Lake Water Level Research Articles

Impact of Rivers on Seasonal Changes in Water Level

This study was conducted regarding the impact of rivers on Darbandikhan Lake's seasonal water level change. The aim of this study is to illustrate the impact of rivers on the lake through seasonal water level change from its tributaries. To achieve the research objectives, descriptive and quantitative methods were used. The findings showed that the highest water level of the lake was 467.72 m (a.s.l.) in Spring and that its storage capacity and area were 1.2554 billion m3 and 46.26 km2, respectively. In contrast, the lowest water level of the lake was 458.62 m (a.s.l.) in Autumn, with storage capacity and area of 908.2 million m3 and 32.38 km2, respectively. The river's share of discharge into the lake varied annually. The Tanjaro River averaged 6 m3/s (19.6%), the Chaqan River 0.47 m3/s (1.52%), the Zalm River 1.17 m3/s (3.77%), the Sirwan River 20.55 m3/s (66.28%), and the Zmkan River 2.74 m3/s (8.83%), all of which contributed to the lake annually. River discharge was influenced by physical factors such as topography, slope, and climatic elements, in addition to human factors such as water resource management policies, population and land use/land cover in the study area.

Analysis of Optimal Water Levels in the Great Lakes Based on A Comprehensive Satisfaction Function

The purpose of this paper is to analyze the optimal water levels in the Great Lakes to meet the needs of multiple stakeholders and thus achieve sustainable development in the region. The Weight-Function-Synthesis-Optimization (WFSO) model was used in this study. By classifying the water level needs of different interest groups, their respective satisfaction functions are established, and the overall satisfaction function is formed by weighting. In addition, the quadratic polynomial function is used to fit the optimal monthly water level time series of Lake Ontario, and the fitting effect is good, fully reflecting the periodic and seasonal changes in the water level of Lake Ontario. By considering the priorities of various stakeholders, this paper defines an overall satisfaction metric, "allsa," and identifies a time series of monthly optimal water levels for Lake Ontario and beyond. The ideal water level was determined to be 74.9 meters. The study results suggest that maintaining this water level can maximize the satisfaction of ecosystems, transportation, real estate, terminals, and hydropower companies, and promote sustainable development of the region.

Local and catchment-scale effects of water retention measures at Lake Velence

AbstractClimate change manifested its adverse impacts last year, with an extreme drought leading to a drastically low water level in Lake Velence, Hungary. Nature-based solutions have the potential to alleviate these impacts locally. While a few initiatives have been implemented in Hungary, widespread adoption of these solutions is expected to be a goal for the more distant future.This research focuses on one catchment at Lake Velence to evaluate decision-maker's readiness and urban water management infrastructure for broadly implementing nature-based solutions. Methods include delineating the stormwater system and creating a numerical model to evaluate rainfall-runoff processes and the possible impacts of nature-based retentions. Surveys among local mayors were conducted to assess their perception of existing water infrastructures and implementations of nature-based solutions. Its widespread use may become significant, but its effect on the lake's water level remains negligible.

Insights into permafrost aggradation and thawing: A case study of Zonag Lake on the Qinghai–Tibet Plateau using SAR interferometry

AbstractLakes on the Qinghai–Tibet Plateau (QTP) are important indicators of climate change. The water level and area of Zonag Lake have increased sharply since 2000, and the west lake bottom area was exposed after an outburst of flood in September 2011. The drained basin was exposed to the cold environment, causing rapid permafrost forming. In this paper, time‐series synthetic aperture radar interferometry (InSAR) was utilized to monitor the surface deformation of the drained basin using Sentinel‐1A images within a 4‐year period (2017–2020). This research focused on characterizing the spatial and temporal dynamics of ground deformation and exploring changes in the permafrost base around the drained basin based on the deformation information. The experimental results revealed spatially variable ground deformation, with long‐term deformation rates ranging from −78.3 mm/year to 72.8 mm/year and seasonal amplitudes of 0–14 mm. Subsidence was prominent in the lakeshore, the thermokarst region and the slope drainage area, indicative of the thawing of ice‐rich permafrost. Conversely, uplift was concentrated in the exposed unconsolidated sediment region, with a maximum displacement of 110 mm over the nearly 4‐year observation period, suggesting ground ice aggradation. The uplift information was used to estimate the change in the permafrost base in the drained basin, yielding a maximum value of 4 m. The estimated permafrost base change closely aligned with simulated results, with an error of 0.28 m. This study demonstrates the capability of InSAR in monitoring permafrost stability and improving the understanding of the dynamic permafrost formation process.

