Assessment of Seasonal Rainfall Prediction in Ethiopia: Evaluating a Dynamic Recurrent Neural Network to Downscale ECMWF-SEAS5 Rainfall

Seasonal rainfall plays a vital role in both environmental dynamics and decision-making for rainfed agriculture in Ethiopia, a country often impacted by extreme climate events such as drought and flooding. Predicting the onset of the rainy season and providing localized rainfall forecasts for Ethiopia is challenging due to the changing spatiotemporal patterns and the country's rugged topography. The Climate Hazards Group Infra Red Precipitation with Station Data(CHIRPS), ERA5-Land total precipitation and temperature data are used from 1981–2022 to predict spatial rainfall by applying an artificial neural network(ANN). The recurrent neural network(RNN) is a nonlinear autoregressive network with exogenous input(NARX), which includes feed-forward connections and multiple network layers, employing the Levenberg Marquart algorithm. This method is applied to downscale data from the European Centre for Medium-range Weather Forecasts fifth-generation seasonal forecast system(ECMWF-SEAS5) and the Euro-Mediterranean Centre for Climate Change(CMCC) to the specific locations of rainfall stations in Ethiopia for the period 1980–2020. Across the stations, the results of NARX exhibit strong associations and reduced errors. The statistical results indicate that, except for the southwestern Ethiopian highlands, the downscaled monthly precipitation data exhibits high skill scores compared to the station records, demonstrating the effectiveness of the NARX approach for predicting local seasonal rainfall in Ethiopia's complex terrain. In addition to this spatial ANN of the summer season precipitation, temperature, as well as the combination of these two variables, show promising results.

Application of ATOVS Microwave Radiance Assimilation to Rainfall Prediction in Summer 2004

Experiments are performed in this paper to understand the influence of satellite radiance data on the initial field of a numerical prediction system and rainfall prediction. First, Advanced Microwave Sounder Unit A （AMSU-A） and Unit B （AMSU-B） radiance data are directly used by three-dimensional variational data assimilation to improve the background field of the numerical model. Then, the detailed effect of the radiance data on the background field is analyzed. Secondly, the background field, which is formed by application of Advanced Television and Infrared Observation Satellite Operational Vertical Sounder （ATOVS） microwave radiance assimilation, is employed to simulate some heavy rainfall cases. The experiment results show that the assimilation of AMSU-A （B） microwave radiance data has a certain impact on the geopotential height, temperature, relative humidity and flow fields. And the impacts on the background field are mostly similar in the different months in summer. The heavy rainfall experiments reveal that the application of AMSU-A （B） microwave radiance data can improve the rainfall prediction significantly. In particular, the AMSU-A radiance data can significantly enhance the prediction of rainfall above 10 mm within 48 h, and the AMSU-B radiance data can improve the prediction of rainfall above 50 mm within 24 h. The present study confirms that the direct assimilation of satellite radiance data is an effective way to improve the prediction of heavy rainfall in the summer in China.

Improving Multi-model Ensemble Probabilistic Prediction of Yangtze River Valley Summer Rainfall

Seasonal prediction of summer rainfall over the Yangtze River valley（YRV） is valuable for agricultural and industrial production and freshwater resource management in China, but remains a major challenge. Earlier multi-model ensemble（MME） prediction schemes for summer rainfall over China focus on single-value prediction, which cannot provide the necessary uncertainty information, while commonly-used ensemble schemes for probability density function（PDF） prediction are not adapted to YRV summer rainfall prediction. In the present study, an MME PDF prediction scheme is proposed based on the ENSEMBLES hindcasts. It is similar to the earlier Bayesian ensemble prediction scheme, but with optimization of ensemble members and a revision of the variance modeling of the likelihood function. The optimized ensemble members are regressed YRV summer rainfall with factors selected from model outputs of synchronous 500-h Pa geopotential height as predictors. The revised variance modeling of the likelihood function is a simple linear regression with ensemble spread as the predictor. The cross-validation skill of 1960–2002 YRV summer rainfall prediction shows that the new scheme produces a skillful PDF prediction, and is much better-calibrated, sharper, and more accurate than the earlier Bayesian ensemble and raw ensemble.

