Term Forecasting Research Articles

Combining vLAPS and Nudging Data Assimilation

The combination of techniques that incorporate observational data may improve numerical weather prediction forecasts; thus, in this study, the methodology and potential value of one such combination were investigated. A series of experiments on a single case day was used to explore a 3DVAR-based technique (the variational version of the Local Analysis and Prediction System; vLAPS) in combination with Newtonian relaxation (observation and analysis nudging) for simulating moist convection in the Advanced Research version of the Weather Research and Forecasting model. Experiments were carried out with various combinations of vLAPS and nudging for a series of forecast start times. A limited subjective analysis of reflectivity suggested all experiments generally performed similarly in reproducing the overall convective structures. Objective verification indicated that applying vLAPS analyses without nudging performs best during the 0–2 h forecast in terms of placement of moist convection but worst in the 3–5 h forecast and quickly develops the most substantial overforecast bias. The analyses used for analysis nudging were at much finer temporal and spatial scales than usually used in pre-forecast analysis nudging, and the results suggest that further research is needed on how to best apply analysis nudging of analyses at these scales.

Bayesian Networks modeling of diarrhetic shellfish poisoning in Mytilus edulis harvested in Bantry Bay, Ireland

Diarrhetic Shellfish Poisoning (DSP) results from the human consumption of contaminated shellfish with marine biotoxins, which are produced by some species of marine dinoflagellates, mainly belonging to the genus Dinophysis. Shellfish contamination with marine biotoxins not only pose a threat to human health, but also lead to financial loss to aquaculture operations from the temporary closure of production areas when toxin concentrations exceed regulatory levels. In this study, we developed a Bayesian Network (BN) model for forecasting the short-term variations of DSP toxins in blue mussels (Mytilus edulis) from Bantry Bay, Southwest Ireland. Data inputs to a BN model from 10 production sites in Bantry Bay included plankton cell densities in sea water, DSP toxin concentration in mussels and sea surface temperature. The model was trained with data from 2014 to 2018, and validated with data of 2019. Validation consisted of predicting the DSP toxin concentration at one production site using the model parameters from the other locations as input values. Model validation showed that the prediction accuracy was higher than 86%. Sensitivity analysis indicated that in general, DSP toxin concentration was more relevant than plankton abundance. This initial work has demonstrated the usefulness of BN modeling as an approach to short term forecasting. Further work is ongoing to use the model for scenario testing and to increase the number of environmental parameters used as inputs to the model.

Long term and short term forecasting of horticultural produce based on the LSTM network model

Long term and short term forecasting of horticultural produce based on the LSTM network model

Improved Trend-Aware Postprocessing of GCM Seasonal Precipitation Forecasts

Abstract Climate trends have been observed over the recent decades in many parts of the world, but current global climate models (GCMs) for seasonal climate forecasting often fail to capture these trends. As a result, model forecasts may be biased above or below the trendline. In our previous research, we developed a trend-aware forecast postprocessing method to overcome this problem. The method was demonstrated to be effective for embedding observed trends into seasonal temperature forecasts. In this study, we further develop the method for postprocessing GCM seasonal precipitation forecasts. We introduce new formulation and evaluation features to cater for special characteristics of precipitation amounts, such as having a zero lower bound and highly positive skewness. We apply the improved method to calibrate ECMWF SEAS5 forecasts of seasonal precipitation for Australia. Our evaluation shows that the calibrated forecasts reproduce observed trends over the hindcast period of 36 years. In some regions where observed trends are statistically significant, forecast skill is greatly improved by embedding trends into the forecasts. In most regions, the calibrated forecasts outperform the raw forecasts in terms of bias, skill, and reliability. Wider applications of the new trend-aware postprocessing method are expected to boost user confidence in seasonal precipitation forecasts.