Modelling And Research on Water Level Control of Great Lakes Based on Neural Network PID Algorithm

The Great Lakes are situated in the border region between the United States and Canada, which has a significant impact on the climate and the lives of those living in the surrounding areas. The objective of this paper is to establish a network of the Great Lakes through Pearson's correlation coefficient analysis and to construct a two-tank water level model based on a PID control system in order to effectively manage the dynamics of the Great Lakes. Firstly, the strength and direction of the linear relationship between two variables is quantified through Pearson's correlation coefficient analysis, which involves the collection of observational data and the calculation of mean values. This analysis serves as a fundamental basis for predictive modelling and hypothesis testing. Secondly, based on the flow balance principle, mathematical expressions are constructed to simulate the water flow, and a PID control system is constructed to achieve optimal water level maintenance. By analysing the Pearson's correlation coefficient, the interrelationships among the variables in the Great Lakes network can be understood, thereby providing guidance for scientific research and decision-making. The results demonstrate that the constructed two-tank water level model combined with the PID control system and SHAP algotithm can effectively manage the water level of the Great Lakes and achieve optimal water level regulation.

Forecasting Lake Nokoué Water Levels Using Long Short-Term Memory Network

The forecasting of hydrological flows (rainfall depth or rainfall discharge) is becoming increasingly important in the management of hydrological risks such as floods. In this study, the Long Short-Term Memory (LSTM) network, a state-of-the-art algorithm dedicated to time series, is applied to predict the daily water level of Lake Nokoué in Benin. This paper aims to provide an effective and reliable method to enable the reproduction of the future daily water level of Lake Nokoué, which is influenced by a combination of two phenomena: rainfall and river flow (runoff from the Ouémé River, the Sô River, the Porto-Novo lagoon, and the Atlantic Ocean). Performance analysis based on the forecasting horizon indicates that LSTM can predict the water level of Lake Nokoué up to a forecast horizon of t + 10 days. Performance metrics such as Root Mean Square Error (RMSE), coefficient of correlation (R2), Nash–Sutcliffe Efficiency (NSE), and Mean Absolute Error (MAE) agree on a forecast horizon of up to t + 3 days. The values of these metrics remain stable for forecast horizons of t + 1 day, t + 2 days, and t + 3 days. The values of R2 and NSE are greater than 0.97 during the training and testing phases in the Lake Nokoué basin. Based on the evaluation indices used to assess the model’s performance for the appropriate forecast horizon of water level in the Lake Nokoué basin, the forecast horizon of t + 3 days is chosen for predicting future daily water levels.

Untangling the coupling effect of water quality and quantity on lake algal blooms in Lake Hulun from a dual perspective of remote sensing and sediment cores

Algal blooms and sediment diatoms are crucial indicators of lake water ecology, influenced by water quantity and quality. However, the coupled effects of water quality and quantity changes on algal blooms are still unclear, especially for lakes in cold and arid regions. This study assessed the long-term variations in algal blooms in Hulun Lake using a novel approach combining remote sensing and sediment core samples for diatom analysis. Two mutation points from the structural change test were identified in approximately 2000 and 2010 for algal bloom area (MBE) and sediment diatom richness, indicating asynchronous algal blooms. A structural equation model (SEM) demonstrated that water level (WL) changes were the dominant influencing factor, co-driving the variations in algal blooms and sediment diatoms in conjunction with total nitrogen (TN), total phosphorus (TP), and chemical oxygen demand (COD). The results revealed nonlinear relationships between the lake WL, TN, Chla, and COD. The water level of 543 m emerged as a critical threshold affecting the relationship between water quality and quantity. Distinct differences in this relationship were observed when water levels were above or below this threshold. These variations became particularly pronounced during periods of high and low water levels. The results provide novel insights into the dynamics of algal blooms and can further support lake ecosystem conservation and management.