A Statistical-Dynamical Scheme for the Extraseasonal Prediction of Summer Rainfall for 160 Observation Stations across China

The purpose of this study was to design and test a statistical-dynamical scheme for the extraseasonal（one season in advance） prediction of summer rainfall at 160 observation stations across China.The scheme combined both valuable information from the preceding observations and dynamical information from synchronous numerical predictions of atmospheric circulation factors produced by an atmospheric general circulation model.First,the key preceding climatic signals and synchronous atmospheric circulation factors that were not only closely related to summer rainfall but also numerically predictable were identified as the potential predictors.Second,the extraseasonal prediction models of summer rainfall were constructed using a multivariate linear regression analysis for 15 subregions and then 160 stations across China.Cross-validation analyses performed for the period 1983-2008 revealed that the performance of the prediction models was not only high in terms of interannual variation,trend,and sign but also was stable during the whole period.Furthermore,the performance of the scheme was confirmed by the accuracy of the real-time prediction of summer rainfall during 2009 and 2010.

An Approach for Improving Short-Term Prediction of Summer Rainfall over North China by Decomposing Interannual and Decadal Variability

A statistical downscaling approach was developed to improve seasonal-to-interannual prediction of summer rainfall over North China by considering the effect of decadal variability based on observational datasets and dynamical model outputs.Both predictands and predictors were first decomposed into interannual and decadal components.Two predictive equations were then built separately for the two distinct timescales by using multivariate linear regressions based on independent sample validation.For the interannual timescale,850-hPa meridional wind and 500-hPa geopotential heights from multiple dynamical models＇ hindcasts and SSTs from observational datasets were used to construct predictors.For the decadal timescale,two well-known basin-scale SST decadal oscillation （the Atlantic Multidecadal Oscillation and the Pacific Decadal Oscillation） indices were used as predictors.Then,the downscaled predictands were combined to represent the predicted/hindcasted total rainfall.The prediction was compared with the models＇ raw hindcasts and those from a similar approach but without timescale decomposition.In comparison to hindcasts from individual models or their multi-model ensemble mean,the skill of the present scheme was found to be significantly higher,with anomaly correlation coefficients increasing from nearly neutral to over 0.4 and with RMSE decreasing by up to 0.6 mm d-1.The improvements were also seen in the station-based temporal correlation of the predictions with observed rainfall,with the coefficients ranging from-0.1 to 0.87,obviously higher than the models＇ raw hindcasted rainfall results.Thus,the present approach exhibits a great advantage and may be appropriate for use in operational predictions.

A STATISTICAL PREDICTION MODEL FOR SPRING RAINFALL OVER NORTHERN CHINA BASED ON THE INTERANNUAL INCREMENT APPROACH

The spring (March-April-May) rainfall over northern China (SPRNC) is predicted by using the interannual increment approach. DY denotes the difference between the current year and previous years. The seasonal forecast model for the DY of SPRNC is constructed based on the data that are taken from the 1965-2002 period (38 years), in which six predictors are available no later than the current month of February. This is favorable so that the seasonal forecasts can be made one month ahead. Then, SPRNC and the percentage anomaly of SPRNC are obtained by the predicted DY of SPRNC. The model performs well in the prediction of the inter-annual variation of the DY of SPRNC during 1965-2002, with a correlation coefficient between the predicted and observed DY of SPRNC of 0.87. This accounts for 76% of the total variance, with a low value for the average root mean square error (RMSE) of 20%. Both the results of the hindcast for the period of 2003-2010 (eight years) and the cross-validation test for the period of 1965-2009 (45 years) illustrate the good prediction capability of the model, with a small mean relative error of 10%, an RMSE of 17% and a high rate of coherence of 87.5% for the hindcasts of the percentage anomaly of SPRNC.

