Search for: All records

Abstract Levees play a critical role in safeguarding communities and assets from flooding, acting as essential defenses against the devastating impacts of inundation. Yet, earthen levees are prone to breaches, especially in the face of swift floodwaters. Distributed low-cost sensor networks offer the potential to generate spatial maps illustrating soil moisture levels. Long-term monitoring of these spatial maps could identify vulnerable zones in the levee while providing an understanding of how climate change affects levee stability. This study presents an investigation into spatial monitoring of soil saturation in levees using a wireless network of UAV-deployable sensing spike packages. The goal of this paper is to demonstrate the use of these sensors for assessing soil conductivity levels in sand-filled embankments. The obtained soil conductivity levels are crucial for determining soil saturation. The developed sensing spikes consist of a spike that penetrates the ground and measures conductivity between two electrically conductive contacts. The sensing spike consists of microprocessors for edge computing, and wireless data communication systems that report data to a way station in real-time. To validate the efficacy of the developed sensors, a flume test is developed as a replica of a levee and monitored under controlled water flow conditions. The analysis of data at different times revealed the progression of moisture throughout the earthen embankment. Initially, the soil is almost dry. As the controlled water flow proceeds, the soil becomes partially saturated, with the final stage showing a dominant presence of saturated soil. The collected data sampled at the measurement points is expanded to a continuous moisture profile using kriging. Gaussian kriging, also known as ordinary kriging, is one of the commonly used variants of the kriging method. In Gaussian kriging, the estimation of values at unsampled locations is based on a linear combination of nearby data points, with weights determined by their spatial relationships. The Gaussian assumption implies that the errors in the estimation process follow a normal distribution. The extended knowledge about saturation levels obtained through kriging can lead to insights for predicting vulnerable areas and preempting potential failures. Overall, this study paves the way for further development of a wireless network of sensing spike packages as a UAV-deployable system for levee health assessment and improved infrastructure management. 

Heavy rains and tropical storms often result in floods, which are expected to increase in frequency and intensity. Flood prediction models and inundation mapping tools provide decision-makers and emergency responders with crucial information to better prepare for these events. However, the performance of models relies on the accuracy and timeliness of data received from in situ gaging stations and remote sensing; each of these data sources has its limitations, especially when it comes to real-time monitoring of floods. This study presents a vision-based framework for measuring water levels and detecting floods using computer vision and deep learning (DL) techniques. The DL models use time-lapse images captured by surveillance cameras during storm events for the semantic segmentation of water extent in images. Three different DL-based approaches, namely PSPNet, TransUNet, and SegFormer, were applied and evaluated for semantic segmentation. The predicted masks are transformed into water level values by intersecting the extracted water edges, with the 2D representation of a point cloud generated by an Apple iPhone 13 Pro lidar sensor. The estimated water levels were compared to reference data collected by an ultrasonic sensor. The results showed that SegFormer outperformed other DL-based approaches by achieving 99.55 % and 99.81 % for intersection over union (IoU) and accuracy, respectively. Moreover, the highest correlations between reference data and the vision-based approach reached above 0.98 for both the coefficient of determination (R2) and Nash–Sutcliffe efficiency. This study demonstrates the potential of using surveillance cameras and artificial intelligence for hydrologic monitoring and their integration with existing surveillance infrastructure. 

Abstract Levees are built to safeguard human lives, essential infrastructure, and farmland. However, failure of levees can have catastrophic impacts due to a fast rate of inundation in areas protected by levees. Earthen levees are prone to failure due to excessive moisture content that reduces the shear strength of the soil. The use of levee monitoring systems has demonstrated the ability to reduce the likelihood of failure by creating maps that depict the saturation levels of the surface of the levee, both in terms of space and time. By utilizing extensive sensor networks to continuously monitor these geo-infrastructure systems, the structural deterioration attributed to changing climate can be studied. Measuring environmental parameters surrounding such structures provides insight into the potential stressors that cause structural failure. Steps can then be taken to mitigate those effects on the levees and maintain structural integrity. However, the massive scale of levees makes it difficult to monitor with conventional wired sensors. This paper presents a preliminary investigation into the development and validation of UAV-deployable smart sensing spikes for soil conductivity levels in levees, which is a measurement modality for determining soil saturation levels. For this work, Gaussian process regression (also known as kriging) is used to model the soil saturation levels between sensing spikes obtaining a continuous moisture map of the levees. The expanded data is then categorized using a clustering-based machine learning approach with conductivity data from sensing spikes as model inputs. The machine learning model output is sorted into three categories: dry, partially saturated, and saturated soil. The findings of a laboratory study are presented, and the implications of the raw and expanded data are discussed. This work will aid in predicting potential levee failure risks and maintenance requirements based on the analysis of the soil conditions using a network of smart sensing spikes.