Hydrological Processes最新文献

Although being a viable substrate to reduce soil erosion, the effects of biochar from solid waste of olive oil mills (OMSW biochar) on soil detachment capacity in rills (D_c) has never been explored. Furthermore, no equations have been proposed to predict important parameters (soil erodibility factor, K_r, shear stress, τ, and its critical value, τ_c, for rills) of this hydrological process in treated soils. This study was conducted in agro-forest sites of Northern Iran to evaluate D_c and key properties of soils–organic carbon (OC), aggregate stability (MWD), bulk density (BD) and carbon exchange capacity (CEC)—treated with OMSW biochar in comparison to untreated sites through flume experiments. Moreover, regression models were developed to predict D_c, K_r and τ_c for both treated and untreated soils. Compared with the untreated soil, the application of this biochar noticeably increased OC (+85%), MWD (+51%) and CEC (+101%), and reduced BD (−11%) and D_c (−31%). The correlation analysis revealed significant but not high associations between physical properties on one the hand, and soil detachment capacity in rills, on the other hand. Overall, the soil treatment with OMSW biochar impacted agro-forest soils to a severe extent. Treated soils were discriminated from untreated sites into two distinct groups by the principal component analysis and agglomerative hierarchical cluster analysis. The linear equations interpolating D_c and τ estimated K_r, and τ_c with accuracy in treated and untreated soils (r² > 0.74, p < 0.05). The best prediction capacity of D_c was given by power equations applied to the stream power (r² > 0.78, p < 0.05). The multiple regression equation developed to estimate D_c from the water flow rate and soil slope was also very accurate (r² > 0.95 and NSE, coefficient of Nash and Sutcliffe, > 0.89). These results help land managers and hydrologists to control and predict rill detachment in long and steep hillslopes against the risk of soil erosion.

Coastal groundwater is a vital resource for coastal communities around the globe, and submarine groundwater discharge (SGD) delivers nutrients to coastal marine ecosystems. Climatic changes and anthropogenic actions alter coastal hydrology, causing seawater intrusion (SWI) globally. However, the selection of SWI and SGD study sites may be highly biased, limiting our process knowledge. Here, we analyse hydroenvironmental characteristics of coastal basins studied in 1298 publications on SGD and SWI to understand these potential biases. We find that studies are biased towards basins with gross domestic product per capita below (SWI) and above (SGD) the median of all global coastal basins. Urban coastal basins are strongly overrepresented compared to rural coastal basins, limiting our progress in understanding undisturbed natural processes. Despite the connection between anthropogenic activity and coastal groundwater issues, and the consequential overrepresentation of urban basins in coastal groundwater studies, perceptual (or conceptual) models of coastal groundwater rarely include anthropogenic influences aside from pumping (e.g., subsidence, land use change). Taking a holistic view on coastal groundwater flows, we have developed an editable perceptual model illustrating the current understanding, including both natural and anthropogenic drivers. As SGD and SWI in new areas of the globe are studied, we advocate for researchers to utilise and further edit this perceptual model to openly communicate our process understanding and study assumptions.

Modelling precipitation inputs in mountainous terrain is challenging for water resource managers given sparse monitoring sites and complex physical hydroclimatic processes. Government of Alberta weather station uncorrected and bias-corrected precipitation datasets were used to examine elevational precipitation gradients (EPGs) and seasonality of EPGs for six South-Saskatchewan River headwater sites (alpine, sub-alpine, valley). January EPG from valley to alpine sites (730 m elevation difference) using uncorrected precipitation was 19 mm/100 m. Corrected EPG was approximately three times greater (61 mm/100 m). The valley received more precipitation than the alpine (inverse EPG) in late spring and summer. A seasonal signal was present whereby all sites demonstrated 50%–70% lower summertime precipitation relative to winter months, with the greatest seasonal variance at the alpine site. Winter watershed-level spatialized precipitation volume was compared to modelled snow water equivalent (SWE) associated with two late-winter airborne lidar surveys. Uncorrected volumes (2020: 64.0 × 10⁶m³, 2021: 63.2 × 10⁶m³) were slightly higher than modelled mean SWE (2020: 51.6 × 10⁶m³, 2021: 44.2 × 10⁶m³) whereas bias-corrected (2020: 120.5 × 10⁶m³, 2021: 119.7 × 10⁶m³) almost doubled the estimate. Corrected precipitation is assumed closer to the true value. Cumulative sublimation, evaporation and snowmelt losses result in ground-level snowpack yield that deviates from total atmospheric precipitation in an increasingly negative manner. The 2020/2021 simulations suggest wintertime atmospheric precipitation exceeds late-winter snowpack accumulation by up to 57% and 63%, respectively. A loss of 16 × 10⁶m³ (7%) watershed SWE from the alpine zone was partially attributed to redistribution downslope to the treeline-ecotone. Physical snowpack losses from sublimation and melt, or modelling uncertainty due to precipitation correction and alpine snow-density uncertainties can also contribute to observed discrepancies between in situ SWE and cumulative precipitation. Ignoring bias-correction in headwater precipitation estimates can greatly impact headwater precipitation volume estimates and ignoring EPG seasonality is likely to result in under-estimated winter and over-estimated summer yields.

