Skill of 6HR and 12HR thunderstorm forecasts based on assessment of larger-scale information

Summary The relationship between information, contained in aerological data from the European area, and a thunderstorm occurrence in the area of the Czech Republic was investigated with input data from the period of May–September 1989–1991. SYNOP reports from Czech ground stations were utilized to assess event occurrence. TEMP 00UTC and TEMP 12UTC reports from European stations were used to determine potential diagnostic predictors, and the TEMP00 data served as the input data set for the 12hr mesoscale model forecast to gain prognostic predictors. Each of the two diagnostic data sets from 00UTC and 12UTC and of the prognostic data set comprised about 400 predictors/predictand elements. The categorical forecast of thunderstorm occurrence, based on the application of linear regression and a simple version of pattern recognition, is discussed. The critical success index was determined for every type of forecast and used to assess forecast skill.     

Statistical Postprocessing of NWP Model Outputs Applied to Maximum and Minimum Temperature Forecasts

Statistical postprocessing of NWP model outputs is applied to maximum and minimum temperature forecasts. Two approaches to its application are effected to local short-range weather forecasts of minimum and maximum temperatures: Model Output Statistics and modified Perfect Prognosis. The modified Perfect Prognosis method is restricted to the first step of PP because of the significant difference between the horizontal resolution of the available objective analyses and the NWP model outputs. The modified Perfect Prognosis method uses actual data from the objective analysis related to the forecast period instead of the NWP forecast. The results are compared with a simple statistical prognostic model, which does not utilize the NWP model outputs, and with simple reference methods. The forecast is verified using ground station measurements from stations providing SYNOP reports. The results show that the predictive accuracy of the Model Output Statistics method is not very different from that of the modified Perfect Prognosis method, and both are significantly more accurate than the direct predictions of the NWP model. The results have confirmed that statistical postprocessing is able to make localized predictions even if lowresolution data are used.     

River flow forecasting using wavelet and cross‐wavelet transform models

In this study, short‐term river flood forecasting models based on wavelet and cross‐wavelet constituent components were developed and evaluated for forecasting daily stream flows with lead times equal to 1, 3, and 7 days. These wavelet and cross‐wavelet models were compared with artificial neural network models and simple perseverance models. This was done using data from the Skrwa Prawa River watershed in Poland. Numerical analysis was performed on daily maximum stream flow data from the Parzen station and on meteorological data from the Plock weather station in Poland. Data from 1951 to 1979 was used to train the models while data from 1980 to 1983 was used to test the models. The study showed that forecasting models based on wavelet and cross‐wavelet constituent components can be used with great accuracy as a stand‐alone forecasting method for 1 and 3 days lead time river flood forecasting, assuming that there are no significant trends in the amplitude for the same Julian day year‐to‐year, and that there is a relatively stable phase shift between the flow and meteorological time series. It was also shown that forecasting models based on wavelet and cross‐wavelet constituent components for forecasting river floods are not accurate for longer lead time forecasting such as 7 days, with the artificial neural network models providing more accurate results. Copyright © 2008 John Wiley & Sons, Ltd.     

The binary decision tree: The growing algorithm and application to thunderstorm forecasting

Summary The paper deals with the probabilistic prediction of event occurrence with use of the binary decision tree which is grown from the learning sample. The tree growing algorithm consists in recursive partition of the predictor space by either single-predictor-based (SP) splits or by hyperplanes perpendicular to the best linear discriminant function (BLDF), and is intended to maximally effectively discriminate the elements of the learning sample with event occurrence from the elements without event occurrence.The predictand is the thunderstorm occurrence in the afternoon in Prague, the set of predictors includes variables derived from a midday single-station TEMP-A data (Perfect Prog approach), persistence predictors and predictors related to passages of the fronts across Prague. The experiments are designed to test the performance of the tree growing algorithm — with a stress upon indeterminateness following from the limited size of the learning sample — and to evaluate the predictive potential of the predictors for thunderstorm forecasting. The stability of the tree structure, the optimal size of the tree and the related prognostic skill score increase with increasing size of the learning sample. Employment of the BLDF splits allows quicker and more effective partition of the predictor space on the assumption that the predictor vector has lower dimension and is well behaved (preferably normally distributed). The stability indices of Faust, Showalter and Adedokun were found to be the most effective predictors. Persistence and frontal predictors only slightly contribute to the total prediction skill of the decision tree. The optimally sized tree has only five splitting nodes and employs three thermodynamical predictors, one frontal and one persistence predictor.     