Predicting daily water tank level fluctuations by using ARIMA model. A case study

The intrinsic dynamical features of water demand highlight the need of proper operational management of tanks in water distribution networks. In addition, due to the water resource scarcity, sustainable management of urban systems is essential. For this purpose, the aid of a predictive model is crucial since it allows to give short term forecasts that can be used to predict the oscillations of relevant parameters, i.e. tanks level and/or water demand. Urban water managers can use these predictions to implement actions aimed at the optimisation of the network function. Among several modelling techniques, the univariate time series analysis is instrumental since it allows forecasting the studied parameter by using the measurements of the parameter itself. In this paper, an autoregressive integrated moving average (ARIMA) model is calibrated on water levels data, measured in an urban tank in Benevento, Campania region (Italy) and then tested on a large dataset not used to tune the parameters. The validation and forecast phases show good performances of the model on a short-term forecast horizon demonstrating the excellent potentiality of this techniques. Finally, the residuals and errors analysis complete the work suggesting possible future implementations and improvements of this technique.

The probabilistic solar particle event forecasting (PROSPER) model

The Probabilistic Solar Particle Event foRecasting (PROSPER) model predicts the probability of occurrence and the expected peak flux of solar energetic particle (SEP) events. Predictions are derived for a set of integral proton energies (i.e., E > 10, > 30, and > 100 MeV) from characteristics of solar flares (longitude, magnitude), coronal mass ejections (width, speed), and combinations of both. Herein the PROSPER model methodology for deriving the SEP event forecasts is described, and the validation of the model, based on archived data, is presented for a set of case studies. The PROSPER model has been incorporated into the new operational advanced solar particle event casting system (ASPECS) tool to provide nowcasting (short term forecasting) of SEP events as part of ESA’s future SEP advanced warning system (SAWS). ASPECS also provides the capability to interrogate PROSPER for historical cases via a run-on-demand functionality.

Cognitive consequences of COVID-19

An analysis of foreign publications illustrates changes in ideas about the etiology, pathogenesis, target organs, leading pathophysiological syndromes and clinical manifestations of nervous system damage in COVID-19 over 2.5 years of the pandemic. It has been demonstrated that, according to modern concepts, COVID-19 is considered as a pathology with a tendency to multiple organ damage, where the symptoms of damage to the nervous system not only determine the severity of the course of the disease, but also indirectly affect the development of severe somatic manifestations, such as distress syndrome. Particular attention is paid to the analysis of publications summarizing the available data on the frequency of cognitive deficit in the acute period of COVID-19 and in the post-COVID period, risk factors, recovery prospects. Putative mechanisms underlying cognitive deficit in COVID-19 are considered. Despite a significant number of observations and publications, many mechanisms of nervous system damage in COVID-19 and the selectivity of the cognitive deficit pattern remain unexplored. In modern publications, there is no information and long term forecast of the cognitive state dynamics, which is an important argument in favor of the need to continue research.

Assessing Forecasting Performance of Daily Mean Temperature at 1st and 2nd Perak Station, Surabaya Using ARIMA and VARIMA Model with Outlier Detection

Air temperature is an important data for several sectors. The demand of fast, exact and accurate forecast on temperature data is getting extremely important since it is useful for planning of several important sectors. In order to forecast mean daily temperature data at 1st and 2nd Perak BMKG Station in Surabaya, this study used the univariate method, ARIMA model and multivariate method, VARIMA model with outlier detection. The best ARIMA model was selected using in-sample criteria, i.e. AIC and BIC. While for VAR model, the minimum information criterion namely AICc value was considered. The RMSE values of several forecasting horizons of out-sample data showed that the overall best model for mean daily temperature at 1st and 2nd Perak Station was the multivariate model, i.e. VARX (10,1) with four outliers incorporated in the model, indicated that it was necessary to consider the temperature from the nearest stations to improve the forecasting performance. This study recommends performing the overall best model only for short term forecasting, i.e. two weeks at maximum. By using the one week-step ahead and one day-step ahead forecasting scheme, the forecasting performance is significantly improved compared to default the k-step ahead forecasting scheme.

