Missing Rainfall Data Research Articles

Assessment of Different Methods for Estimation of Missing Rainfall Data

Missing data is a common problem encountered in various fields, including clinical research, environmental sciences and hydrology. In order to obtain reliable results from the analysis, the data inventory must be completed. This paper presents a methodology for addressing the missing data problem by examining the missing data structure and missing data techniques. Simulated datasets were created by considering the number of missing data, missing data pattern and missing data mechanism of real datasets containing missing values, which are often overlooked in hydrology. Considering the missing data pattern, the most commonly used methods for missing data analysis in hydrology and other fields were applied to the created simulated datasets. Simple imputation techniques and expectation maximization (EM) were implemented in SPSS software and machine learning techniques such as k-nearest neighbor (kNN), together with the hot-deck were implemented in the Python programming language. In the performance evaluation based on error metrics, it is concluded that the EM method is the most suitable completion method. Homogeneity analyses were performed in the Mathematica programming language to identify possible changes and inconsistencies in the completed rainfall dataset. Homogeneity analyses revealed that most of the completed rainfall datasets are homogeneous at class 1 level, consistent and reliable and do not show systematic changes in time.

Comparison of the missing rainfall data method in enhancing discharge estimation

Abstract Missing rainfall data can cause the results of an analysis to be inaccurate. This study aims to compare the performance of different methods in estimating missing rainfall data and to evaluate the estimated missing rainfall data method on hydrograph estimation. The methods used to estimate missing rainfall data in three selected stations include the arithmetic mean (AM) method, the inverse squared distance (ISD) method, and replacing missing value with zero value (ZERO). The study employs root mean square error (RMSE) to assess the accuracy and reliability of the methods. The estimated rainfall data are then utilized in Stormwater Management Modelling Software 5 (SWMM5) to evaluate the data impact on the discharge estimation. The results indicated that the AM method is the best method to estimate missing rainfall data by considering the lowest RMSE value equal to 1.185. In conclusion, the comparison of missing rainfall data methods in this study has shed light on their performance in enhancing discharge estimation.

Correction to: Multilayer perceptron-based predictive model using wavelet transform for the reconstruction of missing rainfall data

Correction to: Multilayer perceptron-based predictive model using wavelet transform for the reconstruction of missing rainfall data

Improving prediction of missing rainfall data by identifying best Artificial Neural Network model

Improving prediction of missing rainfall data by identifying best Artificial Neural Network model

Comparative evaluation of techniques for missing rainfall data estimation in arid regions: case study of Al-Madinah Al-Munawarah, Saudi Arabia

Comparative evaluation of techniques for missing rainfall data estimation in arid regions: case study of Al-Madinah Al-Munawarah, Saudi Arabia

A Comparison of Different Methods for Rainfall Imputation: A Galician Case Study

With the ultimate goal of developing models that involve the use of environmental variables, a GIS-based application is being developed that is circumscribed to the region of Galicia (Spain). Ten-minute data of six meteorological variables were collected from 150 stations of the MeteoGalicia network over a period of 18 years, but the time series data are not complete. In order to estimate missing rainfall data, four imputation methods were evaluated in this study: missForest, MICE, Amelia II, and inverse distance weighting (IDW). Crossvalidation results show that the precipitation is out of phase in the different stations due to their geographical locations, and the imputation can be improved with a displacement of the time series; on the other hand, the missForest method provided better results in the imputation of this meteorological variable than the MICE, Amelia, or IDW.

Comparative assessment of univariate and multivariate imputation models for varying lengths of missing rainfall data in a humid tropical region: a case study of Kozhikode, Kerala, India

Comparative assessment of univariate and multivariate imputation models for varying lengths of missing rainfall data in a humid tropical region: a case study of Kozhikode, Kerala, India

Multilayer perceptron-based predictive model using wavelet transform for the reconstruction of missing rainfall data

Multilayer perceptron-based predictive model using wavelet transform for the reconstruction of missing rainfall data

ANN in forecasting Missing Rainfall Data

ANN has been used to estimate rainfall data by analysing patterns and mapping the correlation between historical data and weather patterns. ANNs are a type of machine learning algorithm that is modelled on the structure of the human brain, which makes them particularly effective for solving complex problems that involve large amounts of data. Overall, the use of ANNs for estimating rainfall data is a promising area of research that has the potential to provide valuable insights into weather patterns and their impacts on the environment and human society. The potential of ANN in the estimation of missing precipitation values has been investigated in detail. This study proposes to compare the performance of conventional methods as well as different algorithms of the Artificial Neural Network method to predict missing rainfall in the Upper Tapi catchment area in the West region of India. It was found that the ANN method has an edge over the conventional methods and proved to be a better method of finding the missing rainfall data values.

