Search

The objective of the current study is to present a comparison of techniques for the forecasting of low frequency climate oscillation indices with a focus on the Great Lakes system. A number of time series models have been tested including the traditional Autoregressive Moving Average (ARMA) model, Dynamic Linear model (DLM), Generalized Autoregressive Conditional Heteroskedasticity (GARCH) model, as well as the nonstationary oscillation resampling (NSOR) technique. These models were used to forecast the monthly El Niño-Southern Oscillation (ENSO) and Pacific Decadal Oscillation (PDO) indices which show the most significant teleconnection with the net basin supply (NBS) of the Great Lakes system from a preliminary study. The overall objective is to predict future water levels, ice extent, and temperature, for planning and decision making purposes. The results showed that the DLM and GARCH models are superior for forecasting the monthly ENSO index, while the forecasted values from the traditional ARMA model presented a good agreement with the observed values within a short lead time ahead for the monthly PDO index.

In locations where measured timeseries are not available or not sufficiently long, reliable predictions of the rainfall depth associated with a given duration and exceedance probability may be obtained through regional frequency analysis (RFA). The scientific literature reports on a large number of different approaches to RFA of rainfall extremes, each one characterized by specific advantages and disadvantages. One of the most common drawbacks is that regional models specifically refer to a single duration or a single exceedance probability. Second, several approaches require the definition of a homogeneous region where the model is trained; this leads to higher accuracy, but also the applicability of the model is limited to those locations that are hydrologically similar to the homogeneous group used in the training. Moreover, most models require filtering the available gauged stations based on the length of the measured timeseries to perform reliable frequency analysis. These aspects lead to discard a significant amount of data, which could turn out to be detrimental to the accuracy of the regional prediction in some cases.We set up a few alternative models aiming to investigate and discuss a different and innovative approach for RFA of rainfall extremes. We want to address three main research questions: (1) Can a single model represent the frequency of extreme rainfall events over a large, climatically, and morphologically complex geographical area? (2) Can a single RFA model handle all sub-daily durations (i.e., from 1 to 24h)? (3) Is it possible to exploit all available annual maximum series, regardless of their length (i.e., very short ones too)? We select a large study area that is located in north-central Italy. We make use of more than 2300 Annual Maximum Series of rainfall depth for different time-aggregation intervals between 1 and 24 hours, that have been collected between 1928 and 2011 in the Italian Rainfall Extreme Dataset (I2-RED). For each gauged station, several morpho-climatic descriptors are retrieved (e.g., minimum distance to the coast, elevation of orographic barriers, aspect, terrain slope, etc.) and used as covariates for the prediction models. Our models are based on ensembles of unsupervised artificial neural networks (ANNs) and are able to predict parameters of a Gumbel distribution for any location and any duration in the 1-24 hours range based on the morphoclimatic descriptors. Through the analysis of results over 100 gauged validation stations, a profitable discussion is enabled on the potential and drawbacks of ensembles of unsupervised ANNs for regional frequency analysis of sub-daily rainfall extremes.

Australian summer heat events have become more frequent and severe in recent times. Temperature-duration-frequency (TDF) curves connect the severity of heat episodes of various durations to their frequencies and thus can be an effective tool for analysing the heat extremes. This study examines Australian heat events using data from 82 meteorological stations. TDF curves have been developed under stationary and non-stationary conditions. Generalised Extreme Value (GEV) distribution is considered to estimate extreme temperatures for return periods of 2, 5, 10, 25, 50 and 100 years. Three major climate drivers for Australia have been considered as potential covariates along with Time to develop the non-stationary TDF curves. According to the Akaike information criterion, the non-stationary framework for TDF modelling provides a better fit to the data than its stationary equivalent. The findings can be beneficial in offering new information to aid climate adaptation and mitigation at the regional level in Australia.