Suspended sediment is one of the major contributors to the total sediment load transported by rivers. Suspended sediment transport is highly variable both in time and space, driven by complex interactions between tectonics, climate change, and anthropogenic activity. Large waterways often underlie strong anthropogenic impact, for example, to ensure navigability and pursue economic objectives. However, stability and ecological integrity of the river system are equally important. For both, economic and ecologic objectives, processes related to the transport, deposition, and resuspension of fine sediment must be understood and quantified. Starting in the 1960s the Waterways and Shipping Administration (WSV) and the Federal Institute of Hydrology (BfG) started an extensive monitoring program to quantify suspended sediment transport in German navigable waterways. To cope with the large spatiotemporal variability of suspended sediment transport the WSV combined work-daily single point measurements and infrequent multi-point measurements. The aim of our study is to quantify the vertical and lateral variability of suspended sediment transport along the largest waterway in Germany, the Rhine River, and investigate drivers of cross-sectional variability. We link results from multi-point measurements with single-point measurements to assess the representativity of surface samples compared to cross-sectional means of suspended sediment concentration. The comparison of these two monitoring programs reveals that surface samples strongly underestimate suspended sediment loads. Main drivers that could be quantified are vertical gradients of suspended sediment concentration by means of Rouse profiles and lateral variability which is partially explained by mean channel curvature and related to the underestimation of suspended sediment transport relying on surface samples. Further, we observe that the magnitude of lateral variability is comparable to vertical variability, but often neglected in suspended sediment monitoring. Our study contributes to the refinement of existing monitoring schemes and shows how empirical data verifies and falsifies transport dynamics and processes.

There is relatively little research on infiltration into seasonally frozen soils on mountain hillslopes and few evaluations of infiltration model performance in this environment exist. As a result, the application of existing infiltration estimation methods developed in level environments is uncertain for estimating spring runoff in mountain basins. A field study was conducted in the Canadian Rockies using 8 years of snowpack, liquid soil moisture, and temperature profile observations from steep north-facing and south-facing slopes. Seasonal infiltration was calculated using soil freezing characteristic curves, timeseries of soil volumetric water content and temperature. Infiltration was found to primarily follow the limited case postulated by Popov (1972), with only 1 year at one site undergoing unlimited infiltration where nearly all meltwater infiltrated. Infiltration was estimated using an equation for the limited case developed from extensive observations of seasonal infiltration, initial soil saturation, and peak SWE in Canadian prairie agricultural fields. Whilst this equation accurately estimated infiltration depths on these mountain hillslope sites, it was unsuitable for application due to a statistical association between its driving variables. Initial soil saturation had no influence on infiltration depths at these sites and so a simpler single-variable infiltration equation to estimate infiltration depths based on peak SWE was developed and found to have good predictive capability. Alternative approaches using modelled cumulative melt or infiltration opportunity time also had good predictability. Runoff depths estimated from a water balance, assuming negligible evaporation and sub-surface drainage, were reliably predicted using peak SWE or cumulative melt depths by single-variable infiltration equations in the absence of soil moisture, texture, aspect, or slope information. The results provide insights into estimating snowmelt runoff on hillslopes from snowpack accumulation that has been observed in cold region mountains, despite the complexity of hillslope hydrology and frozen soil infiltration processes.

Monitoring river connectivity across large regions is essential for understanding hydrological processes and environmental management. However, comprehensive assessments of river connectivity are often hindered by inaccurate dam databases, which are biased towards larger dams while overlooking smaller or low-head dams. To enhance the accuracy of river connectivity assessments, we developed three advanced convolutional neural networks (CNNs; YOLOv5, Advance-You Only Look Once [YOLO], and Faster R-CNN) to accurately classify dams and evaluate river connectivity using high-resolution (1 m) remote sensing imagery. The evaluation results showed that Advance-YOLO performs best with an average mean average precision (mAP) of 86.6%, while Faster R-CNN performs mediocrely with an average mAP of 77.9%. Applying the well-trained model in the Tarim River Basin (China), one of the largest inland river basins around the globe, we found that there are currently 135 dams in total on the Tarim River and its sources. Conversely, the existing public dam database underestimates 85.9% of the dams. Notably, we found a 14.3% decline in river connectivity of the Tarim River over the past decade, and the current dam density of the Tarim River and its four source rivers is 1.12 per 10 000 km². However, the existing public dam database overestimated river connectivity by 83.9%. The model developed here enhances river connectivity assessment across larger areas over a long period, thereby fostering more advanced research on hydrological processes and effective water resource management.

