Monthly Rainfall Data Research Articles

Machine Language Approach for Modeling and Predicting Rainfall in Different Zones of Kerala

The forecasting of rainfall is said to be the most difficult of other hydrological processes due to sudden changes in atmospheric processes. The rainfall directly and indirectly influences agriculture and allied sectors. Sudden changes in rainfall or an uneven distribution of rainfall can lead to crop loss. In order to avoid such problems and take the necessary precautions, it is mandatory to forecast the rainfall using various models with maximum precision. In this study, rainfall for northern, central and southern Kerala, India, was predicted using an artificial neural network (ANN) with a multi-layer perceptron (MLP) feed-forward neural network and an extreme learning machine (ELM) neural network. The monthly rainfall data was collected for a period of 39 years (1982–2020) from the regional agricultural research stations (RARS), Pilicode and Pattambi, for the northern and central zones of Kerala, respectively, whereas for the southern zone of Kerala, data was collected from RARS, Vellayani, for a period of 36 years (1985–2020). For the rainfall data collected from three different zones of Kerala, the MLP and ELM were applied. The comparison and validation of MLP and ELM models was done based on the error values of mean square error (MSE), root mean square error (RMSE), and mean absolute error(MAE). The results indicated that for three different zones in Kerala, ANN with MLP showed better performance in forecasting rainfall compared to the ELM model. The best-selected model was also used for forecasting the next 5 years rainfall in each zone of Kerala.

ARIMA-Based forecasting of monthly rainfall in Mandi district, Himachal Pradesh

ABSTRACT To anticipate monthly rainfall time series, autoregressive integrated moving average (ARIMA) modelling was performed using R studio. For a period of 30 years (in mm), the monthly rainfall data were gathered for eight localities in the Mandi district of Himachal Pradesh: Bharol, Chachiyot, Jogindernagar, Karsog, Kataula, Mandi, Sarkaghat, and Sundernagar. The augmented Dickey–Fuller test was performed to determine whether the rainfall data were steady before applying the ARIMA model. The minimal values of the Akaike information criterion, Bayesian information criterion, autocorrelation function, and partial autocorrelation function were used to select the best models. The best fit was determined to be ARIMA (0,0,2) (2,1,0)12 for five stations and ARIMA (0,0,0) (1,1,0)12 for the remaining three stations. A box test is performed to validate the ARIMA model and verify the autocorrelation in the data. Using the ARIMA model, forecasts are made for the 10 years (2021–2030). The validity and performance of the ARIMA models were evaluated using error statistics (mean absolute error = 34.65, root-mean-square error = 33.58, and R2 = 0.93). The projected data's mean and standard deviation were close to the actual data. As a result, the ARIMA model's monthly rainfall prediction parameters were satisfactory.

Analysis of rainfall erosivity factor (R) on prediction of erosion yield using USLE and RUSLE Model’s; A case study in Mayang Watershed, Jember Regency, Indonesia

The Rainfall erosivity has a relatively high effect on soil erosion, in addition to being very difficult to predict and control. The Universal Soil Loss Equation (USLE) and The Revised Universal Soil Loss Equation (RUSLE) model are commonly used to predict erosion yield in Indonesia. However, these models have several erosivity formulations that give different results. In this sense, identifying the sensitivity of different erosivity formulations in both models above is important. The aim of this study is to analyze soil erosion yield prediction influenced by the difference in erosivity equation on the same rainfall data used in the models while other parameters used are the same. The monthly rainfall and annual rainfall data were tested using the erosivity formulas. The (1) Bols and (2) Utomo equations were tested using monthly rainfall data, while the (3) Bols and (4) Hurni equations were tested using annual rainfall data. The results show that the prediction of soil erosion yields estimates using monthly rainfall data in both models have no significant differences. On the other hand, soil erosion estimates using annual rainfall data in the models have significant differences, whereas the USLE model estimation results in 63% erosion yield on low classification (0-15 ton ha<sup>-1 </sup>year<sup>-1</sup>). Meanwhile, the RUSLE model estimates only 59% erosion yield on low classifications. Another result is that the USLE model estimates lower erosion yield than the RUSLE model when the models use annual rainfall data, which may give significantly different recommendations for soil conservation in Indonesia, especially in reducing erosion yield at the Watershed level.