Improvement of Local Categorical Precipitation Forecasts from an NWP Model by Various Statistical Postprocessing Methods

Several statistical postprocessing methods are applied to results from a numerical weather prediction (NWP) model to test the potential for increasing the accuracy of its local precipitation forecasts. Categorical (Yes/No) forecasts for 12hr precipitation sums equalling or exceeding 0.1, 2.0 and 5.0 mm are selected for improvement. The two 12hr periods 0600-1800 UTC and 1800-0600 UTC are treated separately based on NWP model initial times 0000 UTC and 1200 UTC, respectively. Input data are taken from three successive summer seasons, April-September, 1994-96. The forecasts are prepared and verified for five synoptic stations, four located in the western Czech Republic, and one in Germany near the Czech-German border. Two approaches to statistical postprocessing are tested. The first uses Model Output Statistics (MOS) and the second modifies the MOS approach by applying a successive learning technique (SLT). For each approach several statistical models for the relationship between NWP model predictors and predictand were studied. An independent data set is used for forecast verification with the skill measured by a True Skill Score. The results of the statistical postprocessing are compared with the direct model precipitation forecasts from gridpoints nearest the stations, and they show that both postprocessing approaches provide substantially better forecasts than the direct NWP model output. The relative improvement increases with increasing precipitation amount and there is no significant difference in performance between the two 12hr periods. The skill of the SLT does not depend significantly on the size of the initial learning sample, but its results are nevertheless comparable with the results obtained from the MOS approach, which requires larger developmental samples.     

Improvement of mid- to long-term runoff forecasting based on physical causes: application in Nenjiang basin,China

Abstract

An artificial neural network, mid- to long-term runoff forecasting model of the Nenjiang basin was established by deciding predictors using the physical analysis method, combined with long-term hydrological and meteorological information. The forecasting model was gradually improved while considering physical factors, such as the main flood season and non-flood season by stage, runoff sources and hydrological processes. The average relative errors in the simulation tests of the prediction model were 0.33 in the main flood season and 0.26 in the non-flood season, indicating that the prediction accuracy during the non-flood season was greater than that in the main flood season. Based on these standards, forecasting accuracy evaluation was conducted by comparing forecasting results with actual conditions: for 2001 to 2003 data, the pass rate of forecasting in the main flood season was 50%, while it was 93% in the non-flood season; for 2001–2010, the respective values were 45% and 72%. The accuracy of prediction was found to decrease as the length of record increases.

Editor D. Koutsoyiannis, Associate editor A. Viglione

Citation Li, H.-Y. Tian, L., Wu, Y., and Xie, M., 2013. Improvement of mid- to long-term runoff forecasting based on physical causes: application in Nenjiang basin, China. Hydrological Sciences Journal, 58 (7), 1414–1422.     

Deep echo state network: a novel machine learning approach to model dew point temperature using meteorological variables

ABSTRACT

The potential of different models – deep echo state network (DeepESN), extreme learning machine (ELM), extra tree (ET), and regression tree (RT) – in estimating dew point temperature by using meteorological variables is investigated. The variables consist of daily records of average air temperature, atmospheric pressure, relative humidity, wind speed, solar radiation, and dew point temperature (T_dew) from Seoul and Incheon stations, Republic of Korea. Evaluation of the model performance shows that the models with five and three-input variables yielded better accuracy than the other models in these two stations, respectively. In terms of root-mean-square error, there was significant increase in accuracy when using the DeepESN model compared to the ELM (18%), ET (58%), and RT (64%) models at Seoul station and the ELM (12%), ET (23%), and RT (49%) models at Incheon. The results show that the proposed DeepESN model performed better than the other models in forecasting T_dew values.     

