Journal of Hydrology X最新文献

This paper presents a methodology for the creation of dynamic reservoir rule curves on the basis of the results of implicit stochastic optimization coupled with optimized demand hedging embedded as constraints to optimization. The novelty of the method is a dynamic rule curve that always starts from the current storage level and projects a range of anticipated target levels in the immediate future based on the statistical analyses of the results of implicit stochastic optimization. The method is particularly useful in dry years when storage is not completely filled at the end of wet seasons. Such situations cannot be addressed with standard traditional rule curves, thus causing reservoir operators to base their decisions on mere judgment. The proposed method can be helpful in such situations. The method has been demonstrated on the Tawa reservoir in the Narmada River Basin in India.

Tile drains constitute a shortcut from agricultural fields to surface water systems, significantly altering the transport pathways and fate of nitrate during transport. A correct representation of tile drainage flow is thus crucial for estimating nitrate load at the catchment scale and to identify optimal locations for N-mitigation measures. Drainage is a local process, controlled by local properties and drain configurations, which are rarely known for individual fields, making drainage flow and transport a challenging task in catchment scale models. This study tests the potential for improving drainage flow dynamics at catchment scale, by utilising local drainage flow measurements in a spatial calibration scheme. A distributed hydrological model, MIKE SHE, for the agricultural-dominated Norsminde catchment (145 km²) in Denmark, was calibrated using spatially distributed surrogate parameters (pilot points) to represent heterogeneity in the soil (top 3 m) and the deeper geology below 3 m. The model was calibrated using hydraulic heads, stream discharge, and measured drainage flow from eight drain catchments. Drain measurements were very important in guiding the calibration of top 3 m and subsurface pilot points located in the drainage fields, showing that drain flow hold information on both local (shallow) and regional (deeper) flow patterns. Contrarily, pilot points located outside the drained fields were mainly sensitive to the hydraulic head measurements and the summer water balance of the stream discharge on a catchment scale. Consequently, incorporation of the drain data improved local performance, but did not improve the parameterization and drain description of the entire catchment. Exploitation of the drain flow information is thus difficult beyond the drain catchments, and other approaches are needed to extrapolate and exploit the local data.

Agriculture in regions with limited water availability is possible because of irrigation. Irrigated croplands are expanding, and irrigation water demand is increasing. Nevertheless, there is a limited understanding of how much water is consumed for irrigation and how effective irrigation increases crop productivity in various climates. In this study, we aim to understand how irrigation water affects crop productivity in different climates. To achieve this goal, we developed a simple approach to quantify irrigation quantities from SMAP satellite soil moisture observations based on a zero-dimensional bucket-type hydrology model. The central assumption is that irrigation quantities can be estimated from the gap between the modeled and observed soil moisture by iteratively providing irrigation as a model input until the soil moisture simulations agree well with the observations. We then used the estimated amount of irrigation to simulate water, energy, and carbon fluxes at two agricultural sites on the west coast of the US: one that was water-limited (Central Valley, CA) and one that was energy-limited (Eugene, OR). An agroecosystem model, AgroIBIS-VSF, was used to conduct simulations. To verify our simulations, we used data from two AmeriFlux Eddy covariance towers at each site. We found that incorporating estimated irrigation amounts into our simulations improved the accuracy of energy balance components and soil moisture predictions, reducing the root-mean-square error of soil moisture predictions by up to 22%. We also discovered that the irrigation value, in terms of increased productivity of actual irrigation water used, is more than five times more valuable at the energy-limited site than at the water-limited site. Soil hydraulic properties have a strong influence on irrigation water valuation. Our study highlights the potential of satellite soil moisture observations to improve our understanding of water productivity in different climates. By better understanding the efficiency of resources used for crop production, we can ensure the sustainability and resilience of agricultural systems, leading to better management practices.

The increase in extreme rainfall events due to climate change, combined with urbanisation, leads to increased risks to urban infrastructure and human life. Physically based urban flood models capable of producing water depth maps with sufficient spatial and temporal resolution are generally too slow for decision makers to react in time during an extreme event. We present a surrogate model with high temporal and spatial resolution for real-time prediction of water levels during a pluvial urban flood. We used machine learning techniques to achieve short computation times. The recursive approach used in this work combines convolutional and fully coupled multilayer architectures. The database for the machine learning was pre-simulated results from a physically based urban flood model. The forcing input of the prediction is precipitation and the output is water level maps with a temporal resolution of 5 min and a spatial resolution of 6 x 6 meters. The prediction performance can be considered promising for testing the model in real operational applications.

