Infrastructure is a significant factor in economic growth for systems of government. In order to increase economic productivity, maintaining infrastructure quality is essential. One of the elements of infrastructure is roads. Roads are means which help local and national economies be more productive. Furthermore, road damage such as potholes, debris, or cracks is the cause of many on-road accidents that have cost the lives of many drivers. In this paper, we propose a system that uses Convolutional Neural Networks to detect road degradations without data pre-processing. We utilize the state-of-the-art object detection algorithm, YOLO detector for the system. First, we developed a basic system working on data collecting, pre-processing, and classification. Secondly, we improved the classification performance achieving 97.98% in the overall model testing, and then we utilized pixel-level classification and detection with a method called semantic segmentation. We were able to achieve decent results using this method to detect and classify four different classes (Manhole, Pothole, Blurred Crosswalk, Blurred Street Line). We trained a segmentation model that recognizes the four classes mentioned above and achieved great results with this model allowing the machine to effectively and correctly identify and classify our four classes in an image. Although we obtained excellent accuracy from the detectors, these do not perform particularly well on embedded systems due to their network size. Therefore, we opted for a smaller, less accurate detector that will run in real time on a cheap embedded system, like the Google Coral Dev Board, without needing a powerful and expensive GPU.

During every Atlantic hurricane season, storms represent a constant risk to Texan coastal communities and other communities along the Atlantic coast of the United States. A storm surge refers to the abnormal rise of sea water level due to hurricanes and storms; traditionally, hurricane storm surge predictions are generated using complex numerical models that require high amounts of computing power to be run, which grow proportionally with the extent of the area covered by the model. In this work, a machine-learning-based storm surge forecasting model for the Lower Laguna Madre is implemented. The model considers gridded forecasted weather data on winds and atmospheric pressure over the Gulf of Mexico, as well as previous sea levels obtained from a Laguna Madre ocean circulation numerical model. Using architectures such as Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) combined, the resulting model is capable of identifying upcoming hurricanes and predicting storm surges, as well as normal conditions in several locations along the Lower Laguna Madre. Overall, the model is able to predict storm surge peaks with an average difference of 0.04 m when compared with a numerical model and an average RMSE of 0.08 for normal conditions and 0.09 for storm surge conditions.

There is increasing interest in using pyrogenic carbon as an adsorbent for aqueous contaminants in stormwater. The objective of this study was to investigate pyrogenic carbon materials as an amendment to geomedia to reduce nitrate leaching. Batch adsorption and column experiments were conducted to evaluate the performance of a commercial activated carbon and two biochars incorporated (5% by weight) into sand and pumice columns. The batch adsorption with 50 mg L−1 of nitrate solution showed that only activated carbon resulted in a substantial adsorption for nitrate up to 41%. Tested biochars were not effective in removing aqueous nitrate and even released nitrate (<1%) with 1 h reaction time. Column experiment with a pulse input of nitrate solution (50 mg L−1) confirmed that the sand or pumice columns amended with biochars were not as effective as those amended with activated carbon for reducing nitrate leaching. Our results suggested that net negatively charged surfaces of biochar may inhibit nitrate anion adsorption while activated carbon has reactive sites containing acidic functional groups to improve nitrate retention. There was no difference between sand and pumice for nitrate retention in any of the carbon amendments. Additional surface activation process during biochar production may be needed to improve adsorptive capacity of biochar for aqueous nitrate removal.

The intensification of climatic changes, mainly natural geophysical hazards like hurricanes, are of great interest to the South Texas region. Scientists and engineers must protect essential resources from coastal threats, such as storm surge. This study presents the development process and improvements of a hydrodynamic finite element model that covers the South Texas coast, specifically the Lower Laguna Madre, for the aid of local emergency management teams. Four historical tropical cyclone landfalls are evaluated and used as a means of verification of the hydrodynamic model simulation results. The parameters used to improve the accuracy of the model are the tidal harmonic constituents and the surface roughness coefficient, or manning's n value. A total of four different scenarios that use a variety of tidal constituent combinations and nodal attribute files were developed to identify the best case. Statistical evaluation, such as regression analysis, normalized root mean square error, and scatter index, was used to determine the significance of each hydrodynamic computational storm surge result with observed historical water surface elevations. In an effort to improving all models locally, using seven tidal constituents combinations along with a surface roughness nodal attribute grid that assigns values with respect to bathymetric data improves the accuracy of the storm surge model and should, therefore, be implemented for future hydrodynamic studies in the South Texas region.

Sift Desk Research Page is fully open access journals.

Small reservoirs in semi-arid regions experience substantial evaporative losses but are rarely monitored at daily scales. A multi-reservoir machine learning (ML) framework was developed to estimate daily open-water evaporation. Empirical models (Penman, Penman-Monteith, Priestley-Taylor, Bowen Ratio Energy Budget) andabenchmark combination method (Daily Lake Evaporation Model-DLEM) were compared against ML models. Predictors combined gridded meteorology (gridMET) with reservoir attributes (surface area, average depth, maximum depth, and fetch). ML models (Random Forest-RF, Decision Tree-DT, K-Nearest Neighbor-KNN, and Support Vector Regression-SVR) were trained on four reservoirs using data from 2018 to 2025. Results from ML models were further validated using both DLEM and TexasETNet observations at Delta Lake. On multi-reservoir tests, RF and SVR perform best (R² = 0.67; RMSE ~ 1.5 mm d^− 1; NSE = 0.67). RF aligns most closely with DLEM (R2 = 0.78; RMSE = 1.22 mm d^− 1), whereas SVR aligns better with TexasETNet (R² = 0.33) and shows lowest bias among ML models. Mean annual evaporation from ML and DLEM is 1620–1698 mm yr⁻¹, which exceeds predictions from TexasETNet (1452 mm yr⁻¹) and exhibits narrower interannual variability (± 300–350 mm yr ¹ vs. ±700 mm yr⁻¹). Feature importance and SHAP analyses identified shortwave radiation, temperature (Tmax/Tmin), and ET₀ as dominant drivers. The proposed transferable ML framework delivers accurate daily estimates where observations are sparse, while dual validation clarifies differences between process-based and station datasets, supporting reservoir operations and evaporation mitigation planning in under monitored, semi-arid regions.