2024 Authors: Dumlu E., Ozeren Y., Rébillout L., Al-Hamdan M., Al-Ostaz A.
The state of Mississippi has around 1,000 miles of earthen embankment levees and more than 6,000 earthen dams. More than half of these dams and levees are older than 50 years and are at risk of failure. One of the most common types of levee failures is due to overtopping. This study performed a series of overtopping simulations to investigate the potential use of the open-source geotechnical software Anura3D. The levee was constantly considered saturated and cohesive and had a height of 0.2 m, a crest width of 0.15 m, and a slope of 1V:3H. To control the inflow discharge, an artificial reservoir is installed upstream of the levee and initial reservoir elevations were set at 0.25, 0.30, 0.35, 0.40, and 0.55 m. The gravity values were also varied for some experiments to investigate the scaling effects. During the simulations, discharges were obtained for various locations and compared to each other. Those dimensional values were non-dimensionalized to establish a relationship with Froude similarities. A similar relationship was investigated for the levee's erosion rates for each simulation. Preliminary results showed that Froude scaling is well suited to match the discharge values when the initial reservoir levels were set to 0.30, 0.35, and 0.40 m. Also, increasing the acceleration of gravity resulted in higher discharge values and erosion rates. The implications of these findings, along with potential future implementations of the simulations will be also discussed during the presentation.
Improving Water Availability Modeling Capacity Through the Collection of Aquaculture and Irrigation Water Use Data
Year: 2024 Authors: Stocks S., White V., Duncan L., Kress W.
Traditionally, groundwater flow models have used United States Geological Survey (USGS) water use reports to estimate pumping for aquaculture and irrigation. These reports were compiled every five years since 1950 and provided an annual water withdrawal estimate at the county level for the report's year of study. When used as inputs to groundwater flow models, water use for intervening years need to be estimated. Since water use can vary significantly from year to year depending on weather, this leads to higher uncertainties during the intervening years. The introduction of flowmeters on aquaculture and irrigation wells through the Mississippi Voluntary Metering Program (VMP) administered by Mississippi Department of Environmental Quality (MDEQ) in 2014 has greatly improved the accuracy of water use estimates. The USGS Aquaculture Irrigation Water Use model (AIWUM 2.0) utilizes the annual self-reported water use data from the VMP as input to a Distributed Random Forest Machine Learning algorithm to predict annual water use values by major commodity. Additionally, AIWUM 2.0 uses continuous flowmeter water use data provide the basis for a monthly distribution percentage calculation to estimate the monthly breakdown of the annual water-use predictions. These data have been used by supporting models to provide gridded aquaculture and irrigation water use data for the Delta groundwater flow model. Under the VMP, annual reports of metered water use for 10% or more of the permitted Mississippi River Valley Alluvial (MRVA) aquifer wells in each county within the Delta Region are submitted to MDEQ. In 2016, the USGS Mississippi Alluvial Plain (MAP) project enhanced this water use network by adding data collection platforms to flowmeters at 25 locations on sites with commodities of corn, soybeans, rice, catfish, and wildlife. In 2020 MDEQ expanded the network of continuous sites to include data collection platforms on an additional 23 flowmeter sites through a grant from the USGS Water-Use Data and Research program (WUDR). These locations collect continuous water use data on sites with soybeans, catfish, and waterfowl irrigation data and continue to be operated through the support of Delta Farmers Advocating Resource Management (F.A.R.M). In 2022 and 2024, the USGS MAP project added rain gages to 18 continuous flowmeter water use sites to improve understanding of weather patterns on irrigation scheduling. Annual data provided by the VMP network combined with data from continuous water use sites are supporting higher spatial and temporal resolutions in groundwater flow models for water availability modeling in the Mississippi Delta region.
Life After the Flood: An Automated, Low-Water-Use Rice Production System
Year: 2024 Authors: Oakley G., Spencer D., Gholson D., Krutz J.
