High-water marks provide valuable data for understanding recent and historical flood events. The proper collection and recording of high-water mark data from perishable and preserved evidence informs flood assessments, research, and water resource management. Given the high cost of flooding in developed areas, experienced hydrographers, using the best available techniques, can contribute high-quality data toward efforts such as public education of flood risk, flood inundation mapping, flood frequency computations, indirect streamflow measurement, and hazard assessments.This manual presents guidance for skilled high-water mark identification, including marks left behind in natural and man-made environments by tranquil and rapid flowing water. This manual also presents pitfalls and challenges associated with various types of flood evidence that help hydrographers identify the best high-water marks and assess the uncertainty associated with a given mark. Proficient high-water mark data collection contributes to better understanding of the flooding process and reduces risk through greater ability to estimate flood probability.The U.S. Geological Survey, operating the Nation’s premier water data collection network, encourages readers of this manual to familiarize themselves with the art and science of high-water mark collection. The U.S. Geological survey maintains a national database at http://water.usgs.gov/floods/FEV/ that includes high-water mark information for many flood events, and local U.S. Geological Survey Water Science Centers can provide information to interested readers about participation in data collection and flood documentation efforts as volunteers or observers.

Stream gauge height (water level) is a significant indicator for forecasting future floods. Flooding occurs when the water level exceeds the flood stage. Predicting imminent floods can save lives, protect infrastructure, and improve road traffic management and transportation. Deep neural networks have been increasingly used in this domain due to their predictive capabilities in capturing complex features and interdependencies. This study employs four distinct models—Multi-Layer Perceptron (MLP), Long Short-Term Memory (LSTM), transformer, and LSTNet—with MLP serving as the baseline model to forecast water levels. The models are trained using data from 20 distinct river gages across the state of Missouri to ensure consistent performance. Random search optimization is employed for hyperparameter tuning. The prediction intervals are set at 4, 6, 8, and 10 (each interval equivalent to 30 min) to ensure that performance results are robust and not due to random weight initialization or suboptimal hyperparameters and are consistent throughout different prediction intervals. The findings of this study indicate that the LSTNet model leads to a better performance than the other models, with a median RMSE of 0.00724, 0.00959, 0.01204, and 0.01230 for the 4, 6, 8, and 10 intervals, respectively. As climate change leads to localized storms driven by atmospheric shifts, water level fluctuations are becoming increasingly extreme, further exacerbating data drift in real-world datasets. The LSTNet model demonstrates superior performance in terms of RMSE, MAE, and the correlation coefficient across all prediction intervals when forecasting water levels under data drift conditions.

This research examined the influences of outflow characteristics affecting riverbank stability. The 130-km stretch of the Lower Osage River downstream from Bagnell Dam (Missouri, USA) provided an excellent case study for this purpose. The integrated BSTEM model with the HEC-RAS model was accurately calibrated and validated with data from the US Geological Survey. Then, the outflow characteristics (peak flow duration, flow drawdown rate, and low flow duration) were investigated individually. The results of this study showed that: (1) riverbank stability is little affected by the duration time of the peak flow, especially on the reaches far from the dam; (2) sudden flow drawdown significantly reduces riverbank stability; however, the impact of the drawdown rate decreases with distance from the dam; and (3) the duration of the low flow after peak flow influences the riverbank stability value proportional to the distance from the dam. The time of low flow before failure increases as the distance from the dam increases.