Analyzing the Applicability of Random Forest-Based Models for the Forecast of Run-of-River Hydropower Generation

Analyzing the impact of climate variables into the operational planning processes is essential for the robust implementation of a sustainable power system. This paper deals with the modeling of the run-of-river hydropower production based on climate variables on the European scale. A better understanding of future run-of-river generation patterns has important implications for power systems with increasing shares of solar and wind power. Run-of-river plants are less intermittent than solar or wind but also less dispatchable than dams with storage capacity. However, translating time series of climate data (precipitation and air temperature) into time series of run-of-river-based hydropower generation is not an easy task as it is necessary to capture the complex relationship between the availability of water and the generation of electricity. This task is also more complex when performed for a large interconnected area. In this work, a model is built for several European countries by using machine learning techniques. In particular, we compare the accuracy of models based on the Random Forest algorithm and show that a more accurate model is obtained when a finer spatial resolution of climate data is introduced. We then discuss the practical applicability of a machine learning model for the medium term forecasts and show that some very context specific but influential events are hard to capture.

Improving Atmospheric Angular Momentum Forecasts by Machine Learning

AbstractEarth angular momentum forecasts are naturally accompanied by forecast errors that typically grow with increasing forecast length. In contrast to this behavior, we have detected large quasi‐periodic deviations between atmospheric angular momentum wind term forecasts and their subsequently available analysis. The respective errors are not random and have some hard to define yet clearly visible characteristics which may help to separate them from the true forecast information. These kinds of problems, which should be automated but involve some adaptation and decision‐making in the process, are most suitable for machine learning methods. Consequently, we propose and apply a neural network to the task of removing the detected artificial forecast errors. We found that a cascading forward neural network model performed best in this problem. A total error reduction with respect to the unaltered forecasts amounts to about 30% integrated over a 6‐days forecast period. Integrated over the initial 3‐days forecast period, in which the largest artificial errors are present, the improvements amount to about 50%. After the application of the neural network, the remaining error distribution shows the expected growth with forecast length. However, a 24‐hourly modulation and an initial baseline error of 2 × 10−8 became evident that were hidden before under the larger forecast error.

Quality of the Wind Wave Forecast in the Black Sea Including Storm Wave Analysis

This paper presents the results of wind wave forecasts for the Black Sea. Three different versions utilized were utilized: the WAVEWATCH III model with GFS 0.25 forcing on a regular grid, the WAVEWATCH III model with COSMO-RU07 forcing on a regular grid, and the SWAN model with COSMO-RU07 forcing on an unstructured grid. AltiKa satellite altimeter data were used to assess the quality of wind and wave forecasts for the period from 1 April to 31 December 2017. Wave height and wind speed forecast data were obtained with a lead time of up to 72 h. The presented models provide an adequate forecast in terms of modern wave modeling (a correlation coefficient of 0.8–0.9 and an RMSE of 0.25–0.3 m) when all statistics were analyzed. A clear improvement in the wave forecast quality with the high-resolution wind forecast COSMO-RU07 was not registered. The bias error did not exceed 0.5 m in an SWH range from 0 to 3 m. However, the bias sharply increased to −2 or −3 m for an SWH range of 3–4 m. Wave forecast quality assessments were conducted for several storm cases.