Improvements in Long-Lead Prediction of Early-Summer Subtropical Frontal Rainfall Based on Arctic Sea Ice

Seasonal prediction of East Asia(EA) summer rainfall, especially with a longer-lead time, is in great demand, but still very challenging. The present study aims to make long-lead prediction of EA subtropical frontal rainfall(SFR) during early summer(May-June mean, MJ) by considering Arctic sea ice(ASI) variability as a new potential predictor. A MJ SFR index(SFRI), the leading principle component of the empirical orthogonal function(EOF) analysis applied to the MJ precipitation anomaly over EA, is defined as the predictand. Analysis of 38-year observations(1979-2016) revealed three physically consequential predictors. A stronger SFRI is preceded by dipolar ASI anomaly in the previous autumn, a sea level pressure(SLP) dipole in the Eurasian continent, and a sea surface temperature anomaly tripole pattern in the tropical Pacific in the previous winter. These precursors foreshadow an enhanced Okhotsk High, lower local SLP over EA, and a strengthened western Pacific subtropical high. These factors are controlling circulation features for a positive SFRI. A physical-empirical model was established to predict SFRI by combining the three predictors. Hindcasting was performed for the 1979-2016 period, which showed a hindcast prediction skill that was, unexpectedly, substantially higher than that of a four-dynamical models’ ensemble prediction for the 1979-2010 period(0.72 versus 0.47). Note that ASI variation is a new predictor compared with signals originating from the tropics to mid-latitudes. The long-lead hindcast skill was notably lower without the ASI signals included, implying the high practical value of ASI variation in terms of long-lead seasonal prediction of MJ EA rainfall.

Seasonal Prediction for May Rainfall over Southern China Based on the NCEP CFSv2

In this study,we assess the prediction for May rainfall over southern China(SC)by using the NCEP CFSv2 outputs.Results show that the CFSv2 is able to depict the climatology of May rainfall and associated circulations.However,the model has a poor skill in predicting interannual variation due to its poor performance in capturing related anomalous circulations.In observation,the above-normal SC rainfall is associated with two anomalous anticyclones over the western tropical Pacific and northeastern China,respectively,with a low-pressure convergence in between.In the CFSv2,however,the anomalous circulations exhibit the patterns in response to the El Ni?o-Southern Oscillation(ENSO),demonstrating that the model overestimates the relationship between May SC rainfall and ENSO.Because of the onset of the South China Sea monsoon,the atmospheric circulation in May over SC is more complex,so the prediction for May SC rainfall is more challenging.In this study,we establish a dynamic-statistical forecast model for May SC rainfall based on the relationship between the interannual variation of rainfall and large-scale ocean-atmosphere variables in the CFSv2.The sea surface temperature anomalies(SSTAs)in the northeastern Pacific and the centraleastern equatorial Pacific,and the 500-h Pa geopotential height anomalies over western Siberia in previous April,which exert great influence on the SC rainfall in May,are chosen as predictors.Furthermore,multiple linear regression is employed between the predictors obtained from the CFSv2 and observed May SC rainfall.Both cross validation and independent test show that the hybrid model significantly improve the model’s skill in predicting the interannual variation of May SC rainfall by two months in advance.

A WD-GA-LSSVM model for rainfall-triggered landslide displacement prediction

This paper proposes a WD-GA-LSSVM model for predicting the displacement of a deepseated landslide triggered by seasonal rainfall,in which wavelet denoising(WD)is used in displacement time series of landslide to eliminate the GPS observation noise in the original data,and genetic algorithm(GA)is applied to obtain optimal parameters of least squares support vector machines(LSSVM)model.The model is first trained and then evaluated by using data from a gentle dipping(~2°-5°)landslide triggered by seasonal rainfall in the southwest of China.Performance comparisons of WD-GA-LSSVM model with Back Propagation Neural Network(BPNN)model and LSSVM are presented,individually.The results indicate that the adoption of WD-GA-LSSVM model significantly improves the robustness and accuracy of the displacement prediction and it provides a powerful technique for predicting the displacement of a rainfall-triggered landslide.