Regional frequency analysis (RFA) aims to estimate quantiles of extreme hydrological variables (e.g. floods or low-flows) at sites where little or no hydrological data is available. This information is of interest for the optimal planning and management of water resources. A number of regional estimation models are evaluated and compared in this study and then used for regional estimation of flood quantiles at ungauged catchments located in the Montreal region in southern Quebec, Canada. In this study, two neighborhood approaches using canonical correlation analysis (CCA) and the region of influence (ROI) method are applied to delineate homogenous regions. Three regression methods namely log-linear regression model (LLRM), generalized additive models (GAM), and multivariate adaptive regression splines (MARS), recently introduced in the RFA context, are considered for regional estimation. These models are also applied considering all stations (ALL). The considered models, especially MARS, have never been used previously in a concrete application. Results indicate that MARS and GAM have comparable predictive performances, especially when applied with the whole dataset. Results also show that MARS used in combination with the CCA approach provide improved performances compared to all considered regional approaches. This may reflect the flexibility of the combination of these two approaches, their robustness, and their ability to better reproduce the hydrological phenomena, especially in real-world conditions when limited data are available.

In flood frequency analysis (FFA), annual maximum (AM) model is widely adopted in practice due to its straightforward sampling process. However, AM model has been criticized for its limited flexibility. FFA using peaks-over-threshold (POT) model is an alternative to AM model, which offers several theoretical advantages; however, this model is currently underemployed internationally. This study aims to bridge the current knowledge gap by conducting a scoping review covering several aspects of the POT approach including model assumptions, independence criteria, threshold selection, parameter estimation, probability distribution, regionalization and stationarity. We have reviewed the previously published articles on POT model to investigate: (a) possible reasons for underemployment of the POT model in FFA; and (b) challenges in applying the POT model. It is highlighted that the POT model offers a greater flexibility compared to the AM model due to the nature of sampling process associated with the POT model. The POT is more capable of providing less biased flood estimates for frequent floods. The underemployment of POT model in FFA is mainly due to the complexity in selecting a threshold (e.g., physical threshold to satisfy independence criteria and statistical threshold for Generalized Pareto distribution – the most commonly applied distribution in POT modelling). It is also found that the uncertainty due to individual variable and combined effects of the variables are not well assessed in previous research, and there is a lack of established guideline to apply POT model in FFA.

Stochastically simulated data have been employed for hydrological variables in critical water-related risk management. The simulated data can be utilized to assess the existing flood protection structure and future mitigation frameworks. Disaggregation of the simulated annual data to a lower time scale is often required since water resource management and flood mitigation plans should be done in a fine scale such as a monthly or quarter-monthly. In the current study, the randomized random block length was proposed for the nonparametric disaggregation model since one of the major weakness points for the nonparametric disaggregation model is repetition of similar patterns in the disaggregated data. Furthermore, long-term dependence structure was also mainly focused to preserve since consistent high-flow results devastating damages to inundated area. The proposed model was compared with the existing parametric and nonparametric disaggregation models. The annual net basin supplies (NBS) of the Lake Champlain–Richelieu River (LCRR) Basin was employed to test the performance of the proposed model by reproducing the critical statistics of the 2011 flood in the LCRR Basin. The 2011 flood occurred and was sustained for a few months. The results show that the existing parametric and nonparametric models have limitations and shortcoming and do not provide sufficient temporal dependence. In contrast, the proposed random block-based nonparametric disaggregation (RB-NPD) model with further model enhancement by the genetic algorithm mixture illustrates that the proposed RB-NPD model can be a comparable alternative and that its enhancement is suitable for disaggregating the annual NBS data for the LCRR Basin.

