Comparison of transfer functions in statistical downscaling models for daily temperature and precipitation over Canada

This study compares three linear models and one non-linear model, specifically multiple linear regression (MLR) with ordinary least squares (OLS) estimates, robust regression, ridge regression, and artificial neural networks (ANNs), to identify an appropriate transfer function in statistical downscaling (SD) models for the daily maximum and minimum temperatures (T_max and T_min) and daily precipitation occurrence and amounts (P_occ and P_amount). This comparison was made over twenty-five observation sites located in five different Canadian provinces (British Columbia, Saskatchewan, Manitoba, Ontario, and Québec). Reanalysis data were employed as atmospheric predictor variables of SD models. Predictors of linear transfer functions and ANN were selected by linear correlations coefficient and mutual information, respectively. For each downscaled case, annual and monthly models were developed and analysed. The monthly MLR, annual ANN, annual ANN, and annual MLR yielded the best performance for T_max, T_min, P_occ and P_amont according to the modified Akaike information criterion (AICu). A monthly MLR is recommended for the transfer functions of the four predictands because it can provide a better performance for the T_max and as good performance as the annual MLR for the T_min, P_occ, and P_amount. Furthermore, a monthly MLR can provide a slightly better performance than an annual MLR for extreme events. An annual MLR approach is also equivalently recommended for the transfer functions of the four predictands because it showed as good a performance as monthly MLR in spite of its mathematical simplicity. Robust and ridge regressions are not recommended because the data used in this study are not greatly affected by outlier data and multicollinearity problems. An annual ANN is recommended only for the T_min, based on the best performance among the models in terms of both the RMSE and AICu.     

Non-linear canonical correlation analysis in regional frequency analysis

Hydrological processes are complex non-linear phenomena. Canonical correlation analysis (CCA) is frequently used in regional frequency analysis (RFA) to delineate hydrological neighborhoods. Although non-linear CCA (NL-CCA) is widely used in several fields, it has not been used in hydrology, particularly in RFA. This paper presents an overview of techniques used to reproduce non-linear relationships between two sets of variables. The approaches considered in this work are based on NL-CCA using neural networks (CCA-NN), coupled to a log-linear regression model for flood quantile estimation. In order to demonstrate the usefulness of these approaches in RFA, a comparative study between the latter and linear CCA is performed using three different databases from North America. Results show that CCA-NN is more robust and can better reproduce the non-linear relationship structures between physiographical and hydrological variables. This reflects the high flexibility of this approach. Results indicate that for all three databases, it is more advantageous to proceed with the non-linear CCA approach.     

Generalized autoregressive conditional heteroscedasticity modelling of hydrologic time series

The existence of time‐dependent variance or conditional variance, commonly called heteroscedasticity, in hydrologic time series has not been thoroughly investigated. This paper deals with modelling the heteroscedasticity in the residuals of the seasonal autoregressive integrated moving average (SARIMA) model using a generalized autoregressive conditional heteroscedasticity (GARCH) model. The model is applied to two monthly rainfall time series from humid and arid regions. The effect of Box–Cox transformation and seasonal differencing on the remaining seasonal heteroscedasticity in the residuals of the SARIMA model is also investigated. It is shown that the seasonal heteroscedasticity in the residuals of the SARIMA model can be removed using Box–Cox transformation along with seasonal differencing for the humid region rainfall. On the other hand, transformation and seasonal differencing could not remove heteroscedasticity from the residuals of the SARIMA model fitted to rainfall data in the arid region. Therefore, the GARCH modelling approach is necessary to capture the heteroscedasticity remaining in the residuals of a SARIMA model. However, the evaluation criteria do not necessarily show that the GARCH model improves the performance of the SARIMA model. Copyright © 2012 John Wiley & Sons, Ltd.     

