Many coastal cities are facing frequent flooding from storm events that are made worse by sea level rise and climate change. The groundwater table level in these low relief coastal cities is an important, but often overlooked, factor in the recurrent flooding these locations face. Infiltration of stormwater and water intrusion due to tidal forcing can cause already shallow groundwater tables to quickly rise toward the land surface. This decreases available storage which increases runoff, stormwater system loads, and flooding. Groundwater table forecasts, which could help inform the modeling and management of coastal flooding, are generally unavailable. This study explores two machine learning models, Long Short-term Memory (LSTM) networks and Recurrent Neural Networks (RNN), to model and forecast groundwater table response to storm events in the flood prone coastal city of Norfolk, Virginia. To determine the effect of training data type on model accuracy, two types of datasets (i) the continuous time series and (ii) a dataset of only storm events, created from observed groundwater table, rainfall, and sea level data from 2010–2018 are used to train and test the models. Additionally, a real-time groundwater table forecasting scenario was carried out to compare the models’ abilities to predict groundwater table levels given forecast rainfall and sea level as input data. When modeling the groundwater table with observed data, LSTM networks were found to have more predictive skill than RNNs (root mean squared error (RMSE) of 0.09 m versus 0.14 m, respectively). The real-time forecast scenario showed that models trained only on storm event data outperformed models trained on the continuous time series data (RMSE of 0.07 m versus 0.66 m, respectively) and that LSTM outperformed RNN models. Because models trained with the continuous time series data had much higher RMSE values, they were not suitable for predicting the groundwater table in the real-time scenario when using forecast input data. These results demonstrate the first use of LSTM networks to create hourly forecasts of groundwater table in a coastal city and show they are well suited for creating operational forecasts in real-time. As groundwater table levels increase due to sea level rise, forecasts of groundwater table will become an increasingly valuable part of coastal flood modeling and management.
more »
« less
Long Short-Term Memory Based Subsurface Drainage Control for Rainfall-Induced Landslide Prevention
Subsurface drainage has been widely accepted to mitigate the hazard of landslides in areas prone to flooding. Specifically, the use of drainage wells with pumping systems has been recognized as an effective short-term solution to lower the groundwater table. However, this method has not been well considered for long-term purposes due to potentially high labor costs. This study aims to investigate the idea of an autonomous pumping system for subsurface drainage by leveraging conventional geotechnical engineering solutions and a deep learning technique—Long-Short Term Memory (LSTM)—to establish a geotechnical cyber-physical system for rainfall-induced landslide prevention. For this purpose, a typical soil slope equipped with three pumps was considered in a computer simulation. Forty-eight cases of rainfall events with a wide range of varieties in duration, total rainfall depths, and different rainfall patterns were generated. For each rainfall event, transient seepage analysis was performed using newly proposed Python code to obtain the corresponding pump’s flow rate data. A policy of water pumping for maintaining groundwater at a desired level was assigned to the pumps to generate the data. The LSTM takes rainfall event data as the input and predicts the required pump’s flow rate. The results from the trained model were validated using evaluation metrics of root mean square error (RMSE), mean absolute error (MAE), and R2. The R2-scores of 0.958, 0.962, and 0.954 for the predicted flow rates of the three pumps exhibited high accuracy of the predictions using the trained LSTM model. This study is intended to make a pioneering step toward reaching an autonomous pumping system and lowering the operational costs in controlling geosystems.