Suitability of North American Regional Reanalysis (NARR) output for hydrologic modelling and analysis in mountainous terrain

Meteorological observations at high elevations in mountainous regions are often lacking. One opportunity to fill this data gap is through the use of downscaled output from weather reanalysis models. In this study, we tested the accuracy of downscaled output from the North American Regional Reanalysis (NARR) against high‐elevation surface observations at four ridgetop locations in the southern Coast Mountains of British Columbia, Canada. NARR model output was downscaled to the surface observation locations through three‐dimensional interpolation for air temperature, vapour pressure and wind speed and two‐dimensional interpolation for radiation variables. Accuracy was tested at both the 3‐hourly and daily time scales. Air temperature displayed a high level of agreement, especially at the daily scale, with root mean square error (RMSE) values ranging from 0.98 to 1.21 °C across all sites. Vapour pressure downscaling accuracy was also quite high (RMSE of 0.06 to 0.11 hPa) but displayed some site specific bias. Although NARR overestimated wind speed, there were moderate to strong linear relations (r² from 0.38 to 0.84 for daily means), suggesting that the NARR output could be used as an index and bias‐corrected. NARR output reproduced the seasonal cycle for incoming short‐wave radiation, with Nash–Sutcliffe model efficiencies ranging from 0.78 to 0.87, but accuracy suffered on days with cloud cover, resulting in a positive bias and RMSE ranged from 42 to 46 Wm^{? 2}. Although fewer data were available, incoming long‐wave radiation from NARR had an RMSE of 19 Wm^{? 2} and outperformed common methods for estimating incoming long‐wave radiation. NARR air temperature showed potential to assist in hydrologic analysis and modelling during an atmospheric river storm event, which are characterized by warm and wet air masses with atypical vertical temperature gradients. The incorporation of a synthetic NARR air temperature station to better represent the higher freezing levels resulted in increased predicted peak flows, which better match the observed run‐off during the event. Copyright © 2016 John Wiley & Sons, Ltd.     

Efficient estimation of flood forecast prediction intervals via single‐ and multi‐objective versions of the LUBE method

Prediction intervals (PIs) are commonly used to quantify the accuracy and precision of a forecast. However, traditional ways to construct PIs typically require strong assumptions about data distribution and involve a large computational burden. Here, we improve upon the recent proposed Lower Upper Bound Estimation method and extend it to a multi‐objective framework. The proposed methods are demonstrated using a real‐world flood forecasting case study for the upper Yangtze River Watershed. Results indicate that the proposed methods are able to efficiently construct appropriate PIs, while outperforming other methods including the widely used Generalized Likelihood Uncertainty Estimation approach. Copyright © 2016 John Wiley & Sons, Ltd.     

Ensemble rainfall forecasting with numerical weather prediction and radar‐based nowcasting models

The overall objective of this study is to improve the forecasting accuracy of the precipitation in the Singapore region by means of both rainfall forecasting and nowcasting. Numerical Weather Predication (NWP) and radar‐based rainfall nowcasting are two important sources for quantitative precipitation forecast. In this paper, an attempt to combine rainfall prediction from a high‐resolution mesoscale weather model and a radar‐based rainfall model was performed. Two rainfall forecasting methods were selected and examined: (i) the weather research and forecasting model (WRF); and (ii) a translation model (TM). The WRF model, at a high spatial resolution, was run over the domain of interest using the Global Forecast System data as initializing fields. Some heavy rainfall events were selected from data record and used to test the forecast capability of WRF and TM. Results obtained from TM and WRF were then combined together to form an ensemble rainfall forecasting model, by assigning weights of 0.7 and 0.3 weights to TM and WRF, respectively. This paper presented results from WRF and TM, and the resulting ensemble rainfall forecasting; comparisons with station data were conducted as well. It was shown that results from WRF are very useful as advisory of anticipated heavy rainfall events, whereas those from TM, which used information of rain cells already appearing on the radar screen, were more accurate for rainfall nowcasting as expected. The ensemble rainfall forecasting compares reasonably well with the station observation data. Copyright © 2012 John Wiley & Sons, Ltd.     

Fuzzy computing based rainfall–runoff model for real time flood forecasting

This paper analyses the skills of fuzzy computing based rainfall–runoff model in real time flood forecasting. The potential of fuzzy computing has been demonstrated by developing a model for forecasting the river flow of Narmada basin in India. This work has demonstrated that fuzzy models can take advantage of their capability to simulate the unknown relationships between a set of relevant hydrological data such as rainfall and river flow. Many combinations of input variables were presented to the model with varying structures as a sensitivity study to verify the conclusions about the coherence between precipitation, upstream runoff and total watershed runoff. The most appropriate set of input variables was determined, and the study suggests that the river flow of Narmada behaves more like an autoregressive process. As the precipitation is weighted only a little by the model, the last time‐steps of measured runoff are dominating the forecast. Thus a forecast based on expected rainfall becomes very inaccurate. Although good results for one‐step‐ahead forecasts are received, the accuracy deteriorates as the lead time increases. Using the one‐step‐ahead forecast model recursively to predict flows at higher lead time, however, produces better results as opposed to different independent fuzzy models to forecast flows at various lead times. Copyright © 2004 John Wiley & Sons, Ltd.     