Monitoring the newly infected cases of COVID-19 data weekly: A Survival Data Analysis (SDA) perspective

This paper attempts to fit the best survival model distribution for the Malaysian COVID-19 new infections experience of Wave I/II and Wave III using the well-known Survival Data Analysis (SDA) procedures. The purpose of fitting such models is to reduce the complexity and frequency of the COVID-19 new infections data into a single measure of scale and shape parameters to enable monitoring of weekly trends, undertake short term forecasts and estimate duration when the virality will be contained. The analysis showed a Weibull distribution is the best statistical fit for Malaysia’s new infections COVID-19 data. The estimates of scale and shape parameters for Wave I/II was 0.05901 and 2.48956 and for Wave III was 0.06463 and 2.5693, respectively. Much higher hazard force in Wave III is due to weaker control in the implementation of cordon sanitaire measures imposed in containing the virality spread. Based on the survival function the short-term forecasts showed that the number of new infections projected to decline from 23,282 cases in 28th week to 22,017 cases in 31st week. Similarly, based on the cumulative hazard function the duration estimated for containing the virality completely projected to stretch over another 19.6 weeks under the prevailing conditions.

Framework based on multiplicative error and residual analysis to forecast bitcoin intraday-volatility

This paper deals with bitcoin volatility forecasting, through a framework based on Long-Short Term Memory, multiplicative error and residual analysis. This framework is based on modeling the nonlinear behavior of the multiplicative error in order to improve forecast accuracy. The rapid growth of the cryptocurrency market, its high volatility and its applications in different commercial transactions have attracted the attention of academics and investors. Among cryptocurrencies, bitcoin is the one with the highest volume of trading. For this reason, it is important for investors and companies that use bitcoin as a means of payment to be able to predict its volatility more accurately. Nonetheless, this presents a great challenge given the chaotic behavior of the bitcoin volatility series. The results indicate that the proposed model is able to better capture non-linear behavior of the volatility time series, allowing a more accurate forecast in terms of MSE.

Forecaster Efficiency, Accuracy, and Disagreement: Evidence Using Individual‐Level Survey Data

AbstractTheories of expectations formation sometimes suppose that agents make efficient forecasts given their information sets. We use individual‐level data to test whether survey respondents' forecasts are efficient. We assess whether there are systematic differences between forecasters in terms of their degrees of contrarianism, and the accuracy of their forecasts, and whether these are explicable by inefficiencies in the use of information. We find that forecaster inefficiency cannot explain persistence in levels of disagreement across forecasters, but there is evidence that the inefficient use of information is responsible for persistent differences in accuracy across forecasters.

Energy Load Forecasting Using a Dual-Stage Attention-Based Recurrent Neural Network.

Providing a stable, low-price, and safe supply of energy to end-users is a challenging task. The energy service providers are affected by several events such as weather, volatility, and special events. As such, the prediction of these events and having a time window for taking preventive measures are crucial for service providers. Electrical load forecasting can be modeled as a time series prediction problem. One solution is to capture spatial correlations, spatial-temporal relations, and time-dependency of such temporal networks in the time series. Previously, different machine learning methods have been used for time series prediction tasks; however, there is still a need for new research to improve the performance of short-term load forecasting models. In this article, we propose a novel deep learning model to predict electric load consumption using Dual-Stage Attention-Based Recurrent Neural Networks in which the attention mechanism is used in both encoder and decoder stages. The encoder attention layer identifies important features from the input vector, whereas the decoder attention layer is used to overcome the limitations of using a fixed context vector and provides a much longer memory capacity. The proposed model improves the performance for short-term load forecasting (STLF) in terms of the Mean Absolute Error (MAE) and Root Mean Squared Errors (RMSE) scores. To evaluate the predictive performance of the proposed model, the UCI household electric power consumption (HEPC) dataset has been used during the experiments. Experimental results demonstrate that the proposed approach outperforms the previously adopted techniques.

Solar power plant generation forecasting using NARX neural network model: A case study

New technologies have been developed and adopted to generate energy from renewable sources to satisfy the increasing demand without causing environmental damage. However, estimating the power output of inherently intermittent, weather-driven, and non-dispatchable renewable energy sources is a major scientific and societal concern. In this study, a neural network model to enable short-to-middle term forecasts of a photovoltaic (PV) power system is provided. Using historical weather and power generation data, a non-linear autoregressive network with exogenous input (NARX) model is built to forecast the non-linear photovoltaic system output. The performance of the model is then analyzed by different statistical evaluation parameters. It is shown that the PV system power output estimation method can be successfully employed.