Comparison of Monthly Rainfall Prediction using Long Short Term Memory and Multi Layer Perceptron Methods in South Tangerang City

Rainfall is one of the meteorological and climatological parameters whose information must be disseminated to the public and related stakeholders. Rainfall information has an important role in the sectors of people's lives. In agriculture, the amount of rainfall has an important role in determining the planting season, so that this can prevent potential crop failure. On Disaster, South Tangerang City during the 2016-2021 period experienced floods, landslides, and droughts. Therefore, the importance of rainfall prediction information can improve meteorological and climatological information services in various sectors. Nevertheless, it is still difficult for the community and stakeholders to get monthly rainfall predictions with high accuracy in the long term. In this research, monthly rainfall prediction is designed using MLP (Multi Layer Perceptron) and LSTM (Long Short Term Memory). The data used is the monthly rainfall data of Climate Hazards Group InfraRed Precipitations (CHIRPS) for 42 years (period 1981-2022) with coordinate boundaries according to the research location, namely South Tangerang City, which is located between 106.625 º - 106.825 º East and 6.4 ° - 6.2 ° LS as many as 16 grids with a resolution of 0.05 ° each grid. Monthly rainfall prediction using MLP produces an RMSE value of 90.19, and a MAPE of 40.55, while the LSTM method produces an RMSE value of 88.12 and a MAPE of 40.49. Monthly rainfall prediction results using the LSTM method are better than the MLP method; this can be seen from the RMSE value of the LSTM method is smaller than MLP.

Predicting Drought in East Nusa Tenggara: A Novel Approach Using Wavelet Fuzzy Logic and Support Vector Machines

The water crisis, or what is hereinafter referred to as drought, has become a systemic and crucial problem in several regions in Indonesia. Indonesia is an agricultural country, where the presence of water is very influential so that drought can become a natural disaster if it starts to cause an area to lose its source of income due to disturbances in agriculture and the ecosystem it causes. Drought forecasting can provide support solutions in preventing the impact of drought. In this paper, we compare the performance of wavelet fuzzy logic and the support vector machine (SVM) as a supervised learning method for drought forecasting in East Nusa Tenggara. This study examines the monthly rainfall data for 1999-2015 which is the basis for calculating the drought index based on the Standardized Precipitation Index (SPI). The SPI value used is SPI-3 at a station in East Nusa Tenggara. The performance of models is compareded on R2. The results showed that R2 of wavelet fuzzy logic is smaller than one of SVMVM is better than the wavelet fuzzy logic for forecasting SPI value of drought in East Nusa Tenggara.

Assessing the applicability of rainfall index spatial interpolation in predicting landslide susceptibility: A study in Brunei Darussalam

Abstract Rainfall is crucial in meteorology and hydrology, significantly impacting landslide risk assessment. This study focused on evaluating rainfall’s role in determining landslide susceptibility in Brunei Darussalam’s Jalan Kota Batu-Subok and Jalan Jangsak-Tutong regions. Using monthly rainfall data (2008 – 2018) from four weather stations, three spatial interpolation methods inverse distance weighting (IDW), radial basis function (RBF), and global polynomial interpolation (GPI) were assessed. The RBF proved superior in predicting rainfall distribution, evidenced by lower error metrics and higher correlation coefficients. The landslide susceptibility index (LSI) derived from the RBF’s rainfall interpolation showed high accuracy in identifying landslide-prone areas, with success rates between 89 % and 94.3 %, and prediction rates from 85.2 % to 95.9 % across the two areas studied. These findings suggest that the RBF-derived LSI is a reliable tool for landslide risk assessment. However, the LSI’s stability, irrespective of the rainfall data or interpolation method used, indicates that factors like terrain and human activities might have a more significant impact on landslide risks than rainfall alone. This research highlights the importance of considering various factors in landslide risk management and land-use planning, offering valuable insights for policymakers and local authorities.