Probabilistic forecasting of hydrological events using geostatistical analysis / Prévision probabiliste d'événements hydrologiques par analyse géostatistique

Abstract

A method is introduced for probabilistic forecasting of hydrological events based on geostatistical analysis. In this method, the predictors of a hydrological variable define a virtual field such that, in this field, the observed dependent variables are considered as measurement points. Variography of the measurement points enables the use of the system of kriging equations to estimate the value of the variable at non-measured locations of the field. Non-measured points are the forecasts associated with specific predictors. Calculation of the estimation variance facilitates probabilistic analysis of the forecast variables. The method is applied to case studies of the Red River in Manitoba, Canada and Karoon River in Khoozestan, Iran. The study analyses the advantages and limitations of the proposed method in comparison with a K-nearest neighbour approach and linear and nonlinear multiple regression. The utility of the proposed method for forecasting hydrological variables with a conditional probability distribution is demonstrated.     

A high-resolution real-time forecasting system for predicting the fate of oil spills in the Strait of Bonifacio (western Mediterranean Sea)

The Strait of Bonifacio is a long and narrow area between Corsica and Sardinia. To manage environmental emergencies related to the spill of oil from vessels, an innovative forecasting system was developed. This tool is capable of operationally predicting the dispersion of hydrocarbon spills in the coastal area of the Bonifacio Strait, either from an instantaneous or continuous spill and either in forward or backward mode. Experimental datasets, including ADCP water current measurements and the trajectories of drifter buoys released in the area, were used to evaluate the accuracy of this system. A comparison between the simulation results and experimental data revealed that both the water circulation and the surface transport processes are accurately reproduced by the model. The overall accuracy of the system in reproducing the transport of an oil spill at sea was estimated for both forward and backward prediction mode and in relation to different forecasting time lags.     

A comparative study of extensive machine learning models for predicting long-term monthly rainfall with an ensemble of climatic and meteorological predictors

Rainfall prediction is of vital importance in water resources management. Accurate long-term rainfall prediction remains an open and challenging problem. Machine learning techniques, as an increasingly popular approach, provide an attractive alternative to traditional methods. The main objective of this study was to improve the prediction accuracy of machine learning-based methods for monthly rainfall, and to improve the understanding of the role of large-scale climatic variables and local meteorological variables in rainfall prediction. One regression model autoregressive integrated moving average model (ARIMA) and five state-of-the-art machine learning algorithms, including artificial neural networks, support vector machine, random forest (RF), gradient boosting regression, and dual-stage attention-based recurrent neural network, were implemented for monthly rainfall prediction over 25 stations in the East China region. The results showed that the ML models outperformed ARIMA model, and RF relatively outperformed other models. Local meteorological variables, humidity, and sunshine duration, were the most important predictors in improving prediction accuracy. 4-month lagged Western North Pacific Monsoon had higher importance than other large-scale climatic variables. The overall output of rainfall prediction was scalable and could be readily generalized to other regions.     

Application of integrated back‐propagation network and self organizing map for flood forecasting

Much of the nonlinearity and uncertainty regarding the flood process is because hydrologic data required for estimation are often tremendously difficult to obtain. This study employed a back‐propagation network (BPN) as the main structure in flood forecasting to learn and to demonstrate the sophisticated nonlinear mapping relationship. However, a deterministic BPN model implies high uncertainty and poor consistency for verification work even when the learning performance is satisfactory for flood forecasting. Therefore, a novel procedure was proposed in this investigation which integrates linear transfer function (LTF) and self‐organizing map (SOM) to efficiently determine the intervals of weights and biases of a flood forecasting neural network to avoid the above problems. A SOM network with classification ability was applied to the solutions and parameters of the BPN model in the learning stage, to classify the network parameter rules and to obtain the winning parameters. The outcomes from the previous stage were then used as the ranges of the parameters in the recall stage. Finally, a case study was carried out in Wu‐Shi basin to demonstrate the effectiveness of the proposal. Copyright © 2009 John Wiley & Sons, Ltd.     