Compound extreme natural events cause a significantly larger impact than individual extreme events. Therefore, the urgency of exploring the climatology of compound events is growing. This paper is aimed to identify the current hotspots of compound heatwaves and droughts (CHD) and trends in their occurrence in Southeast Australia. In this context, 61 stations were selected from the study area, and analyses were carried out over the extended summer season of the time period 1971–2021. The hotspots of CHDs in southeast Australia were identified considering both the total count of CHD days and 90th percentile of CHDs during the study period. The study period was divided into two periods 1971–2000 and 2001–2021, to assess the change in hotspots spatially and temporally. Four different attributes of CHDs based on the number, duration, severity and amplitude of CHDs were also calculated, and MK test followed by Sen's slope was adopted to detect the trends in all these four attributes of CHDs. Trends in CHD indices were also calculated for the two periods of 1971–2000 and 2001–2021. For calculating the CHD, EHF was used to identify the heatwaves. In the case of drought identification, SPEI and SPI drought indices were adopted with aggregation of 3 and 6 antecedent months, and three different threshold values were selected to consider three levels of dry conditions, e.g., 0, -0.5 and − 1.0. It has been found that more CHDs occurred on the eastern side of the NSW and Queensland states. Furthermore, the total count of CHD days increased notably during the last two decades. The trends in CHD indices were found to be significant in the recent period (2001–2021), and there was no trend in old period (1971–2000). The findings of this study will help to plan heat and drought related emergency management in the study area.

Regional Frequency Analysis (RFA) relies on a wide range of physiographical and meteorological variables to estimate hydrological quantiles at ungauged sites. However, additional catchment characteristics related to its drainage network are not yet fully understood and integrated in RFA procedures. The aim of the present paper is to propose the integration of several physiographical variables characterizing the drainage network systems in RFA, and to evaluate their added value in predicting quantiles at ungauged sites. The proposed extended dataset (EXTD) includes several variables characterising drainage network characteristics. To evaluate the new variables, a number of commonly used RFA approaches are applied to the extended data representing 151 stations in Quebec (Canada) and compared to a standard dataset (STA) that excludes the new variables. The considered RFA approaches include the combination of two neighborhood methods namely the canonical correlation analysis (CCA) and the region of influence (ROI) with two regional estimation (RE) models which are the log-linear regression model (LLRM) and the generalized additive model (GAM). The RE models are also applied without the hydrological neighborhood. Results show that regional models using the extended dataset lead to significantly better flood quantile predictions, especially for large basins. Indeed, the variable selection performed with EXTD consistently includes some of the new variables, in particular the drainage density, the stream length ratio, and the ruggedness number. Two other new variables are also identified and included in the DHR step: the circularity ratio and the texture ratio. This leads to better predictions with relative errors about 29% for EXTD, versus around 42% for STA in the case of the best combination of RFA approaches. Thus, the proposed new variables allow for a better representation of the physical dynamics within the watersheds.

During the last two decades, a number of countries or cities established heat-health warning systems in order to alert public health authorities when some heat indicator exceeds a predetermined threshold. Different methods were considered to establish thresholds all over the world, each with its own strengths and weaknesses. The common ground is that current methods are based on exposure-response function estimates that can fail in many situations. The present paper aims at proposing several data-driven methods to establish thresholds using historical data of health issues and environmental indicators. The proposed methods are model-based regression trees (MOB), multivariate adaptive regression splines (MARS), the patient rule-induction method (PRIM) and adaptive index models (AIM). These methods focus on finding relevant splits in the association between indicators and the health outcome but do it in different fashions. A simulation study and a real-world case study hereby compare the discussed methods. Results show that proposed methods are better at predicting adverse days than current thresholds and benchmark methods. The results nonetheless suggest that PRIM is overall the more reliable method with low variability of results according to the scenario or case.

Summary In environmental epidemiology, there is wide interest in creating and using comprehensive indices that can summarize information from different environmental exposures while retaining strong predictive power on a target health outcome. In this context, the present article proposes a model called the constrained groupwise additive index model (CGAIM) to create easy-to-interpret indices predictive of a response variable, from a potentially large list of variables. The CGAIM considers groups of predictors that naturally belong together to yield meaningful indices. It also allows the addition of linear constraints on both the index weights and the form of their relationship with the response variable to represent prior assumptions or operational requirements. We propose an efficient algorithm to estimate the CGAIM, along with index selection and inference procedures. A simulation study shows that the proposed algorithm has good estimation performances, with low bias and variance and is applicable in complex situations with many correlated predictors. It also demonstrates important sensitivity and specificity in index selection, but non-negligible coverage error on constructed confidence intervals. The CGAIM is then illustrated in the construction of heat indices in a health warning system context. We believe the CGAIM could become useful in a wide variety of situations, such as warning systems establishment, and multipollutant or exposome studies.