FITTING G-FAMILY PROBABILITY DISTRIBUTION AT THE RAINFALL DATA OF RAJSHAHI DISTRICT, BANGLADESH

Classical statistical distributions were recognized tool in case of different areas such as biological science, economics, finance, engineering and environmental science. But these types of distribution have some limitations for modeling real data. Researchers are trying to find out the best performing statistical distribution for modeling environmental data by adding extra parameters in these existing distributions which provide super flexibility in modeling. These extend to new families of these distributions generally defined as G-family distribution. Rainfall is one of the most important natural resources for agricultural production. The excess and deficit amount of rainfall sometimes be dangerous for agriculture production. So, the prior knowledge about the pattern of rainfall will help to solve food insecurity, water erosion and water pollution problems. In this study we compare the performance of Gamma uniform G, Log gamma G type-I, Log gamma G type-II , Marshall-Olkin G and Weibull G distribution where we consider lomax, f, chen and Weibull distribution as baseline distribution for modeling the rainfall data in case of Rajshahi district Bangladesh. The monthly rainfall data from January 1971 to December 2020 consider as study period. The empirical study shows that Marshall-Olkin Chen distribution gives better fitting results in case of Rajshahi district. This study will help policy makers to become aware of upcoming situations for solving water related problems.

Rainfall erosivity assessment over a flooding basin, Kelani River basin, Sri Lanka

ABSTRACT This study evaluates the rainfall erosivity (RE) and erosivity density (ED) over the Kelani River basin, Sri Lanka for a period of 31 years (1990–2020). The river basin is well known for its annual floods during the southwestern monsoon season and severe erosion including landslides can be observed. The catchment was analyzed for its RE using the Wischmeier and Smith algorithm and for its ED using Kinnel's algorithm. The monthly rainfall data spreading over the river basin were used to analyze the monthly, seasonal, and annual RE and ED. Interestingly, the annual RE showed a linear increasing trend line over 31 years, and a maximum value of 2,831.41 MJ mm ha−1 h−1 yr−1 was able to be observed in the year 2016. The RE peaks in May which is in the southwestern monsoon season. This reveals that the risk of soil erosion in the basin is high in the southwestern monsoon season. In addition, land use and land cover changes over the years have adversely impacted the erosion rates. Therefore, it is highly recommended to investigate soil erosion in-depth and then implement relevant regulations to conserve the soil layers upstream of the river basin.

Assessment of climate change induced rainfall trend and variability with non-parametric and linear approach for Sirajganj district, Bangladesh

The monthly and annual trends and variance of rainfall have been studied for five stations in an economically important Bangladeshi district named Sirajganj since 1965 to 2021. Natural disasters have prevalent in Sirajganj which is indispensable to assess. But, several researchers have been normally focused on river bank management and flood risk assessment. However, no extensive research has been conducted on Sirajganj based on non-normally distributed time series meteorological data such as rainfall time series so the current study is very important. In this study, the non-parametric Mann-Kendall and Sen's methods have been used to determine the statistical significance of a positive or negative trend in rainfall data. Also, cumulative sum charts and bootstrapping, one-way ANOVA, Tukey's range tests, and linear regression have been used to discover the incidence of abrupt changes, compare the significant difference in monthly and annual rainfall data, multiple comparisons amidst mentioned stations to find changes, and to investigate the changeover on dry and rainy days, respectively. The analysis showed a statistically significant decreasing trends in monthly and annual rainfall series. As well, changes from positive to negative direction have been recognized in the February, May, July, September, and annual rainfall time sequence. Besides, ANOVA and Tukey's range tests revealed a statistically substantial difference in all monthly and annual rainfall volume excluding January, March, and June. Additionally, these two tests demonstrated momentous differences in all monthly and annual frequency of rainfall categories excepting January and April. However, Linear regression analysis revealed that the number of dry days gradually reduced at the end of the dry winter, though the number of rainy days decreased during the rainy season. As in, the number of rainy days replaces the number of dry days during the dry season and vice versa during the rainy season. Even though, with very few exceptions, the volume of rainfall decreases throughout the year. The outcomes of this research might helpful for implementing the planning and evaluating hydrological projects on Sirajganj district.