Hydroclimate variability and long-lead forecasting of rainfall over Thailand by large-scale atmospheric variables

Abstract

The development of statistical relationships between local hydroclimates and large-scale atmospheric variables enhances the understanding of hydroclimate variability. The rainfall in the study basin (the Upper Chao Phraya River Basin, Thailand) is influenced by the Indian Ocean and tropical Pacific Ocean atmospheric circulation. Using correlation analysis and cross-validated multiple regression, the large-scale atmospheric variables, such as temperature, pressure and wind, over given regions are identified. The forecasting models using atmospheric predictors show the capability of long-lead forecasting. The modified k-nearest neighbour (k-nn) model, which is developed using the identified predictors to forecast rainfall, and evaluated by likelihood function, shows a long-lead forecast of monsoon rainfall at 7–9 months. The decreasing performance in forecasting dry-season rainfall is found for both short and long lead times. The developed model also presents better performance in forecasting pre-monsoon season rainfall in dry years compared to wet years, and vice versa for monsoon season rainfall.

Editor Z.W. Kundzewicz

Citation Singhrattna, N., Babel, M.S. and Perret, S.R., 2012. Hydroclimate variability and long-lead forecasting of rainfall over Thailand by large-scale atmospheric variables. Hydrological Sciences Journal, 57 (1), 26–41.     

A hybrid approach of artificial neural network and multiple regression to forecast typhoon rainfall and groundwater-level change

ABSTRACT

A forecasting model is developed using a hybrid approach of artificial neural network (ANN) and multiple regression analysis (MRA) to predict the total typhoon rainfall and groundwater-level change in the Zhuoshui River basin. We used information from the raingauge stations in eastern Taiwan and open source typhoon data to build the ANN model for forecasting the total rainfall and the groundwater level during a typhoon event; then we revised the predictive values using MRA. As a result, the average accuracy improved up to 80% when the hybrid model of ANN and MRA was applied, even where insufficient data were available for model training. The outcome of this research can be applied to forecasts of total rainfall and groundwater-level change before a typhoon event reaches the Zhuoshui River basin once the typhoon has made landfall on the east coast of Taiwan.     

Adjusting wavelet‐based multiresolution analysis boundary conditions for long‐term streamflow forecasting

We propose a novel technique for improving a long‐term multi‐step‐ahead streamflow forecast. A model based on wavelet decomposition and a multivariate Bayesian machine learning approach is developed for forecasting the streamflow 3, 6, 9, and 12 months ahead simultaneously. The inputs of the model utilize only the past monthly streamflow records. They are decomposed into components formulated in terms of wavelet multiresolution analysis. It is shown that the model accuracy can be increased by using the wavelet boundary rule introduced in this study. A simulation study is performed to evaluate the effects of different wavelet boundary rules using synthetic and real streamflow data from the Yellowstone River in the Uinta Basin in Utah. The model based on the combination of wavelet and Bayesian machine learning regression techniques is compared with that of the wavelet and artificial neural networks‐based model. The robustness of the models is evaluated. Copyright © 2015 John Wiley & Sons, Ltd.     

Real‐time recurrent learning neural network for stream‐flow forecasting

Various types of neural networks have been proposed in previous papers for applications in hydrological events. However, most of these applied neural networks are classified as static neural networks, which are based on batch processes that update action only after the whole training data set has been presented. The time variate characteristics in hydrological processes have not been modelled well. In this paper, we present an alternative approach using an artificial neural network, termed real‐time recurrent learning (RTRL) for stream‐flow forecasting. To define the properties of the RTRL algorithm, we first compare the predictive ability of RTRL with least‐square estimated autoregressive integrated moving average models on several synthetic time‐series. Our results demonstrate that the RTRL network has a learning capacity with high efficiency and is an adequate model for time‐series prediction. We also investigated the RTRL network by using the rainfall–runoff data of the Da‐Chia River in Taiwan. The results show that RTRL can be applied with high accuracy to the study of real‐time stream‐flow forecasting networks. Copyright © 2002 John Wiley & Sons, Ltd.