Risk assessment of a glacial lake with abruptly slowing expansion, Jiongpu, Southeastern Tibet

The expansion of potentially hazardous glacial lakes is a symptom of global warming during this interglacial period. A pertinent example is the Jiongpu glacial lake in southeastern Tibet, the area of which has expanded approximately fivefold in the last half-century. However, recently, the glacier tongue has retreated to a high steep slope, and the rate of retreat of the glacier and expansion of the lake have temporarily slowed. The risk of a glacier tongue landslide after glacier detachment and subsequent glacial lake outburst flood (GLOF) needs to be assessed. In this study, we employed a combination of unmanned aerial vehicle (UAV), sonar, geological radar, remote sensing, field investigation, sampling, drilling, and dating techniques to determine the critical parameters of potential GLOFs, including glacier tongue geometry, lake bathymetry, and moraine dam geometry and composition. Utilizing empirical models and multiphase flow models, we identified the most hazardous triggers and simulated the processes of a glacier tongue landslide into the lake, moraine dam overtopping by a displacement wave, and subsequent flood evolution. The results showed that the most hazardous trigger in volume is a glacier tongue landslide, accounting for 58.29 % of all triggers associated with potential GLOFs. Lapped by the largest glacier tongue landslide impulse wave, the moraine dam would not fail because the minimum safety factor is approximately 1.66 ± 0.7 according to empirical methods and geological slope simulation. However, overtopping would occur, resulting in a peak discharge of approximately 9740 ± 4137 m3/s at the moraine dam based on r.avaflow calculations. The flood would reach the densely populated Jinling township and inundate approximately 46 ± 4.55 % of the houses according to HEC-RAS. Reducing the water level of the glacial lake represents an effective strategy for mitigating potential losses. This concise, physics-based method effectively assesses GLOF triggers and processes and can be applied to risk assessments of other expanding glacial lakes.

Changes monitoring in Hongjiannao Lake from 1987 to 2023 using Google Earth Engine and analysis of climatic and anthropogenic forces

The research aimed to quantify the lake area dynamics, evaluate the changes in distance and rate of lake shorelines quantitatively and spatially and investigate the key factors influencing the Hongjiannao Lake (HL) area shrinkage. The study used remote sensing (RS) data from Landsat TM/ETM+ and OLI images and Google Earth Engine (GEE), a cloud platform for obtaining the lake surface area and island information from 1987 to 2023. A modified normalized difference water index (MNDWI) was applied to water area extraction. Digital Shoreline Analysis System (DSAS) was employed to assess net shoreline movement (NSM) and depict the lake shoreline length and rate changes. Furthermore, the water level was derived by ASTER GDEM V2 using the waterline method and lake boundaries. Six climatic features (temperature, precipitation, potential and actual evaporation, aridity index, and actual water difference) were investigated to find the driving factors of lake area shrinkage by correlation and factor analysis. The results reveal that during 1987–2023, the HL area has undergone four separate phases: stable (1987–1997), shrinkage (1998–2015), growth (2016–2019), and reduction (2020–2023). The most substantial negative change (−7.45%) in the HL area was observed in 1998. NSM analysis demonstrates that the lake has experienced both expansion and shrinkage at various times and locations. According to Water Balance Method, the water volume of HL exhibited variations, ranging from −0.1895 to −0.009 km³. The average yearly change in lake volume, water level, and area displayed similar characteristics with high inconstancy. Correlation and factor analysis of lake area and climatic factors demonstrate that higher precipitation, low temperatures, less potential evaporation level, lower actual evaporation rates, and more minor differences in water levels are associated with an increase in lake area. In contrast, the opposite conditions lead to a reduction in lake size.