Power Data Access Viewer-Based Meteorological Drought Analysis and Rainfall Variability in the Nile River Basin

Meteorological drought poses a frequent challenge in the Nile River basin, yet its comprehensive evaluation across the basin has been hindered by insufficient recorded rainfall data. Common indices like the standard precipitation index, coefficients of variation, and precipitation concentration index serve as pivotal tools in gauging drought severity. This research aimed to assess the meteorological drought status in the Nile River basin by using the Power Data Access Viewer product rainfall data. Bias correction procedures were implemented to refine the monthly rainfall data for Bahirdar, Markos, Nekemt, and Muger stations, resulting in notable improvements in the coefficient of determination (R2) that were increased from 0.74 to 0.93, 0.72 to 0.89, 0.71 to 0.96, and 0.69 to 0.84, respectively. The average spatial distribution of drought in the Nile basin was classified as extremely wet (3.81%), severely wet (9.01%), moderately wet (7.36%), near normal (9.97%), moderately drought (21.20%), severely drought (17.11%), and extremely drought (31.54%). Approximately 10.33% of the Nile River basin was situated in regions characterized by high rainfall variability, while around 21.17% was located in areas with a notably irregular precipitation concentration index. Overall, this study sheds light on the prevailing meteorological drought patterns in the Nile River basin, emphasizing the significance of understanding and managing these phenomena for the sustainable development of the region.

FORECASTING RAINFALL VOLUME IN SELANGOR WITH A COMBINED ARIMA MODEL

Flash flood is the most hazardous type of flooding, mainly caused by extensive rainfall. It also can cause significant harm to a community’s economy, ecology, and society without warning at an irrational pace. Therefore, this study was conducted to detect the time series element within the rainfall data, select the optimal model, and make predictions about the volume of rainfall in Selangor. A variety of univariate time series models were utilized, including the naïve model, decomposition model, Autoregressive Integrated Moving Average (ARIMA) model, exponential models, and combined models. Historical monthly rainfall data collected from Petaling station and Subang station from 2018 to 2022 were used to estimate the parameters of the models, and the model was evaluated for the smallest error of measurements. Previous research mostly focused on complex methodologies for forecasting rainfall. However, this research aimed to identify a simple tool for fast prediction of rainfall. The results showed that the combination of the ARIMA (2,0,3) model from Petaling Station and the ARIMA (4,0,4) model from Subang station were able to capture the trends and seasons in the time series with the lowest error of measurement on short-term predictions of rainfall volume. Furthermore, the study delves into the concept of combined time series models, which are blended using weighted performance measures to enhance prediction accuracy further. The research acknowledges certain limitations of univariate time series models, notably their inability to account for intricate interactions among environmental variables and potential long-term trends, such as those stemming from climate change. Overall, the study explores the potential of combining models to refine predictions for forecasting rainfall volume in Klang Valley.

Liver condemnation due to bovine and buffalo fasciolosis in Antioquia province, Colombia