Groundwater resources are being depleted by irrigation practices in the mid-southern US. This study was conducted to determine whether automated technologies and low-water-use techniques can be coupled to reduce water use in the drill-seeded, delayed flood rice (Oryza sativa L.) production system. The effects of autonomously managing the intermittent flood depth on water use, grain yield, and rice milling quality were investigated at Shaw, MS on a Sharkey clay (very-fine, smetitic, thermic Chromic Epiaquerts). Automating an intermittent flood decreased the volume of water applied 23.5% when compared to a continuous flood. Rice grain yield did not differ between a continuous and intermittent flood; however, a 4% yield reduction was observed in the bottom paddy relative to the top and middle portions of the field. Intermittent flooding did not affect the quantity of whole milled rice kernels. Chalk content in rice grown under an intermittent flood production system decreased as distance from the top of the field increased, while chalk content in a continuous flood was similar across the field. Implementing automation with intermittent flood management appears to be an effective strategy to reduce irrigation water applied while having no effect on rice grain yield and quality.
Development of the Agricultural Integrated Management System (AIMS): A web-based decision support tool for Watershed Management in the United States
Year: 2024 Authors: Ozeren Y., Rébillout L., Sahin A., Al-Hamdan M., Bingner R.
The Agricultural Integrated Management System (AIMS) is an automated web-based decision support tool designed to evaluate the impact of agricultural and channel conservation practices on watersheds in the United States. AIMS features a user-friendly Web-GIS interface, enabling interaction with geospatial data, automatic input data preparation, and the visualization and download of watershed simulation results on any internet-enabled device. AIMS utilizes the USDA-ARS Annualized Agricultural Non-Point Source Pollution (AnnAGNPS) model to estimate runoff, sediment yield, and sediment erosion from agricultural areas. Key datasets available include topographic data generated using TOPAGNPS, soil data from NRCS Soil Data Access, and climate data from NLDAS-2 and DAYMET. A Python library was developed to automatically run AnnAGNPS simulations and store the output datasets in the AIMS database. This tool was used to run AnnAGNPS without routing for any watershed in the United States, with manually generated land use and land management data. AnnAGNPS simulations were also performed using AIMS-generated data for selected watersheds in Mississippi to showcase the system's capabilities. This presentation highlights the development, data preparation, and capabilities of the AIMS web interface.
WaterSTAR: A Foundation for Water Budgets and Shaping Water Policy in Alabama
Year: 2024 Authors: Arnold A.
The Geological Survey of Alabama Groundwater Assessment Program (GSA-GAP) has recently implemented the WaterSTAR data management system developed in conjunction with the Ground Water Protection Council (GWPC) and Coordinate Solutions as a foundation for integrated water well data storage, reporting, and retrieval. This digital system allows for simple input of water well construction and water chemistry analytical data. To date, more than 40,000 well records have been entered into the system. Records contain information on well construction, lithology, electric logs, water levels, and other analytical data. These data can be accessed via attribute tables and displayed through a user-friendly GIS interface. Hydrogeologic projects can be easily created in the system allowing for the seamless retrieval of related information in diverse ongoing GSA-GAP studies. Output information can be displayed in ArcGIS map layers for various projects to effectively represent hydrogeologic conditions. WaterSTAR is a vast improvement from earlier relational databases because the system incorporates a spatial display of county map features, the public land survey system, searchable functions, and links to stored data collected by various state agencies. A powerful outcome of using WaterSTAR is the communication of groundwater information to stakeholders and the public in a timely and interactive manner for water management decision-making purposes. This presentation highlights the GSA-GAP approach to managing a statewide groundwater well database with the goal of enabling other public agencies to integrate some of our learned lessons to facilitate their data management objectives.
Denitrification potential in Lower Mississippi River floodplain lakes during summer cutoff
Year: 2024 Authors: Taylor J., Ochs C., Powell J., Shields, Jr. D.