A limited number of studies have focused on the hydroclimate dynamics of tropical Caribbean islands. The present study aims to analyze the rainfall regime in Barbados. CHIRPS gridded dataset, at a resolution of 0.05°×0.05°, providing daily rainfall data from January 1981 until 2018 was used. The variables analyzed were the annual and seasonal maximum rainfall, the total annual and seasonal rainfall, and the number of rainy days per year and per season. Potential change points in rainfall time-series were detected with a Bayesian multiple change point detection procedure. Time series were then analyzed for detection of trends using the modified Mann-Kendall test. The true temporal slopes of the rainfall time series were obtained with the Theil-Sen’s statistic. The links between rainfall and various global climate oscillation indices were also investigated. Results indicate that no change points or significant trends were observed in the annual rainfall time series. However, it was found that some climate indices have a strong correlation with precipitation on the island, especially for the total rainfall and the number of rainy days. A stationary and non-stationary frequency analysis is carried out on the rainfall annual variables using climate oscillation indices as covariates, and uncertainties on quantile estimates are identified. It is shown that non-stationary models lead to a better fit to rainfall data. Empirical mode decomposition (EMD) is used for the long-term prediction of hydro-climatic time series. Rainfall annual time series were extended with this method for a period of 20 years. Results indicate that, within that period, annual maximum rainfall will increase by about 12 mm (or 0.6 mm/year), total annual rainfall will increase by about 200 mm (or 10 mm/year) and the number of rainy days per year will see a slight decrease by about 3 days (or 0.15 day/year).

Regional frequency analysis (RFA) is widely used in the design of hydraulic structures at locations where streamflow records are not available. RFA estimates depend on the precise delineation of homogenous regions for accurate information transfer. This study proposes new physiographical variables based on river network features and tests their potential to improve the accuracy of hydrological feature estimates. Information about river network types is used both in the definition of homogenous regions and in the estimation process. Data from 105 river basins in arid and semiarid regions of the United States were used in our analysis. Artificial neural network ensemble models and canonical correlation analysis were used to produce flood quantile estimates, which were validated through tenfold cross and jackknife validations. We conducted analysis for model performance based on statistical indices, such as the Nash–Sutcliffe efficiency, root-mean-square error, relative root-mean-square error, mean absolute error, and relative mean bias. Among various combinations of variables, a model with 10 variables produced the best performance. Further, 49, 36, and 20 river networks in the 105 basins were classified as dendritic, pinnate, and trellis networks, respectively. The model with river network classification for the homogenous regions appeared to provide a superior performance compared with a model without such classification. The results indicated that including our proposed combination of variables could improve the accuracy of RFA flood estimates with the classification of the network types. This finding has considerable implications for hydraulic structure design.

Hydrological systems are naturally complex and nonlinear. A large number of variables, many of which not yet well considered in regional frequency analysis (RFA), have a significant impact on hydrological dynamics and consequently on flood quantile estimates. Despite the increasing number of statistical tools used to estimate flood quantiles at ungauged sites, little attention has been dedicated to the development of new regional estimation (RE) models accounting for both nonlinear links and interactions between hydrological and physio-meteorological variables. The aim of this paper is to simultaneously take into account nonlinearity and interactions between variables by introducing the multivariate adaptive regression splines (MARS) approach in RFA. The predictive performances of MARS are compared with those obtained by one of the most robust RE models: the generalized additive model (GAM). Both approaches are applied to two datasets covering 151 hydrometric stations in the province of Quebec (Canada): a standard dataset (STA) containing commonly used variables and an extended dataset (EXTD) combining STA with additional variables dealing with drainage network characteristics. Results indicate that RE models using MARS with the EXTD outperform slightly RE models using GAM. Thus, MARS seems to allow for a better representation of the hydrological process and an increased predictive power in RFA.