Paleoenvironmental reconstruction of Lake Van and Lake Erçek over the last millennium using varved sediments (Eastern Türkiye)

Detailed lithological and geochemical studies of the bottom sediments were conducted based on the original sampling of brackish adjacent lakes: Lake Van and Lake Erçek in Eastern Anatolia, Türkiye. Seasonal variations in the accumulation of weather-driven element content were determined using high-resolution Synchrotron Radiation X-Ray Fluorescence (SR-XRF) analysis of the annual layers in the solid prepare. The results were utilized to approximate the regional paleo-temperature of both lakes and the water level of Lake Van. Visual counting of the varve as well as annual oscillation of measured geochemical series was employed to create a time scale. The element content series were synchronized with recent changes in external climatic environments and lake water level fluctuations. Subsequently, the geochemical series were transformed into climate units, using calibration on synchronous meteorological data. Finally, the original quantitative reconstruction of the regional climate variations was performed over the last millennium period with annual time resolution; for Lake Erçek (840 years ago) and Lake Van (1300 years ago). It was revealed that the sediments of these adjacent lakes (Lake Van and Lake Erçek), located 30 km apart from each other, provided synchronous responses to regional environmental changes, which confirmed the reliability of the reconstructions.

Active remote sensing data and dispersal processes improve predictions for an invasive aquatic plant during a climatic extreme in Great Lakes coastal wetlands

Invasive aquatic plants pose a significant threat to coastal wetlands. Predicting suitable habitat for invasive aquatic plants in uninvaded yet vulnerable wetlands remains a critical task to prevent further harm to these ecosystems. The integration of remote sensing and geospatial data into species distribution models (SDMs) can help predict where new invasions are likely to occur by generating spatial outputs of habitat suitability. The objective of this study was to assess the efficacy of utilizing active remote sensing datasets (synthetic aperture radar (SAR) and light detection and ranging (LiDAR) with multispectral imagery and other geospatial data in predicting the potential distribution of an invasive aquatic plant based on its biophysical habitat requirements and dispersal dynamics. We also considered a climatic extreme (lake water levels) during the study period to investigate how these predictions may change between years. We compiled a time series of 1628 field records on the occurrence of Hydrocharis morsus-ranae (European frogbit; EFB) with nine remote sensing and geospatial layers as predictors to train and assess the predictive capacity of random forest models to generate habitat suitability in Great Lakes coastal wetlands in northern Michigan, USA. We found that SAR and LiDAR data were useful as proxies for key biophysical characteristics of EFB habitat (emergent vegetation and water depth), and that a vegetation index calculated from spectral imagery was one of the most important predictors of EFB occurrence. Our SDM using all predictors yielded the highest mean overall accuracy of 88.3% and a true skill statistic of 75.7%. Two of the most important predictors of EFB occurrence were dispersal-related: 1) distance to the nearest known EFB population (m), and 2) distance to nearest public boat launch (m). The area of highly suitable habitat (pixels assigned ≥0.8 probability) was 74% larger during a climatically extreme high water-level year compared to an average year. Our findings demonstrate that active remote sensing can be integrated into SDM workflows as proxies for important drivers of invasive species expansion that are difficult to measure in other ways. Moreover, the importance of a proxy variable for endogenous dispersal (distance to nearest known population) in these SDMs indicates that EFB is currently spreading, and thereby less influenced by within-site dynamics such as interspecific competition. Lastly, we found that extreme climatic conditions can dramatically change this species’ niche, and therefore we recommend that future studies include dynamic climate conditions in SDMs to more accurately forecast the spread during early invasion stages.

Invasive-Weed-Optimization-Based Extreme Learning Machine for Prediction of Lake Water Level Using Major Atmospheric–Oceanic Climate Scenarios