Denitrification in Lower Mississippi River (LMR) backwater areas has the potential to reduce nitrogen (N) loading to the Gulf of Mexico. We hypothesized that NO3-N introduced to backwater lakes through river connection is denitrified during summer disconnection. We tracked summer dissolved NO3-N, NH4-N and N2-N gas in the hypolimnion of a large oxbow lake with seasonal hydrologic connection to the LMR. Denitrification rates were estimated by measuring changes in hypolimnetic excess N2-N concentrations over time. In both years, hypolimnetic N2-N increased while NO3-N declined to almost zero, with estimated denitrification rates accounting for approximately 40% of the observed NO3-N loss. During a mid-summer survey, we also found N2-N was exceeding equilibrium in seven lakes distributed across 560 km of the LMR floodplain, indicating that denitrification is active in backwater areas throughout the lower basin. However, other results highlight the importance of considering alternative N cycling processes through time and space. For example, NH4-N in the hypolimnion increased throughout the summer in both years demonstrating that processes such as dissimilatory nitrate reduction to ammonium (DNRA) and organic matter decomposition need to be considered. While denitrification is an active N removal process in LMR lakes, incorporating more information on other habitat components, such as the epilimnion and flooded forests, and N cycling processes throughout the whole year is needed to better quantify the contribution of denitrification in backwater lakes N removal in the LMR.
Advancing Modeling Tools for [Water and] Soil Conservation Planning
Year: 2024 Authors: Vieira D., Wells R., Bingner R.
The recent rise in resolution, quality, and overall availability of data required for the analysis and modeling of processes related to water and soil conservation issues led to advancements of modeling tools developed and maintained by the USDA-Agricultural Research Service. In the case of soil erosion in agricultural fields, the Revised Universal Soil Loss Equation – Version 2 is routinely employed across the United States for soil conservation analysis and planning. It estimates how agricultural management impacts erosion for a set of parameters describing climate, soil, vegetation, and farming operations. However, as it is routinely employed, it calculates soil loss over a few, user-defined, one-dimensional flow paths that represent areas where sheet-and-rill erosion predominates. With the wide availability of topographic data from LiDAR-based surveys, its erosion calculation methodology was revised, modified and extended so it could also calculate erosion over entire fields, considering the actual field topography and how it governs runoff concentration, soil detachment and sediment transport. Full-field erosion calculations for individual storm events also provide data for the prediction of erosion by concentrated flows that result in the formation of ephemeral gullies and provide a better estimate of the total soil loss to ditches and streams. That was motivation for the development of supporting technologies to deliver the new capabilities to end-users. New methods were developed to automate the creation of a digital description of the landscape that supports the modeling of physical processes at several spatial and temporal scales, and allow the reuse of proven, existing models. Other tools continue to be developed, which include cloud-based technologies that allow easier management of large datasets, many already available from national databases. Also, machine learning approaches are being tested to retrieve, adapt or generate data for natural resources modeling. The effort will result in better tools for soil erosion prediction that respond to stakeholder needs of improved accuracy, ease of use, and availability that respond to changing agricultural practices and climate conditions.
Low NPDES compliance rates in financially challenged municipalities and adopting natural wetlands assimilation as a nature-based solution for increased compliance in Mississippi
Year: 2024 Authors: Ko J., Day J.
Water infrastructures for drinking water supply and wastewater treatment have been deteriorating over the years, mainly due to financial challenges exposed to municipal governments triggered by dwindling their revenues in Mississippi. One consequence is the low NPDES compliance rates among wastewater treatment facilities, which are significantly lower than the national average and even among the Southern states. The City of Jackson could obtain over $600 million from federal governments to improve its water infrastructure after numerous embarrassing reports on disrupted service and sewage treatment in 2021 and 2022. Mound Bayou could upgrade its wastewater facility after obtaining a federal grant in 2023. Now, its updated system complies with the NEDES permit. However, most communities wait for extra funding rather than securing funding by raising property taxes or issuing municipal bonds to follow the MDEQ's guides. The low NPDES compliance rates continue in the State. Adopting natural wetlands assimilation is a nature-based solution that utilizes ecosystem service to provide the tertiary treatment service as a component of the wastewater treatment system. Due to natural energy utilization, not imported electricity, the method increases the NPDES compliance rates with less financial burden. The State of Louisiana has adopted the wetlands method for multiple towns in the State over the years, contributing to increased compliance with less financial burden to local communities. After emphasizing the need for enhancing efforts for water infrastructure in the State, the presentation will also present selected Louisiana cases, including scientific background, typical NPDES permits when the method is applied, field monitoring reports, and the economic benefits of the wetlands assimilation projects in Louisiana as a potential model for Mississippi communities.