Groundwater, the second largest stock of freshwater on the planet, is an important water source used for municipal water supply, irrigation, or industrial needs. For instance, California’s arid Central Valley relies on groundwater resources to produce a quarter of the United States’ food demand as farmers rely on this precious resource when surface water is scarce. Despite its importance, the nexus between groundwater dynamics and climate drivers remains difficult to quantify, model, and predict because of the lack of a comprehensive observation network. In this study, machine learning techniques were used to predict groundwater levels with a 3-month forecasting horizon for the Sacramento River Basin. For this, publicly available meteorological and hydrological datasets and in-situ well-level measurements were used. Time series, ensemble-based, and deep-learning models including transformers were all tested, with an ensemble-based, XGBoost model, producing the best mean standard deviation percent error (MSPE) of 32.23% and a root mean squared error (RMSE) of 1.05 m (m) when using a 3- month forecasting horizon and when tested using a monthly rolling window over the years 2017–2020. The model proved to be better at predicting into wet months than the dry summer months and was found to be better at extracting seasonality than explaining well-level residuals, with well-specific features, as opposed to exogenous meteorological features specific to the hydrological unit of the well, ranking as the most important features to the model. Though other forecasting horizons were tested, a 3-month look-ahead window resulted in the best balance of precision and accuracy, where smaller forecasting horizons resulted in smaller RMSE but larger MSPE scores and vice-versa for larger forecasting horizons.

Headwater streams (HWS) are ecologically important components of montane ecosystems. However, they are difficult to map and may not be accurately represented in existing spatial datasets. We used topographically resolved climatic water balance data and satellite indices retrieved from Google Earth Engine to model the occurrence (presence or absence) of HWS across Northwest Montana. A multi-scale feature selection (MSFS) procedure and boosted regression tree models/machine learning algorithms were used to identify variables associated with HWS occurrence. In final model evaluation, models that included climatic water balance deficit were more accurate (83.5% ranging from 82.9% to 83.7%) than using only terrain indices (81.1% ranging from 80.7% to 81.4%) and improved upon estimates of stream extent represented by the National Hydrography Dataset Plus High Resolution (NHDPlus HR) (82.7% ranging from 82.5% to 83.1%). Including topoclimate captured the varying effect of upslope accumulated area across a strong moisture gradient. Multi-scale cross-validation, coupled with a MSFS algorithm allowed us to find a parsimonious model that was not immediately evident using standard cross-validation procedures. More accurate spatial model predictions of HWS have potential for immediate application in land and water resource management, where significant field time can be spent identifying potential stream impacts prior to contracting and planning.

Peridotite aquifers are ubiquitous on Earth, but most are in the deep-sea, and thus difficult to access. Ophiolites provide a unique opportunity to study peridotite aquifers, and the Oman Drilling Project established a Multi-Borehole Observatory in a peridotite terrain of the Samail ophiolite. We use the water level response of two 400-m deep boreholes (BA1B, BA1D) to solid Earth, ocean, and atmospheric tides to investigate the hydromechanical structure of the aquifer. The two boreholes are offset by ∼ 100 m but exhibit markedly different tidal responses, indicating a high degree of short-length-scale heterogeneity. Hole BA1B does not respond to tidal strain or barometric loading, consistent with the behavior of an unconfined aquifer. Hole BA1D responds to both tidal strain and barometric loading, indicating some degree of confinement. The response to applied strain, which includes a non-negligible ocean tidal loading component, is consistent with a partially confined, low conductivity aquifer. The response to barometric loading appears to be affected by the complex hydrological structure of the surficial zone and we were not able to fit the observations to within error. Aquifer conductivity estimates for Hole BA1D based on the response to tidal strain are within a factor of ∼ 3 of pumping test estimates.

In-stream habitat models at the meso-scale are increasingly used to quantify the effects of hydro-morphological pressures in rivers. The spatial distributions of water depth and velocity represent key attributes of physical habitat. Choosing between field surveys, hydraulic modeling or their integration is made depending on available tools, technical skills, budget and time. However, the sensitivity to such choices of estimated habitat conditions suitable for biological organisms, such as fish, is poorly known.

In this study, three commonly used approaches in hydraulic-habitat modeling were compared and tested on a mountain stream, the Mareta River (NE Italy). Two approaches were based on 2D hydraulic modeling, calculated on computational meshes with varying resolution and quality: (1) high-resolution meshes derived from topographical data obtained from Airborne Bathymetric LiDAR; (2) a mesh extrapolated from topographical cross-sectional profiles. The third approach (3) was based on in-stream surveys. From these, suitable channel-area for two fish species, the marble trout (juvenile and adult), and the European bullhead (adult), were estimated.

Results showed that decreasing mesh resolution and quality affects the simulated water depth and velocity distributions, both in terms of their average and their standard deviation. The largest differences were found for the in-stream survey-based results. Morphologically complex unit types, such as steps, rapids and pools were more sensitive than simpler mesohabitats, such as glides and riffles. The most sensitive hydro-morphological unit types to the chosen approach were backwaters, glides being the least sensitive, also in terms of their suitability as mesohabitats. Despite that, a key finding is that errors are minimized when deriving habitat - streamflow rating curves at the reach scale, for which all approaches were largely able to reproduce the main characteristics of the curve, i.e. maxima, minima and inflection points.