Background: Fasciola hepatica affects cattle and buffaloes, causing decreased production and liver condemnation at the slaughterhouse level. Objective: To determine liver condemnation frequency, temporal distribution, and associated factors, as well as the spatial distribution and economic impact of F. hepatica infection, based on post mortem inspection in cattle and buffaloes processed between 2010 and 2019 in one of the main slaughterhouses in Colombia. Methods: An observational, descriptive, and cross37 sectional study was carried out, with a convenience non-probability sampling. The temporal distribution of the frequency of liver condemnations was analyzed, the Pearson ꭕ2 test was used to determine the association between the dependent variable (i.e. presence/absence of adult parasites of F. hepatica) and the frequency rates between months and years, and the monthly rainfall data was compared with the accumulated monthly frequency. The association between the dependent variable and the type of production (i.e. industrial-type bovine cattle—intended for milk production, commercial44 type bovine cattle —intended for meat production), weight, and sex (odds ratios, OR and confidence intervals, 95%CI) was explored. Descriptive maps of the municipalities that presented at least one positive case of F. hepatica in 2019 and of the weighted prevalences, both for bovines and buffaloes, were designed. Only losses due to liver condemnation between 2010 and 2019 were considered for the analysis of the economic impact. Results: A total of 1,947,233 animals were processed during the study period, with an overall frequency of fasciolosis of 3.1%. Industrial-type cattle presented a 7.4% higher risk of developing F. hepatica compared to commercial-type cattle (OR=0.074; 95%CI: 0.07-0.08). Likewise, male cattle presented 3.88 times more risk compared to females (OR=3.88; 95%CI: 3.49-4.33). Economic losses due to liver condemnation were estimated at US$1,129,660 for the study period. Conclusion: The frequency of condemnations in Colombia occurs throughout the year, although it increases in the rainy season. The economic losses, only estimated by liver condemnation due to fasciolosis, are important. However, it is necessary to explore such losses considering other productive indicators and in larger animal populations. Fasciola hepatica was reported for the first time in the country in buffaloes. This work demonstrates the importance of systematizing the information generated by animal slaughterhouses.

Homogenizing Monthly Rainfall and Temperature Data Series in Maharashtra & Goa

Annual rainfall and temperature data series of all climate stations in Maharashtra & Goa are statistically tested for data homogeneity. To inspect homogeneity of a station, a two-step approach is followed. First, four homogeneity tests Standard normal homogeneity test, Pettit’s test, Buishand’s range test and Von Neumann ration test at 5% level of significance are used to determine test hypothesis for homogeneity on testing parameters of annual rainfall and temperature. Second, results from all these four tests aggregated together into three different classes as ‘useful’, ‘doubtful’ and ‘suspect’. Here 30 rainfall, 29 maximum and minimum temperature climate stations were tested. The results showed 80% stations as ‘useful’, 7% as ‘suspect’ and 13% as ‘doubtful’ for rainfall, for maximum temperature series these results are 17% as ‘useful’, 7% as ‘suspect’ and 76% as ‘doubtful’, while for minimum temperature series these results are 21% as ‘useful’, 10% as ‘suspect’ and 69% as ‘doubtful’. Further, in this study an attempt is also made to correct the monthly rainfall and temperature data series for homogeneity. Stations categorised as ‘useful’ are used as reference series to remove inhomogeneities from ‘suspect’ and ‘doubtful’ stations. To correct rainfall series ratio’s method is used while for temperature series addition method is used. Correction results showed significant improvement in ‘suspect’ category stations. After correction of inhomogeneous series, the results shows all 100% of rainfall stations and more than 65% of temperature stations are now in ‘useful’ category. The corrected stations may be included in further climate research studies.

Erratic Rainfall and its Upshot on Rice Production in Kerala

In recent years, rainfall irregularities have affected crop production in Kerala. Rice, a water-reliant crop, is the staple food of people in Kerala. It is challenging to revive paddy if the soil loses even 10 to 20 per cent of its moisture. Due to erratic rains and frequent dry spells, rain-dependent paddy cultivation is nearly impossible in Kerala. Among the three seasons of paddy cultivation in Kerala, the autumn season entirely depends on rainfall. According to 2020-21 Kerala agriculture statistics, the majority of the rice is cultivated under irrigated conditions in the state. This loss in rain-fed paddy cultivation was due to the decline in autumn rice cultivation after the mid-1990s. The study analysed the annual and seasonal rainfall variability and their relationships with autumn rice production in Kerala. The variability and concentration in the annual and seasonal rainfall were analysed using the coefficient of variation and the precipitation concentration index, respectively. The results revealed that the overall rainfall distribution pattern in Kerala was moderate to irregular. The seasonal index was estimated for the monthly rainfall data, and it was found that in June and July, the rainfall was the maximum. Production and productivity of autumn rice showed a high correlation with rains in June and July. The variability in rice production was positively influenced by rainfall and negatively influenced by the standard deviation of precipitation.

Hybridized gated recurrent unit with variational mode decomposition and an error compensation mechanism for multi-step-ahead monthly rainfall forecasting.