INTERNATIONAL HYDROLOGICAL DECADE / WORKING GROUP ON “THE INFLUENCE OF MAN ON THE HYDROLOGICAL CYCLE”

Abstract

Floods, as extreme hydrological phenomena, can be described by more than one correlated characteristic, such as peak, volume and duration. These characteristics should be jointly considered since they are generally not independent. For an ungauged site, univariate regional flood frequency analysis (FA) provides a limited assessment of flood events. A recent study proposed a procedure for regional FA in a multivariate framework. This procedure represents a multivariate version of the index-flood model and is based on copulas and multivariate quantiles. The performance of the proposed procedure was evaluated by simulation. However, the model was not tested on a real-world case study data. In the present paper, practical aspects are investigated jointly for flood peak (Q) and volume (V) of a dataset from the Côte-Nord region in the province of Quebec, Canada. The application of the proposed procedure requires the identification of the appropriate marginal distribution, the estimation of the index flood and the selection of an appropriate copula. The results of the case study show that the regional bivariate FA procedure performed well. This performance depends strongly on the performance of the two univariate models and, more specifically, the univariate model of Q. The results show also the impact of the homogeneity of the region on the performance of the univariate and bivariate models.

Editor D. Koutsoyiannis     

Comparison of direct statistical and indirect statistical-deterministic frameworks in downscaling river low-flow indices

ABSTRACT

This work explores the ability of two methodologies in downscaling hydrological indices characterizing the low flow regime of three salmon rivers in Eastern Canada: Moisie, Romaine and Ouelle. The selected indices describe four aspects of the low flow regime of these rivers: amplitude, frequency, variability and timing. The first methodology (direct downscaling) ascertains a direct link between large-scale atmospheric variables (the predictors) and low flow indices (the predictands). The second (indirect downscaling) involves downscaling precipitation and air temperature (local climate variables) that are introduced into a hydrological model to simulate flows. Synthetic flow time series are subsequently used to calculate the low flow indices. The statistical models used for downscaling low flow hydrological indices and local climate variables are: Sparse Bayesian Learning and Multiple Linear Regression. The results showed that direct downscaling using Sparse Bayesian Learning surpassed the other approaches with respect to goodness of fit and generalization ability.

Editor D. Koutsoyiannis; Associate editor K. Hamed     

Regional frequency analysis of autumnal floods in the province of Quebec, Canada

A large number of models have been proposed over the last years for regional flood frequency analysis in northern regions. However, these models dealt generally with snowmelt-caused spring floods. This paper deals with the adaptation, application, and comparison of two regional frequency analysis methods, canonical correlation analysis (CCA) and universal canonical kriging (UCK), on autumnal floods of 29 stations from the C?te-Nord region (QC, Canada). Three possible periods during which autumnal floods can take place are tested. The absolute and specific flood peak and volume quantiles are also studied. A jack-knife resampling procedure is applied to compare the performance of each model according to the selected period and the type of quantile. The period of September 1st to December 15th is found to be optimal to represent autumnal floods and specific quantiles were shown to lead to better results than absolute quantiles. Variables that explain best the autumnal floods are the basin area, the fraction of the area covered with lakes, and the average of mean July, August, and September maximal temperatures. The CCA model performs slightly better than UCK.     

Delineation the active subsurface structures at Fayoum-Cairo district, Egypt, using magnetic and magnetotelluric tools

In this work, we present a reconnaissance study to elucidate and delineate subsurface fault structures for an active tectonics area that lies between Cairo and El Fayoum provinces and consider major sources of seismicity in Egypt. Well logging, aeromagnetic, land magnetic, and magnetotelluric data have been used. The well-logging data were used for several drilled wells along W–E direction. The magnetic data were analyzed using trend analysis, 3D magnetic modeling, and Werner deconvolution techniques. The magnetotelluric data were interpreted using 2D (TM–TE) modeling techniques. The results show that there are eight major fault structures having E–W, N–S, and NW–SE directions. These faults extend downward for about 20 km at the Dahshour and Qatrani areas. The epicenter sources of the earthquakes are clustering around the intersections of these structures. The Kattaniya horst structure has been interpreted as a regional structure that exceeds the limits previously determined by geologists. The depth to this horst reaches about 1.8 km at the NW and more than 4.3 km at the southern parts. The interpreted values of magnetic susceptibility at the horst zone indicate that they are ultrabasic/basic intrusion bodies.     