An Effective Approach for Improving the Real-Time Prediction of Summer Rainfall over China

This paper has two purposes. One is to evaluate the ability of an atmospheric general circulation model (IAP9L-AGCM) to predict summer rainfall over China one season in advance. The other is to propose a new approach to improve the predictions made by the model. First, a set of hindcast experiments for summer climate over China during 1982-2010 are performed from the perspective of real-time prediction with the IAP9L-AGCM model and the IAP ENSO prediction system. Then a new approach that effectively combines the hind-cast with its correction is proposed to further improve the model's predictive ability. A systematic evaluation reveals that the model's real-time predictions for 41 stations across China show significant improvement using this new approach, especially in the lower reaches between the Yellow River and Yangtze River valleys.

Joint Probability Prediction Model of Rainfall Triggered Landslides and Debris Flows

The rainfall induced landslides and debris flows are the major disasters in China, as well in Europe, South America, Japan and Australia. This paper proposes a new type of joint probability prediction model—Double Layer Nested Multivariate Compound Extreme Value Distribution (DLNMCEVD) to predict landslides and debris flows triggered by rainfall. The outer layer of DLNMCEVD is predicting the joint probabilities of different combinations for rainfall characteristics, air temperature and humidity, which should be considered as external load factors with geological and geotechnical characteristics as resistance factors for reliability analysis of slope stability in the inner layer of model. For the reliability and consequence analysis of rainfall-induced slope failure, the Global Uncertainty Analysis and Global Sensitivity Analysis (GUA & GSA) should be taken into account for input-output iterations. Finally, based on the statistics prediction by DLNMCEVD, the geological hazards prevention alarm and regionalization can be provided in this paper.

Seasonal Prediction of June Rainfall over South China:Model Assessment and Statistical Downscaling

The performances of various dynamical models from the Asia-Pacific Economic Cooperation（APEC） Climate Center（APCC） multi-model ensemble（MME） in predicting station-scale rainfall in South China（SC） in June were evaluated.It was found that the MME mean of model hindcasts can skillfully predict the June rainfall anomaly averaged over the SC domain.This could be related to the MME＇s ability in capturing the observed linkages between SC rainfall and atmospheric large-scale circulation anomalies in the Indo-Pacific region.Further assessment of station-scale June rainfall prediction based on direct model output（DMO） over 97 stations in SC revealed that the MME mean outperforms each individual model.However,poor prediction abilities in some in-land and southeastern SC stations are apparent in the MME mean and in a number of models.In order to improve the performance at those stations with poor DMO prediction skill,a station-based statistical downscaling scheme was constructed and applied to the individual and MME mean hindcast runs.For several models,this scheme can outperform DMO at more than 30 stations,because it can tap into the abilities of the models in capturing the anomalous Indo-Paciric circulation to which SC rainfall is considerably sensitive.Therefore,enhanced rainfall prediction abilities in these models should make them more useful for disaster preparedness and mitigation purposes.

Spider Monkey Optimization with Statistical Analysis for Robust Rainfall Prediction

Rainfall prediction becomes popular in real time environment due to the developments of recent technologies.Accurate and fast rainfall predictive models can be designed by the use of machine learning(ML),statistical models,etc.Besides,feature selection approaches can be derived for eliminating the curse of dimensionality problems.In this aspect,this paper presents a novel chaotic spider money optimization with optimal kernel ridge regression(CSMO-OKRR)model for accurate rainfall prediction.The goal of the CSMO-OKRR technique is to properly predict the rainfall using the weather data.The proposed CSMO-OKRR technique encompasses three major processes namely feature selection,prediction,and parameter tuning.Initially,the CSMO algorithm is employed to derive a useful subset of features and reduce the computational complexity.In addition,the KRR model is used for the prediction of rainfall based on weather data.Lastly,the symbiotic organism search(SOS)algorithm is employed to properly tune the parameters involved in it.A series of simulations are performed to demonstrate the better performance of the CSMO-OKRR technique with respect to different measures.The simulation results reported the enhanced outcomes of the CSMO-OKRR technique with existing techniques.