more »
« less
- Award ID(s):
- 1742656
- NSF-PAR ID:
- 10321068
- Date Published:
- Journal Name:
- Geosciences
- Volume:
- 12
- Issue:
- 2
- ISSN:
- 2076-3263
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
more » « less
Abstract Threshold changes in rainfall‐runoff generation commonly represent shifts in runoff mechanisms and hydrologic connectivity controlling water and solute transport and transformation. In watersheds with limited human influence, threshold runoff responses reflect interaction between precipitation event and antecedent soil moisture. Similar analyses are lacking in intensively managed landscapes where installation of subsurface drainage tiles has altered connectivity between the land surface, groundwater, and streams, and where application of fertilizer has created significant stores of subsurface nitrogen. In this study, we identify threshold patterns of tile‐runoff generation for a drained agricultural field in Illinois and evaluate how antecedent conditions—including shallow soil moisture, groundwater table depth, and the presence or absence of crops—control tile response. We relate tile‐runoff thresholds to patterns of event nitrate load observed across multiple storm events and evaluate how antecedent conditions control within‐event nitrate concentration‐discharge relationships. Our results demonstrate that an event tile‐runoff threshold emerges relative to the sum of gross precipitation and indices of antecedent shallow soil moisture and antecedent below‐tile groundwater moisture deficit, indicating that both shallow soil and below‐tile storages must be filled to generate significant runoff. In turn, event nitrate load shows a linear dependence on runoff for most time periods, suggesting that subsurface nitrate export and storage can be estimated using runoff threshold relationships and long‐term average nitrate concentrations. Finally, within‐event nitrate concentration‐discharge relationships are controlled by event size and the antecedent tile flow state because these factors dictate the sequence of flow path activation and tile connectivity over a storm event.
-
Valveless pumping based on Liebau mechanism entails asymmetrical positioning of the compression site relative to the attachment sites of the pump’s elastic segment to the rest of the circuit. Liebau pumping is believed to play a key role during heart development and be involved in several other physiological processes. Until now studies of Liebau pump have been limited to numerical analyses, in silico modeling, experiments using non-biological elements, and a few indirect in vivo measurements. This review aims to stimulate experimental efforts to build Liebau pumps using biologically compatible materials in order to encourage further exploration of the fundamental mechanisms behind valveless pumping and its role in organ physiology. The covered topics include the biological occurrence of Liebau pumps, the main differences between them and the peristaltic flow, and the potential uses and body sites that can benefit from implantable valveless pumps based on Liebau principle. We then provide an overview of currently available tools to build such pumps and touch upon limitations imposed by the use of biological components. We also talk about the many variables that can impact Liebau pump performance, including the concept of resonant frequencies, the shape of the flowrate-frequency relationship, the flow velocity profiles, and the Womersley numbers. Lastly, the choices of materials to build valveless impedance pumps and possible modifications to increase their flow output are briefly discussed.more » « less
-
more » « less
Abstract Many irrigated agricultural areas seek to prolong the lifetime of their groundwater resources by reducing pumping. However, it is unclear how lagged responses, such as reduced groundwater recharge caused by more efficient irrigation, may impact the long‐term effectiveness of conservation initiatives. Here, we use a variably saturated, simplified surrogate groundwater model to: (a) analyze aquifer responses to pumping reductions, (b) quantify time lags between reductions and groundwater level responses, and (c) identify the physical controls on lagged responses. We explore a range of plausible model parameters for an area of the High Plains aquifer (USA) where stakeholder‐driven conservation has slowed groundwater depletion. We identify two types of lagged responses that reduce the long‐term effectiveness of groundwater conservation, recharge‐dominated and lateral‐flow‐dominated, with vertical hydraulic conductivity (
K Z) the major controlling variable. When highK Zallows percolation to reach the aquifer, more efficient irrigation reduces groundwater recharge. By contrast, when lowK Zimpedes vertical flow, short term changes in recharge are negligible, but pumping reductions alter the lateral flow between the groundwater conservation area and the surrounding regions (lateral‐flow‐dominated response). For the modeled area, we found that a pumping reduction of 30% resulted in median usable lifetime extensions of 20 or 25 years, depending on the dominant lagged response mechanism (recharge‐ vs. lateral‐flow‐dominated). These estimates are far shorter than estimates that do not account for lagged responses. Results indicate that conservation‐based pumping reductions can extend aquifer lifetimes, but lagged responses can create a sizable difference between the initially perceived and actual long‐term effectiveness. -
Core Ideas A subsurface drainage‐fed bioreactor was retrofitted with a supplemental surface water pumping system. Design criteria of the pumping system are presented along with challenges and future recommendations. Pumped bioreactor systems show promise for the treatment of alternative nitrate‐laden sources of water. Pumped bioreactors have the potential to remove nitrate beyond the typical drainage season.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.3390/geosciences12020064