Fresh water lakes are vulnerable assets that need to be protected against manmade/natural challenges like climate change and anthropogenesis activities. This study addresses the predictability of the lake water level changes based on the knowledge acquired directly from the climate data. Two fresh water lakes named Lake Iznik and Uluabat, located in Turkey, are addressed. Time series of the lake water levels during October 1990–September 2019 at a monthly scale, along with the corresponding anomalies of 24 Large-Scale Atmospheric–Oceanic Oscillations (LSAOOs) from around the globe, are used in the analysis. The relationship between variables and the structure of the models are initially acquired based on the significance of the dependence between climate indices and lake water levels with consideration of the significance of the Spearman rank-order coefficient. Then, the time series are divided into training (80%) and testing (20%) sets. The Extreme Learning Method (ELM), enhanced with the genetic algorithm (ELM-GA) and Invasive Weed Optimization (ELM-IWO), is then used in the predictive models. Based on the results, Lake Uluabat showed a stronger teleconnection with LSAOOs, while the ELM-GA for Lake Iznik and ELM-IWA for Lake Uluabat depicted the best performance in the prediction of lake water levels. Comparison of the enhanced ELM-IWO to the corresponding ELM-GA illustrates that the ELM-IWO reveals more acceptable results owing to its flexible nature.

A sedimentary DNA perspective about the influence of environmental and food‐web changes on the microbial eukaryotic community of Lake Biwa

Abstract The impacts of environmental change on Lake Biwa have been explored for decades, with water monitoring and palaeolimnological studies revealing how environmental forcing, including climate warming, eutrophication, water level manipulation and human manipulation of fish populations, has influenced the food web of Lake Biwa. However, these studies have rarely accounted for microbial food‐web components. This knowledge gap is mostly due to the lack of time series spanning more than a couple of decades, coupled with the high taxonomical expertise required to identify organisms belonging to very diverse groups. The use of a sedimentary DNA approach allows for the reconstruction of past changes in the diversity, composition and structure of the microbial eukaryotic community of aquatic systems. The application of 18S metabarcoding has been proven successful to describe the response of unicellular eukaryotes (protists) and aquatic fungi in lake ecosystems, encompassing a large taxonomic and functional diversity such as phototrophs, heterotrophs and mixotrophs. We applied 18S metabarcoding to 31 sediment core samples from Lake Biwa, spanning the past 100 years and explored the response of microbial eukaryotic communities to changes in multiple environmental stressors, including nutrient levels, lake water level, climate, as well as fish and zooplankton biomass for the period 1973–2010. We found that the manipulation of the water level and changes in fish community composition were the primary factors impacting (indirectly) the structure of the lake microbial eukaryotic community with minor, but significant, effects of climate warming and phosphorus levels. Co‐occurrence network analysis highlighted the potential food web impacts on the microbial eukaryotic community, suggesting that organisms from this compartment were impacted by both bottom‐up and top‐down processes.

Attribution of hydrological droughts in large river-connected lakes: Insights from an explainable machine learning model

Large lakes play an important role in water resource supply, regional climate regulation, and ecosystem support, but they face threats from frequent extreme drought events, necessitating an understanding of the mechanisms behind these events. In this study, we developed an explainable machine learning (ML) model that combines the Bayesian optimized (BO) long short-term memory (LSTM) model and the integrated gradients (IG) interpretation method to simulate and explain lake water level variations. In addition, the hydrological drought trends and extreme drought events in Poyang Lake from 1960 to 2022 were identified using the standardized water level index (SWI) and run theory. The analysis revealed that the frequency of hydrological droughts in Poyang Lake increased from 1960 to 2022, especially in the autumn after 2003. By selecting the flows of the catchment and the Yangtze River as the input features, the BO-LSTM model accurately predicted the water level of Poyang Lake. The IG method was then used to interpret the prediction results from three aspects: the importance ranking of the input features, their roles in the seasonal drought trends, and their roles in extreme drought events. The results indicate that (1) the most influential factor affecting the water level of Poyang Lake was the inflow of the Ganjiang River in the catchment. (2) The increase in the lake outflow caused by the Yangtze River's draining effect was the reason for the intensification of the autumn drought in Poyang Lake. (3) The extreme hydrological drought events were primarily caused by low catchment inflows. Overall, this research provides a new approach that balances prediction accuracy with interpretability for predicting and understanding the hydrological processes in large river-connected lakes. Moreover, this method was also applied to the attribution analysis of hydrological drought in Poyang Lake, providing theoretical support for regional water resource management.