Mountain ranges cover approximately 24% of the Earth’s land mass. These environments have a special relevance in terms of global water supply. However, historically mountain groundwater processes have been generally overlooked or poorly understood, especially in the Andes cordillera. With this in mind, this work aimed to study hydrological processes in four Andean, semi-arid headwater river basins. Along with monthly stable isotope data collection, we carried out a synoptic surface water sampling programme in each river on four specific dates for ³H analysis. The latter indicated water of similar age in the rivers of three sub-basins (Derecho, Cochiguaz, Incaguaz), but much older in the fourth (Toro). We assessed different possible explanations for these differences such as effects of past mining activities (El Indio mine), physiographic factors and snow accumulation and glacier related factors, but none of these were satisfactory. Instead, our findings point to the activation of faults in response to seismic activity, which induces pumping of fluids (water) from deeper zones, facilitating exfiltration processes in the Toro River sub-basin. This explains the presence of surface waters older than those associated with current meteoric processes. Such geological process should be assessed and eventually accounted for when studying mountain hydrogeological processes, especially in high fractured areas with direct or indirect evidence of geothermal activity.

This commentary explores the potential of artificial intelligence (AI) to transform hydrological modelling workflows. We introduce a prototype AI-assisted framework called INDRA (Intelligent Network for Dynamic River Analysis) that leverages a multi-agent architecture composed of specialised large language models (LLMs) to assist in model conceptualization, configuration, execution, and interpretation. INDRA integrates with CONFLUENCE, a comprehensive modelling framework, to provide context-aware guidance and automation throughout the modelling process. We discuss the opportunities and dangers of AI-augmented research frameworks, emphasising the importance of maintaining human oversight while harnessing AI's potential to enhance efficiency, reproducibility, and scientific understanding. We argue that AI-assisted workflows could democratise advanced hydrological modelling, enabling researchers worldwide to address critical water resources challenges, particularly in understudied regions. While acknowledging potential biases and risks, we advocate for responsible AI integration to catalyse a new paradigm in hydrological science.

While tracing the sources of fluvial sediment using carbon and nitrogen stable isotopic ratios (δ¹³C and δ¹⁵N) has progressed significantly over the last two decades, the conservativeness of these tracers remains questionable. Recent work indicates that δ¹³C and δ¹⁵N alterations in streambed deposition zones likely represent the largest source of uncertainty impacting usefulness of the isotopic ratios as tracers. Here we report a 14-year dataset of δ¹³C and δ¹⁵N of fluvial sediment from a streambed-dominated basin in Kentucky, USA, and employ empirical model decomposition (EMD) to identify dominant temporal trends that may impact conservativeness. Results from EMD show significant seasonality of δ¹³C and δ¹⁵N for sediment as well as underlying multi-year variation. The seasonal and multi-year variance account for 72% and 50% of the total data variation for δ¹³C and δ¹⁵N, respectively. The prominent seasonality for δ¹³C and δ¹⁵N show a mean intra-annual change of 0.6‰ and 1.1‰, respectively, and the seasonal change is attributed to algal accrual and organic matter turnover in the streambed sediment deposits. Mixing model simulations show that the mean streambed isotopic ratios should be separated from other sediment sources by 3.0‰ and 3.6‰ for δ¹³C and δ¹⁵N, respectively, to achieve 90% accuracy in source apportionment when the isotopic ratios are used independently; and the mean streambed value of both isotopic ratios should be separated from other sediment sources by 3.0‰ when δ¹³C and δ¹⁵N are used in combination. Our results lead to the recommendation that isotope ratios of sources be separated by at least 3‰ when the streambed is expected to be a prominent sediment source, which far exceeds the prior recommendation of 1‰ mean separation of sources.

Under more frequent, extreme global drought events, the use of stable isotopes to quantify soil evaporation losses (SEL) is of great significance for understanding the water supply capacity from soil to plants. During March 2017–September 2019, we continuously monitored meteorological factors, soil temperature (ST) and humidity, and collected precipitation and soil water stable isotope data. The Craig-Gordon (C-G) and line-conditioned excess (lc-excess) coupled with the Rayleigh fractionation (RL) models were used to quantify SEL in subtropical secondary forests. The results showed: (1) the theoretical evaporation line (EL) slope negatively correlated with air temperature (AT). Water source isotopic values are more positive in autumn and more negative in spring. The aridity index (AI) and soil evaporation loss ratio (f) from both models indicated drier conditions during March–September 2018 compared to 2017 and 2019; (2) comparative analysis showed the C-G model agreed more closely with measured evapotranspiration (ET₀) and water surface evaporation (E) than the RL model, indicating better suitability of the C-G model in the study region; (3) because the “inverse temperature effect” of the precipitation isotopes, the linear fitting method was not suitable for determining the water source in spring, summer, autumn, and on the annual scale, while the linear fitting method was consistent with the basic principle of soil evaporation in winter. Thus, the theoretical method was more suitable for determining the EL slope in such regions; (4) because of the different fundamentals, the C-G model positively correlated with AT and negatively correlated with relative humidity (h), while the RL model showed the opposite trends, indicating different applicability. The SEL is influenced by soil thickness, atmospheric evaporation and soil water supply capacity. These findings support stable isotope application techniques for quantifying SEL and are crucial for analysis of soil water resources in subtropical secondary forests.