Rainfall Prediction Model Improvement by Fuzzy Set Theory

This paper presents the improvement of the fuzzy inference model primarily developed for predicting rainfall with data from United States Department of Agriculture (USDA) Soil Climate Analysis Network (SCAN) Station at the Alabama Agricultural and Mechanical University (AAMU) Campus for the year 2004. The primary model was developed with Fuzzy variables selected based on the degree of association of different factors with various combinations causing rainfall. An increase in wind speed (WS) and a decrease in temperature (TP) when compared between the ith and (i-1)th day were found to have a positive relation with rainfall. Results of the model showed better performance after introducing the threshold values of 1) relative humidity (RH) of the ith day;2) humidity increase (HI) when compared between the ith and (i-1)th day;and 3) product (P) of increase in wind speed (WS) and decrease in temperature (TP) when compared between the ith and (i-1)th day. In case of the improved model, errors between actual and calculated amount of rainfall (RF) were 1.20%, 2.19%, and 9.60% when using USDA-SCAN data from AAMU campus for years 2003, 2004 and 2005, respectively. The improved model was tested at William A. Thomas Agricultural Research Station (WTARS) and Bragg farm in Alabama to check the applicability of the model. The errors between the actual and calculated amount of rainfall (RF) were 3.20%, 5.90%, and 1.66% using USDA-SCAN data from WATARS for years 2003, 2004, and 2005, respectively. Similarly, errors were 10.37%, 11.69%, and 25.52% when using SCAN data from Bragg farm for years 2004, 2005, and 2006, respectively. The primary model yielded the value of error equals 12.35% using USDA- SCAN data from AAMU campus for 2004. The present model performance was proven to be better than the primary model.

Application of Ordinal Set Pair Analysis in Annual Rainfall Prediction of Liao River Basin

[Objective] The research aimed to study the application of ordinal set pair analysis in the annual precipitation prediction of Liao River basin.[Method] The ordinal theory was introduced into the set pair analysis modeling,and the prediction model of set pair analysis was improved.A kind of rainfall prediction model based on the ordinal set pair analysis (OSPA) was put forward.The time sequence of annual rainfall in the hydrological rainfall station of Liao River basin during 1956-2006 was the research objective.The annual rainfall during 1998-2006 was predicted by the model,and the error analysis was given.[Result] In the relative errors of predicted results by ordinal set pair analysis,there were six relative errors within 5%,which occupied 66.7% of the total prediction number.One relative error was during 5%-10%,which occupied 11.1% of the total prediction number.Two relative errors were during 10%-15%,which occupied 22.2% of the total prediction number.All the relative errors were less than 20%,which met the precision requirement of annual rainfall prediction in Forecast Specification of Hydrological Information.[Conclusion] The rainfall prediction based on the ordinal set pair analysis model had high precision,and the prediction result was ideal.It was suitable for the annual rainfall prediction.

Long-term Prediction and Verification of Rainfall Based on the Seasonal Model

Using the seasonal cross-multiplication trend model, monthly precipitation of eight national basic weather stations of Shaanxi Province from 2005 to 2010 was predicted, and the forecast results were verified using the rainfall scoring rules of China Meteorological Administration. The verification results show that the average score of annual precipitation prediction in recent six years is higher than that made by a professional forecaster, so this model has a good prospect of application. Moreover, the level of making prediction is steady, and it can be widely used in long-term prediction of rainfall.