In order to understand and adequately manage hydrological stress, it is necessary to simulate groundwater levels accurately. In this research, gene expression programming (GEP) and M5 model tree (M...

The proper prediction of water sales revenue allows for pricing policies with a specified trend for the optimized use of water resources. The present work focuses on the prediction of the economic status of water sales revenue in a semi-arid environment. To meet this objective, evaporation data (E), dam input water volume (I), and dam output water volume (O) are used as independent factors to estimate water revenue (R) in the case study of Jiroft Dam, Iran. Different machine learning models are used, including classification and regression tree (CART), Chi-squared automatic interaction detector (CHAID), multi-layer perceptron neural network (MLP), and radial basis function neural network (RBF). The data are obtained daily from 20 March 2012 to 20 March 2015 and defined in six input combinations to the models using multicollinearity analyses. To compare these models, the Nash-Sutcliffe efficiency coefficient (NSEC), the root mean square error (RMSE), and the coefficient of correlation (CC) criteria are employed. All the models act better when records of water sales revenue are incorporated as additional input factors to the machine learning models. The MLP neural-based model indicates the best predicted values for daily water sales revenue (RMSE = 638.3 $ and CC = 0.798) followed by the RBF neural model (RMSE = 655.1 $ and CC = 0.786).

Although the relationship between weather and health is widely studied, there are still gaps in this knowledge. The present paper proposes data transformation as a way to address these gaps and discusses four different strategies designed to study particular aspects of a weather–health relationship, including (i) temporally aggregating the series, (ii) decomposing the different time scales of the data by empirical model decomposition, (iii) disaggregating the exposure series by considering the whole daily temperature curve as a single function, and (iv) considering the whole year of data as a single, continuous function. These four strategies allow studying non-conventional aspects of the mortality-temperature relationship by retrieving non-dominant time scale from data and allow to study the impact of the time of occurrence of particular event. A real-world case study of temperature-related cardiovascular mortality in the city of Montreal, Canada illustrates that these strategies can shed new lights on the relationship and outlines their strengths and weaknesses. A cross-validation comparison shows that the flexibility of functional regression used in strategies (iii) and (iv) allows a good fit of temperature-related mortality. These strategies can help understanding more accurately climate-related health.

BACKGROUND AND AIM: Extreme heat events have important impacts on human life and health, and have been shown to increase daily mortality and hospitalization. The heat waves of recent years led many...

Lake Water Surface Area (WSA) plays a vital role in environmental preservation and future water resource planning and management. Accurately mapping, monitoring and forecasting Lake WSA changes are of great importance to regulatory agencies. This study used the MODIS satellite images to extract a monthly time series of WSA of two lakes located in Iran from 2001 to 2019. Following a consequence of image and time series preprocessing to obtain the preprocessed lake surface area time series, the outcomes were modeled by the Long-Short-Term Memory (LSTM) deep learning (DL) method, the stochastic Seasonal Auto-Regressive Integrated Moving Average (SARIMA) method and hybridization of these two techniques with the objective of developing WSA forecasts. After separate standardization and normalization of AL TS and reevaluation of the preprocessed data, the SARIMA (1, 0, 0) (0, 1, 1)12 model outperformed sole LSTM models with correlation index of (R) 0.819, mean absolute error (MAE) of 49.425 and mean absolute percentage error (MAPE) of 0.106. On the other hand, the hybridization (stochastic-DL) enhanced the reproduction of the primal statistical properties of WSA data and caused better mediation. However, the other accuracy indices did not change markedly (R 0.819, MAE 49.310, MAPE 0.105). The multi-step preprocessing and reevaluation also caused all LSTM models to produce their best results by less than 12 inputs.