Pesticide Runoff from Conventional, Minimum, and No-Tillage Cropping Systems: Meta-Analysis of Published North American Data
Year: 2024 Authors: Fleming D., Krutz J., Spencer D.
Pesticide applications may soon be regulated by laws predicated on the presupposition that reducing tillage and increasing surface residues decreases agrochemical losses in surface runoff. This meta-analysis was conducted to determine whether pesticide transport via surface runoff could be manipulated through tillage. The effects of pesticide and tillage systems on agrochemical runoff were analyzed from experiments conducted in the United States of America and Canada. Transitioning from conventional-tillage to minimum-tillage increased surface residues 5.4-fold while concomitantly reducing runoff, sediment, and pesticide losses 62%, 29%, and 15%, respectively. Conversely, converting from conventional- to no-tillage increased surface residues 16.1-fold, reduced runoff 45%, decreased sediment loss 87%; yet had no effect on pesticide losses because eliminating tillage increased pesticide concentrations in runoff 77%. Consequently, minimizing rather than eliminating tillage may be an effective means to decrease agrochemical losses in surface runoff.
Aquatic Macroinvertebrate Assessment of Restored Wetlands in the Lower Mississippi Alluvial Valley
Year: 2024 Authors: Ousley J., Hall K., Wissmueller A., Entsminger E., Davis J.B.
The U. S. Department of Agriculture Natural Resources Conservation Service implemented the Wetland Reserve Easement (WRE; formerly the Wetland Reserve Program) program in the Mississippi Delta in the early 1990s. The WRE program restores frequently flooded and non-productive agricultural lands to wetlands and bottomland hardwood forests. The WRE program has restored wetland hydrology and reforested over 81,000 ha of marginally productive farmlands in Mississippi as of 2024. After 30 years of program implementation, however, there is limited knowledge of the program's environmental impact. To fill this information void, we assessed ecosystem services of WRE lands in the Mississippi Delta in 2024. One of our primary objectives was to investigate aquatic macroinvertebrate (AMI) communities in WREs, as compared to agricultural drainage ditches and reference wetlands. We sampled 36 sites once in three different aquatic environments, including semi-permanently flooded areas of the WREs (n=27), reference wetlands (n=5), and agricultural cropland drainage ditches (n=4). We collected 63 samples across all sites from March to May 2024. We identified samples to orders and families to estimate indices of abundances, then assigned individual taxa a pollution tolerance value to calculate Hilsenhoff Biotic Indices (HBI) and Shannon-Weiner Diversity Indices. Mean index values for each site category (WRE, reference wetlands, and croplands) were used to explore trends between AMI communities and pollution tolerance of observed taxa. We found that mean specimen tolerance was 6.50 HBI in WRE associated wetlands and less than for both reference wetland (7.07 HBI) and cropland (7.30 HBI) samples, indicating reduced pollution levels in WRE sites. Moreover, the biodiversity index of WRE site samples were nearly identical to reference sites, with means of 1.18 and 1.27, respectively. These preliminary results suggest that WRE implementation positively impacts environmental conditions as indicated by the health and diversity of AMI communities.
WaterAware– A National Weather Service Hydrology Outreach Initiative
Year: 2024 Authors: Roberts A.
Flooding and drought are among the most fatal and costly weather-related phenomena. One of the best ways to diminish damage to life and property due to these disasters is to provide education on flood safety, drought mitigation and the science behind National Weather Service (NWS) river and flood forecasts. WaterAware does just that by serving as the new NWS hydrology outreach program for ages kindergarten through adult. The WaterAware Program supports the development of a more informed water-ready public with outreach materials spanning the range of water resource issues, from flash flooding, flood inundation, and dam breaks to drought, snowmelt runoff, and water supply. WaterAware consists of a group of on-demand NWS hydrology outreach representatives who can be scheduled for in-person or virtual educational sessions. We offer a variety of high-quality hydrology and water resources materials and activities tailored to each age group all the way up to adults! The WaterAware program is much more than just educational outreach sessions. We have additional subgroups that handle topics such as recruiting new NWS hydrologists, minority outreach and recruitment, and graphic design of our materials. The presentation covers the NWS river forecasting process, the many levels of support the NWS river forecast centers (RFCs) provide, as well as introducing the various types of hydrologists within the NWS. We'll also touch on some of Waterware's free programs and outreach materials.