Groundwater abstraction management in Sana’a Basin,Yemen: a local community approach

Overexploitation of groundwater resources in Sana’a Basin, Yemen, is causing severe water shortages associated water quality degradation. Groundwater abstraction is five times higher than natural recharge and the water-level decline is about 4–8 m/year. About 90 % of the groundwater resource is used for agricultural activities. The situation is further aggravated by the absence of a proper water-management approach for the Basin. Water scarcity in the Wadi As-Ssirr catchment, the study area, is the most severe and this area has the highest well density (average 6.8 wells/km²) compared with other wadi catchments. A local scheme of groundwater abstraction redistribution is proposed, involving the retirement of a substantial number of wells. The scheme encourages participation of the local community via collective actions to reduce the groundwater overexploitation, and ultimately leads to a locally acceptable, manageable groundwater abstraction pattern. The proposed method suggests using 587 wells rather than 1,359, thus reducing the well density to 2.9 wells/km². Three scenarios are suggested, involving different reductions to the well yields and/or the number of pumping hours for both dry and wet seasons. The third scenario is selected as a first trial for the communities to action; the resulting predicted reduction, by 2,371,999 m³, is about 6 % of the estimated annual demand. Initially, the groundwater abstraction volume should not be changed significantly until there are protective measures in place, such as improved irrigation efficiency, with the aim of increasing the income of farmers and reducing water use.     

Predicting river water temperatures using stochastic models: case study of the Moisie River (Québec,Canada)

Successful applications of stochastic models for simulating and predicting daily stream temperature have been reported in the literature. These stochastic models have been generally tested on small rivers and have used only air temperature as an exogenous variable. This study investigates the stochastic modelling of daily mean stream water temperatures on the Moisie River, a relatively large unregulated river located in Québec, Canada. The objective of the study is to compare different stochastic approaches previously used on small streams to relate mean daily water temperatures to air temperatures and streamflow indices. Various stochastic approaches are used to model the water temperature residuals, representing short‐term variations, which were obtained by subtracting the seasonal components from water temperature time‐series. The first three models, a multiple regression, a second‐order autoregressive model, and a Box and Jenkins model, used only lagged air temperature residuals as exogenous variables. The root‐mean‐square error (RMSE) for these models varied between 0·53 and 1·70 °C and the second‐order autoregressive model provided the best results. A statistical methodology using best subsets regression is proposed to model the combined effect of discharge and air temperature on stream temperatures. Various streamflow indices were considered as additional independent variables, and models with different number of variables were tested. The results indicated that the best model included relative change in flow as the most important streamflow index. The RMSE for this model was of the order of 0·51 °C, which shows a small improvement over the first three models that did not include streamflow indices. The ridge regression was applied to this model to alleviate the potential statistical inadequacies associated with multicollinearity. The amplitude and sign of the ridge regression coefficients seem to be more in agreement with prior expectations (e.g. positive correlation between water temperature residuals of different lags) and make more physical sense. Copyright © 2006 John Wiley & Sons, Ltd.     

Daily river water temperature forecast model with a k‐nearest neighbour approach

Water temperature is a key abiotic variable that modulates both water chemistry and aquatic life in rivers and streams. For this reason, numerous water temperature models have been developed in recent years. In this paper, a k‐nearest neighbour model (KNN) is proposed and validated to simulate and eventually produce a one‐day forecast of mean water temperature on the Moisie River, a watercourse with an important salmon population in eastern Canada. Numerous KNN model configurations were compared by selecting different attributes and testing different weight combinations for neighbours. It was found that the best model uses attributes that include water temperature from the two previous days and an indicator of seasonality (day of the year) to select nearest neighbours. Three neighbours were used to calculate the estimated temperature, and the weighting combination that yielded the best results was an equal weight on all three nearest neighbours. This nonparametric model provided lower Root Mean Square Errors (RMSE = 1·57 °C), Higher Nash coefficient (NTD = 0·93) and lower Relative Bias (RB = ? 1·5%) than a nonlinear regression model (RMSE = 2·45 °C, NTD = 0·83, RB = ? 3%). The k‐nearest neighbour model appears to be a promising tool to simulate of forecast water temperature where long time series are available. Copyright © 2011 John Wiley & Sons, Ltd.