Simulation of surface water flow and temperature under a non-stationary, anthropogenically impacted climate is critical for water resource decision makers, especially in the context of environmental flow determination. Two climate change scenarios were employed to predict streamflow and temperature: RCP 8.5, the most pessimistic with regards to climate change, and RCP 4.5, a more optimistic scenario where greenhouse gas emissions peak in 2040. Two periods, 2018–2050 and 2051–2100, were also evaluated. In Canada, a number of modelling studies have shown that many regions will likely be faced with higher winter flow and lower summer flows. The CEQUEAU hydrological and water temperature model was calibrated and validated for the Wilmot River, Canada, using historic data for flow and temperature. Total annual precipitation in the region was found to remain stable under RCP 4.5 and increase over time under RCP 8.5. Median stream flow was expected to increase over present levels in the low flow months of August and September. However, increased climate variability led to higher numbers of periodic extreme low flow events and little change to the frequency of extreme high flow events. The effective increase in water temperature was four-fold greater in winter with an approximate mean difference of 4 °C, while the change was only 1 °C in summer. Overall implications for native coldwater fishes and water abstraction are not severe, except for the potential for more variability, and hence periodic extreme low flow/high temperature events.

The Middle East can experience extended wintertime spells of exceptionally hot weather, which can result in prolonged droughts and have major impacts on the already scarce water resources of the region. Recent observational studies point at increasing trends in mean and extreme temperatures in the Middle East, while climate projections seem to indicate that, in a warming weather scenario, the frequency, intensity and duration of warm spells will increase. The nonstationary warm spell frequency analysis approach proposed herein allows considering both climate variability through global climatic oscillations and climate change signals. In this study, statistical distributions with parameters conditional on covariates representing time, to account for temporal trend, and climate indices are used to predict the frequency, duration and intensity of wintertime warm spells in the Middle East. Such models could find a large applicability in various fields of climate research, and in particular in the seasonal prediction of warm spell severity. Based on previous studies linking atmospheric circulation patterns in the Atlantic to extreme temperatures in the Middle East, we use as covariates two classic modes of ‘fast’ and ‘slow’ climatic variability in the Atlantic Ocean (i.e., the Northern Atlantic Oscillation and the Atlantic Multidecadal Oscillation respectively). Results indicate that the use of covariates improves the goodness-of-fit of models for all warm spell characteristics.

The objective of the current study is to build a stochastic model to simulate climate indices that are teleconnected with the hydrologic regimes of large‐scale water resources systems such as the Great Lakes system. Climate indices generally contain nonstationary oscillations (NSOs). We adopted a stochastic simulation model based on Empirical Mode Decomposition (EMD). The procedure for the model is to decompose the observed series and then to simulate the decomposed components with the NSO resampling (NSOR) technique. Because the model has only been previously applied to single variables, a multivariate version of NSOR (M‐NSOR) is developed to consider the links between the climate indices and to reproduce the NSO process. The proposed M‐NSOR model is tested in a simulation study on the Rossler system. The simulation results indicate that the M‐NSOR model reproduces the significant oscillatory behaviors of the system and the marginal statistical characteristics. Subsequently, the M‐NSOR model is applied to three climate indices (i.e., Arctic Oscillation, El Nino‐Southern Oscillation, and Pacific Decadal Oscillation) for the annual and winter data sets. The results of the proposed model are compared to those of the Contemporaneous Shifting Mean and Contemporaneous Autoregressive Moving Average model. The results indicate that the proposed M‐NSOR model is superior to the Contemporaneous Shifting Mean and Contemporaneous Autoregressive Moving Average model for reproducing the NSO process, while the other basic statistics are comparatively well preserved in both cases. The current study concludes that the proposed M‐NSOR model can be a good alternative to simulate NSO processes and their teleconnections with climate indices.