Mississippi Water Stewards: Building capacity for watershed protection through community-based monitoring
Year: 2024 Authors: Baker B., Hill M., Dominguez M., Ruiz-Cordova S.
Community-based water monitoring (CBWM) programs provide mechanisms for education and water quality data collection. In this study, we adapted a long-standing CBWM program from Alabama, the Alabama Water Watch program, to Mississippi as the Mississippi Water Stewards program. Program goals included advancing the protection of Mississippi's freshwater resources and the Gulf of Mexico. We adapted the program by revising traditional educational mediums to hybrid training courses, digital data tracking and digital evaluation. Program implementation was conducted in three pilot watersheds from 2021-2024. We adapted and developed educational content to include the principles and practices of bacteriological, chemistry, and biomonitoring water testing. Training was delivered through online modules and nine in-person workshops, training 151 individuals. We assessed the efficacy of initial program implementation to determine changes in knowledge, skills, and behavior. Results indicate that 94% of survey respondents (n=66) strongly agreed that their knowledge was increased, 96% strongly agreed that they learned a new skill, and 99% reported an intention to start monitoring or engage in stewardship activities. However, of the those who intended to adopt a new behavior, 33% of participants engaged in active monitoring. Citizens who completed all training and engaged in active monitoring collected 1039 water monitoring observations. Our data indicate that while frequency and longevity of data collection varied, those who monitored with a local community-based group were more likely to engage in active monitoring. This study offers data-driven strategies to expand and adopt CBWM into Extension curricula that builds knowledge, skills, and capacity for community water resource conservation.
Phase A Bridge Recommendations Over Town Creek and Town Creek Tributary in Baldwyn, Mississippi
Year: 2024 Authors: Spain C., Hendon D.
MDOT commissioned an extensive hydraulic analysis to assess water surface elevations (WSEs) and velocities near the SR 145 crossings. The study considered various bridge lengths and sizes, accounting for channel degradation and instabilities at the project site. Leveraging the SRH-2D model software, we evaluated multiple scenarios to mitigate erosion and scour risks. Our final recommendations prioritize safety and resilience, addressing the underlying instabilities that could impact the proposed bridges. The presentation will discuss the analysis procedure in detail and explain why the recommendations were chosen.
Exploring hydroecological impact on bald cypress if water is impounded in oxbows for managed aquifer recharge
Year: 2024 Authors: Gleason J., Williams D., Granger J., Davidson G.
Paired tree cores were collected from 40 bald cypress (Taxodium distichum) in an oxbow wetland in the Delta of Mississippi, to compare with a high-resolution 22-yr record of lake levels (measured every 6 hrs), precipitation records and tree age (mostly 100 to 250 yrs). Trees were distributed across an elevation gradient, with the roots of lower-elevation trees sitting in saturated sediments year-round and the roots of higher-elevation trees seasonally aerated to varying degrees. Analysis of tree trees was complicated by an unexpected number of double rings in the same year, a phenomenon typically associated with intermittent dry and wet conditions in the same growing season. For baldcypress trees taken at a higher elevation, an average diameter growth per year of 4.52 mm per year, and for baldcypress tree at a lower elevation, an average diameter growth of 4.89 mm per year. During years with below average water depths, trees at a higher elevation were shown to have an average diameter growth per year of 4.31 mm. The trees at lower elevations were found to have an average diameter growth per year of 4.46 mm. Preliminary findings appear to indicate a water-depth/growth response in the lower elevation trees. Additional tree cores and analyses will be presented at the meeting.