Evaluating D-InSAR performance to detect small water level fluctuations in two small lakes in Sweden

Monitoring lake water level fluctuations is critical for managing water resources, predicting the impacts of climatic change, and preserving ecosystem services lakes provide. However, traditional gauging stations are insufficient to monitor all lakes worldwide due to the large number of existing lakes, the challenges of installation and maintenance, and the remote locations of some. Although satellite altimetry is an alternative for measuring water levels, it cannot monitor small lakes effectively. This study evaluates the potential of Differential Interferometric Synthetic Aperture Radar (D-InSAR) for tracking minor water level changes in small lakes, a method more typically used in wetland studies. We investigate two Swedish lakes using Sentinel-1A and Sentinel-1B data from 2019, generating six-day interferograms and filtering out those with in situ water level changes exceeding one phase cycle. Our results show that D-InSAR can detect small water level changes with Lin’s correlations up to 0.63 and 0.89 and RMSE values of approximately 9 and 4 mm, respectively. These results evidence the potential of future L-band SAR missions with larger wavelengths, such as the NASA-ISRO SAR (NISAR) of the National Aeronautics and Space Administration (NASA) and the Indian Space Research Organisation (ISRO), to track water level changes in lakes and aid water tracking missions such as the SWOT (Surface Water and Ocean Topography).

Preparation of CMC-MF/FeO(OH) by growing ferric nanoparticles on melamine foam in-situ for effective phosphate removal

Nanocomposite adsorbents composed of a bulk matrix and metal nanoparticles have garnered significant attention for phosphorus (P) removal. However, many nanocomposite adsorbents have been synthesized in harsh environments involving high temperature or high energy consumption, which do not meet industrial requirements; on the other hand, lots of adsorbents are synthesized by directly depositing metal hydroxides on bulk matrixs, leading to the active components shedding from the matrix due to the lack of chemical anchoring sites to immobilize metal oxide. To overcome the limitations of the existing methods, we synthesized a nanocomposite adsorbent named CMC-MF/FeO(OH) using a facile method in a mild environment. It involved coating an anchoring film prepared with carboxymethyl cellulose sodium (CMCNa) on melamine foam (MF) for in-situ growth of ferric nanoparticles. Our results demonstrated that the CMC-MF/FeO(OH) adsorbent showed promising efficiency in phosphate removal. The adsorption process was effectively described by Elovich kinetic model and Freundlich isotherm model. The adsorption mechanism was ascribed to the chemisorption by inner-sphere complex along with physical adsorption. A maximum adsorption capacity of 31.579 mg P/g which calculated by Langmuir isotherm model was achieved at 45 °C and natural pH condition in experimental range. The adsorbent exhibited effective P adsorption even in harsh environments, with the process being spontaneous and exothermic. In column applications, the CMC-MF/FeO(OH) adsorbent reduce the P content of wastewater from 4 mg P/L to 0.5 mg P/L within a short empty bed contact time of 190.95 s. Furthermore, the adsorption capacity reached 8.87 mg P/g in a column study, and trace P levels in lake water decreased to undetectable levels in a lake experiment, effectively inhibiting algae growth.

Modeling and Analysis of Lake Water Levels Using System Dynamics and Principal Component Analysis

Understanding the intricate dynamics of the Great Lakes water levels, shaped by environmental factors like rainfall and evaporation, is crucial for devising sustainable water management strategies. This study develops a model to manage the Great Lakes' water levels by analyzing the interconnectedness of the lake network. Utilizing principal component analysis, key variables like rainfall and evaporation were evaluated for their impact on water levels. System dynamics models integrated with flow conservation laws allowed for a detailed hydrodynamic assessment. The study introduces a "narrow seam method" for dry river scenarios, enhancing model accuracy. Annual water level trends from 2000 are depicted, offering insights into future water management practices in response to environmental changes.