China's coal consumption forecasting using adaptive differential evolution algorithm and support vector machine

Coal consumption forecasting is the premise of energy supply structure reform and the basic work of energy planning under the background of climate change. Based on the econometrics which determines the influencing factors, this paper proposes a combined algorithm using Self-adaptive differential evolution (SaDE) algorithm and support vector (SVM) optimization algorithm for coal consumption forecasting. The optimization algorithm reduces the selection problem of SVM oversized hyperplane parameters, improves its global optimization ability, and further improves the prediction accuracy of SVM. The results of China's coal consumption forecasting show that the improved SaDE-SVM algorithm has good adaptability, robustness, faster converging speed and higher prediction accuracy for the prediction of less data samples and multiple influencing factors suitable for relevant medium and long term forecasts. The forecast shows that China's coal consumption is expected to be between 3.92 billion tons and 4.14 billion tons by 2020. By 2030, China's coal consumption will be between 2.195 billion tons and 3.699 billion tons.

A Missing Data on Covid-19 Forecasts

Mathematical and computational studies of Covid-19 have underestimated the influence that other countries have on their daily records. To visualize this, a Granger causality analysis was implemented in Python to determine if the cases registered in Brazil, Chile, Colombia, Ecuador, Panama, Paraguay, Peru and the USA have any effect on Venezuela, and between all of them. Finally, this paper highlights the need to incorporate causality analysis employing only the cases of Covid-19 to improve mid and long term forecasts.

Probabilistic Forecasting of Wind and Solar Farm Output

Accurately forecasting the output of grid connected wind and solar systems is critical to increasing the overall penetration of renewables on the electrical network. This includes not only forecasting the expected level, but also putting error bounds on the forecast. The National Electricity Market (NEM) in Australia operates on a five minute basis. We used statistical forecasting tools to generate forecasts with prediction intervals, trialing them on one wind and one solar farm. In classical time series forecasting, construction of prediction intervals is rudimentary if the error variance is constant—Termed homoscedastic. However, if the variance changes—Either conditionally as with wind farms, or systematically because of diurnal effects as with solar farms—The task is much more complicated. The tools were trained on segments of historical data and then tested on data not used in the training. Results from the testing set showed good performance using metrics, including Coverage and Interval Score. The methods used can be adapted to various time scales for short term forecasting.

Robust Satellite Techniques for mapping thermal anomalies possibly related to seismic activity of March 2021, Thessaly Earthquakes.

In recent years, there is a growing interest concerning the development of a multi-parametric system for earthquakes’ short term forecast identifying those parameters whose anomalous variations can be associated to the complex process of such events. In this context, the Robust Satellite Technique (RST) has been adopted herein with the aim to detect and map thermal anomalies probably related with the strong earthquake of M6.3 occurred near the city of Larissa, Thessaly on March 3rd 2021 10:16:07 UTC. For this purpose, 10 years (2012-2021) of daily Night-time Land Surface Temperature (LST) remotely sensed data from Moderate Resolution Imaging Spectroradiometer (MODIS), were analyzed. Pixels characterized by statistically significant LST variations on a daily scale were interpreted as an indicator of variations in seismic activity. Quite intense (Signal/Noise ratio > 2.5) and rare, spatially extensive and time persistent, TIR signal transients were identified, appearing twenty five days before the Thessaly main shock (pre-seismic anomalies: February 6th, February 11th March 1st), the day of the main earthquake (co-seismic anomaly) and after the main shock (post-seismic anomalies: March 4th, 10th and 17th). The final dataset of thermal anomalies was combined with geological and structural data of the area of interest, such as active faults, composite seismogenic sources, earthquake epicenter and topography in order to perform preliminary spatial analysis.