Case Study: ENSO Events, Rainfall Variability and the Potential of SOI for the Seasonal Precipitation Predictions in Iran

The studies in recent decades show that many natural disasters such as tropical severe storms, hurricanes development, torrential rain, river flooding, and landslides in some regions of the world and severe droughts and wildfires in other areas are due to El Nino-Southern Oscillation (ENSO). This research aims to contribute to an improved definition of the relation between ENSO and seasonal (autumn and winter) variability of rainfall over Iran. The results show that during autumn, the positive phase of SOI is associated with decrease in the rainfall amount in most part of the country;negative phase of SOI is associated with a significant increase in the rainfall amount. It is also found that, during the winter time when positive phase of SOI is dominant, winter precipitation increases in most areas of the eastern part of the country while at the same time the decreases in the amount of rainfall in other parts is not significant. Moreover, with negative phase of SOI in winter season the amount of rainfall in most areas except south shores of Caspian Sea in the north decreases, so that the decrease of rainfall amount in the eastern part is statistically significant.

Data Mining with Comprehensive Oppositional Based Learning for Rainfall Prediction

Data mining process involves a number of steps fromdata collection to visualization to identify useful data from massive data set.the same time,the recent advances of machine learning(ML)and deep learning(DL)models can be utilized for effectual rainfall prediction.With this motivation,this article develops a novel comprehensive oppositionalmoth flame optimization with deep learning for rainfall prediction(COMFO-DLRP)Technique.The proposed CMFO-DLRP model mainly intends to predict the rainfall and thereby determine the environmental changes.Primarily,data pre-processing and correlation matrix(CM)based feature selection processes are carried out.In addition,deep belief network(DBN)model is applied for the effective prediction of rainfall data.Moreover,COMFO algorithm was derived by integrating the concepts of comprehensive oppositional based learning(COBL)with traditional MFO algorithm.Finally,the COMFO algorithm is employed for the optimal hyperparameter selection of the DBN model.For demonstrating the improved outcomes of the COMFO-DLRP approach,a sequence of simulations were carried out and the outcomes are assessed under distinct measures.The simulation outcome highlighted the enhanced outcomes of the COMFO-DLRP method on the other techniques.

Improvement of Rainfall Prediction Model by Using Fuzzy Logic

This paper presents the improvement of the fuzzy inference model for predicting rainfall. Fuzzy rule based system is used in this study to predict rainfall. Fuzzy inference is the actual procedure of mapping with a given set of input and output through a set of fuzzy systems. Two operations were performed on the fuzzy logic model;the fuzzification operation and defuzzification operation. This study is obtaining two input variables and one output variable. The input variables are temperature and wind speed at a particular time and output variable is the amount of predictable rainfall. Temperature, wind speed and rainfall have to construct eight equations for different categories and which are shows the diagram of the graph. Fuzzy levels and membership functions obtained after minimum composition of inference part of the fuzzifications done for temperature and wind speed are considered as they represent the environmental condition enhance a rainfall occurrence which is effect on agricultural production.

Predictability of Winter Rainfall in South China as Demonstrated by the Coupled Models of ENSEMBLES

Winter rainfall over South China shows strong interannual variability,which accounts for about half of the total winter rainfall over South China.This study investigated the predictability of winter （December-January-February; DJF） rainfall over South China using the retrospective forecasts of five state-of-the-art coupled models included in the ENSEMBLES project for the period 1961-2006.It was found that the ENSEMBLES models predicted the interannual variation of rainfall over South China well,with the correlation coefficient between the observed/station-averaged rainfall and predicted/areaaveraged rainfall being 0.46.In particular,above-normal South China rainfall was better predicted,and the correlation coefficient between the predicted and observed anomalies was 0.64 for these wetter winters.In addition,the models captured well the main features of SST and atmospheric circulation anomalies related to South China rainfall variation in the observation.It was further found that South China rainfall,when predicted according to predicted DJF Nifio3.4 index and the ENSO-South China rainfall relationship,shows a prediction skill almost as high as that directly predicted,indicating that ENSO is the source for the predictability of South China rainfall.