Highly accurate monthly rainfall predictions can provide early warnings for rain-related disasters, such as floods and droughts, and allow governments to make timely decisions. This paper proposes a two-phase error compensation model based on a gated recurrent unit (GRU), variational mode decomposition (VMD), and error compensation mechanism (ECM) (GRU-VMD-ECM) for accurate multi-step-ahead monthly rainfall forecasts. In the first phase, the GRU model is used to make an initial monthly rainfall prediction, and the error series is extracted. In the second phase, the error series is decomposed into eight subseries using the VMD method. Each subseries is then input into the GRU model to build different forecasting models. These predicted error sequences are added to the initial prediction results to obtain the final forecast. The model's performance is tested using six evaluation indicators based on Beijing's monthly rainfall data from 1951 to 2018. The results show that the error compensation mechanism significantly improved the prediction accuracy, particularly in the Nash-Sutcliffe efficiency (NSE) of single-step-ahead prediction which recorded a substantial increase of 281.16% from 0.259981 to 0.990944, as well as a decrease in root mean square error (RMSE) from 2.257580 to 0.249746. Furthermore, the GRU-VMD-ECM model outperforms the RF, GRU-CNN, and VMD-GRU models in terms of precision across all forecasting horizons. These findings highlight the potential of the GRU-VMD-ECM model in providing highly accurate monthly rainfall predictions for early warnings and informed decision-making by governments.

Fitting Statistical Distributions to Rainfall Data with Different Estimation Techniques: An Empirical Study from Pabna and Dinajpur Districts

Bangladesh is a country of diverse climatic conditions because of its rainfall and other geographical conditions which have a complex impact on economic and social aspects. The statistical distributions are used in many real life data for modeling and predicting. Knowing the real distribution of rainfall rather than depending on basic summary statistics would improve many uses of rainfall data. The aim of this paper is twofold: first, the performance of different statistical distributions such as Normal, Log-Normal, Gamma Weibull, and Gumbel are compared for modeling the monthly rainfall data from Pabna and Dinajpur districts from January 1971 to December 2015; second, the performance of the Maximum Likelihood Estimation (MLE), Quantile Matching Estimation (QME), and the Maximum Spacing Estimation are also compared for fitting these statistical distributions. The empirical study showed that Gamma distribution performs better for fitting the monthly rainfall data for both Pabna and Dinajpur districts of the three methods like Maximum Likelihood Estimation (MLE) method, the Quantile Matching Estimation (QME), and the Maximum Spacing Estimation (MSE) method. The Normal distribution performs worse of these study areas. By the comparison of these three methods we found that Maximum Likelihood Estimation (MLE) gives better results. This study provides the actual distribution of rainfall data of these study areas. International Journal of Statistical Sciences, Vol. 23(2), November, 2023, pp 87-106

Decadal Rainfall Trends and Variability Across Nigeria

This work examined the decadal trends and variability of rainfall data over Nigeria from 1979 to 2021 (42 years). Observational monthly rainfall data was obtained from the Nigerian Meteorological Agency (NiMET). The data underwent statistical analysis to illustrate its spread and variability using metrics such as mean, standard deviation, and coefficient of variation, linear regression was applied to reveal the trends or changes over time, the coefficient of correlation was employed to assess the statistical relationships of rainfall across the distinct climate regions of Guinea, Savanna, and Sahel, respectively. There were varying levels of annual and seasonal rainfall across the regions, with Port Harcourt receiving the highest annual and seasonal rainfall in the Guinea region. Sokoto and Maiduguri exhibit the highest annual rainfall in the Sahel region. The decadal analysis highlights the fluctuations in rainfall anomalies, as some decades showed surplus rainfall while others displayed negative deviations, indicating the changing nature of regional rainfall patterns.