Modeling of water level trends and characterizing potential influencing factors in Lake Baringo in Kenya

Water plays a significant role in every sector of the ecosystem. The fluctuation of the water levels in lakes is influenced by natural and man-made factors within the water catchment. Lake Baringo, which has no visible outlet, has been rising drastically, causing panic among native communities and businesses on the shores of the lake. Using GIS and Remote Sensing, this study intends to analyze the changes in water lakes using the Automatic water extraction index (AWEI), determine the causes of the fluctuation using Land use Land cover, land surface temperature, soil erosion, and siltation in the lake basin and the lake respectively, precipitation, and later predict of the water level for the year 2030 using the MOLUSCE tool. The tool utilizes an artificial neural network and cellular automata to analyze land use and land cover conveniently. It was found that the lake’s water level has been increasing drastically over the years, and the leading causes of the fluctuations were increased rainfall and human activities within the water basin. There are visible increased human activities within the water basin, such as agriculture, deforestation, settlement, and urbanization. It was also found that there will be a further increase in water level in 2030. With all the above results, it is recommended that better policies be made to conserve the water basin effectively, and a plan should be drawn to re-delineate the new riparian buffer.

Multi-year climate memory in shallow lake water levels

Landscape hydrologic memory of meteorological cycles can have an important impact on catchment hydrological responses by propagating clustering of wet or dry conditions into extreme events. The non-linear and hysteretic hydrologic response functions driven by memory are often only studied at shorter temporal scales (event, season) despite larger interannual hydrologic responses evident in some systems. Within the Canadian Boreal, lakes and lake water levels provide an important indicator that can be used to assess the role of landscape memory on catchment hydrological function. Landscape memory has also been hypothesised to control the hydrological dynamics of shallow lake ecosystems that are also important biogeochemically and ecologically. Here we combine measurements of lake water levels in 26 lakes of varying type at varying temporal frequencies within the glaciated sub-humid Boreal Plain, to examine the impact and variability of interdecadal, decadal, multi-year, and intra-annual memory on lake water levels. We show multi-annual hysteresis of precipitation-lake water levels with varied characteristics in space and time. These spatial variations in landscape memory are driven by differences in storage capacities controlled by heterogeneity in glacial landforms, wetland-forest landcover and lake properties. Thus, the propensity for drought years or wet years to persist or accumulate into extreme landscape drying or wetting varies significantly between different lake-landscape characteristics. We show how landscape memory is crucial to project lake water levels by defining spatial variability of the impact of periods of meteorological drought and deluge vital for understanding system sensitivity, duration of recovery and in turn infer resilience on Boreal Plain hydrology.

Predicting flood stages in watersheds with different scales using hourly rainfall dataset: A high-volume rainfall features empowered machine learning approach

Accurate prediction of instantaneous high lake water levels and flood flows (flood stages) from micro-catchments to big river basins are critical for flood forecasting. Lake Carl Blackwell, a small-watershed reservoir in the south-central USA, served as a primary case study due to its rich historical dataset. Bearing knowledge that both current and previous rainfall contributes to the reservoirs' water body, a series of hourly rainfall features were created to maximize predicting power, which include total rainfall amounts in the current hour, the past 2 h, 3 h, …, 600 h in addition to previous-day lake levels. Notedly, the rainfall features are the accumulated rainfall amounts from present to previous hours rather than the rainfall amount in any specific hour. Random Forest Regression (RFR) was used to score the features' importance and predict the flood stages along with Neural Network – Multi-layer Perceptron Regression (NN-MLP), Support Vector Regression (SVR), Extreme Gradient Boosting (XGBoost), and the ordinary multi-variant linear regression (MLR) together with dimension reduced linear models of Principal Component Regression (PCR) and Partial Least Square Regression (PLSR). The prediction accuracy for the lake flood stages can be as high as 0.95 in R2, 0.11 ft. in mean absolute error (MAE), and 0.21 ft. in root mean square error (RMSE) for the testing dataset by the RFR (NN-MLP performed equally well), with small accuracy decreases by the other two non-linear algorithms of XGBoost and SVR. The linear regressions with dimension reductions had the lowest accuracy. Furthermore, our approach demonstrated high accuracy and broad applicability for surface runoff and streamflow predictions across three different-sized watersheds from micro-catchment to big river basins in the region, with increases of predicting power from earlier rainfall for larger watersheds and vice versa.