The GEO-SPATIAL ASSESSMENT OF THE INFLUENCE OF TEMPERATURE AND RAINFALL ON THE OCCURRENCE OF CHOLERA IN KANO STATE, NIGERIA

Temperature and rainfall variability as an important factor that could affect the occurrence of Cholera becomes one of the major concerns globally. This study assessed the influence of these factors on the occurrence of Cholera in Kano State, Nigeria. The researchers employed the used of Cholera surveillance data obtained from Kano State Ministry of Health, the mean monthly temperature, and total monthly rainfall data downloaded from the climatic research unit of the University of East Anglia archive (2010 to 2019). The data were analyzed using Ordinary Least Squares (OLS) regression and Geographically Weighted Regression (GWR) in ArcMap 10.8 environment. Results showed that an increase in temperature by 1 unit positively influences the occurrence of Cholera by 19.69, 11.7 and 18.7 unit during hot/dry, warm/wet, and cold/dry season respectively. Similarly, an increase in rainfall by 1 unit brings about a rise in the outbreak of Cholera by 0.67 and 0.97 unit during hot/dry and warm/wet season respectively. In addition, highest influence of temperature and rainfall on the outbreak of Cholera during warm/wet season with R2 value of 0.25, and lowest during the cold/dry season with R2 value at 0.06. Northern part of the State experienced the strongest influence (R2 values of 0.2) while Southern part recorded the lowest influence with R2 values of 0.04 to 0.07. Variability in temperature and rainfall in Kano State were directly related with the outbreak of Cholera. Hence, the need for more environmental sanitation programs and public campaign as climate related mitigation and adaptation strategies.

Machine Learning based IoT Flood Rediction Using Data Modeling and Decision Support System

An essential step in supplying data for climate impact studies and evaluations of hydrological processes is rainfall prediction. However, rainfall events are complex phenomenon’s that continue to be difficult to forecast. In this paper , we present unique hybrid models for the prediction of monthly precipitation that include Seasonal Artificial Neural Networks and Discrete wavelet transforms are two pre-processing methods, together with Artificial Neural Networks have two feed forward neural networks. The temporal series of observed monthly rainfall from Vietnam’s Ca Mau hydrological station were decomposed into three subsets by seasonal decomposition and five sub signals and four levels by wavelet analysis. The methods for predicting rainfall that use feed forward artificial neural networks (ANN) and seasonal artificial neural network (SANN) were fed with the processed data. The classic genetic method and simulated annealing method backed by using an integrated moving average and autoregressive moving was contrasted with the predicted models for model evaluation. The results showed that non-stationary regarding issues with non-linear time series, such forecasting rainfall could be satisfactorily simulated. The SANN model was integrated with the wavelet transform and seasonal decomposition are both used. Techniques, however the wavelet transform method produced the most accurate monthly rainfall data, Predictions. Due to the effects of climate change, nations including the Japan, China, the United States of America, and Taiwan, etc., have recently experienced severe and devastating natural disasters. One of the biggest causes of the destruction in Asian nations like china, India, Bangladesh, Sri Lanka, etc. is the flood. The danger of fatality from these floods is increased by 78% as information technology advances; there is a demand for simple access to massive amounts of cloud storage and computing capacity.

Combined use of graphical and statistical approaches for rainfall trend analysis in the Mae Klong River Basin, Thailand

Abstract The main purpose of this paper was to investigate the monthly, seasonal, and annual rainfall variability in the Mae Klong River Basin in Thailand using the Mann–Kendall (MK) test, Sen's slope method, Spearman's Rho (SR) test, and the innovative trend analysis (ITA) method. The monthly rainfall data of eight stations for the period 1971–2015 were used for trend analysis. Datasets with significant serial correlation were corrected by the trend-free pre-whitening (TFPW) approach for statistical methods. The MK test showed increasing rainfall trends for five out of eight stations in the dry season while 50% of stations indicated increasing trends in the wet season. On an annual scale, 75% of the stations exhibited increasing rainfall trends. The results of the SR test were in line with the MK test for seasonal and annual rainfall. The ITA method showed comparable findings with those of the statistical methods. For the entire basin, trend analysis found increasing rainfall on both seasonal and annual scales by all the tests. The findings of this study could benefit water supply and management, drought monitoring, agricultural production activities, and socioeconomic development in the Mae Klong River Basin in the future.