Imputation of missing monthly rainfall data using machine learning and spatial interpolation approaches in Thale Sap Songkhla River Basin, Thailand.

Missing rainfall data has been a prevalent issue and primarily interested in hydrology and meteorology. This research aimed to examine the capability of machine learning (ML) and spatial interpolation (SI) methods to estimate missing monthly rainfall data. Six ML algorithms (i.e. multiple linear regression (MLR), M5 model tree (M5), random forest (RF), support vector regression (SVR), multilayer perceptron (MLP), genetic programming (GP)) and four SI methods (i.e. arithmetic average (AA), inverse distance weighting (IDW), correlation coefficient weighted (CCW), normal ratio (NR)) were investigated and compared in their performance. The twelve rainfall stations, located in the Thale Sap Songkhla river basin and nearby basins, were considered as a study case. Tuning hyper-parameters for each ML method was conducted to get the most suitable model for the data sets considered. Three performance criteria matrices (i.e. NSE, OI, and r) were chosen, and the sum of those three performance criteria matrices was introduced for methods' performance comparison. The experimental results pointed out that selecting neighbouring stations were essential when applying SI methods, but not for the ML method. The overall performance showed ML better imputed missing monthly rainfall than SI due to overcoming spatial constraints. GP provided the highest performance by giving NSE = 0.825, OI = 0.877, and r = 0.909 for the training stage. Those values for the testing stage were 0.796, 0.852, and 0.902, respectively. It was followed by SVR-rbf, SVR-poly, and RF. NR provided the best performance among four SI methods, followed by CCW, AA, and IDW. When applying SI methods, it should contemplate a correlation between the target and neighbouring stations greater than 0.80.

Analisis Debit Banjir Rancangan dengan Data Curah Hujan di DAS Waru Kabupaten Banggai

The meteorological data is important to design and planning the water resources system and irrigation. The complete and long of data is needed to make sure the accuracy of the water avaibility. Unfortunately, the avaibility of the rainfall data is not always complete. So it’s needed the methods to fill the lacking rainfall data. The location of the research is Waru watershed at the Banggai Regency, Central Sulawesi. The cause of repeated flooding in 2019 and 2020 was the disability of the watershed to accommodate higher rainfall intensity. The goals of the research was to compare the design flood discharge values obtained, using the lacking of rainfall data and rainfall data that had been filled in. In filling in the missing rainfall data, 2 methods are used, that are the Algebraic Average method and the Inverse Square Distance method. Then the calculation continues with the design flood discharge, using the Rational method. The final result is the design flood discharge values from the two rainfall data with different conditions will be compared to see the difference between the both of the methods. The value of the difference in design flood discharge obtained from this comparison is 0.0552 m³/s for a 2 year return period, 37.0313 m³/s for a 5 year return period, 79.3389 m³/s for a 10 year return period, 163.1544 m³/s for a 25 year return period, 255.1732 m³/s for the 50-year return period and 380.8701 m³/s for the 100-year return period.

A COMPARATIVE STUDY OF MISSING RAINFALL DATA ANALYSIS USING THE METHODS OF INVERSED SQUARE DISTANCE AND ARITHMETIC MEAN

In water resources planning and management, it is essential to have reliable rainfall data. In many cases, rainfall data under the guardian national/ local institution are incomplete. Some data are missing, both monthly and annually. The missing data may persist due to neither damage nor human error. This study aims to estimate the missing rainfall data using two methods, i.e., the inverse square distance and the arithmetic mean methods. The study compared the two mentioned methods using root mean square error (RMSE) and mean absolute error (MAE) and to determine the consistency of rainfall data in all stations using double mass curve analysis. This study utilized the rainfall data from Tepus, Semanu, Rongkop, and Tanjungsari Stations in Gunung Kidul Regency, Yogyakarta Province, Indonesia. The model performance was tested by the root mean square error (RMSE) and mean absolute error (MAE). The rainfall data consistency was determined by double mass curve analysis. The results showed that the arithmetic mean method performed better rather than the inverse square distance method. The smallest RMSE and MAE values in the arithmetic method at the four stations have confirmed the statement. The rainfall data consistency analyzed by the double mass curve is consistent in all stations except Tepus Station.

Comparison of Normal Ratio Method and Distance Power Method for Estimating Missing Rainfall Data with Three Neighboring Stations

This study compared the performances of Normal Ratio Method and Distance Power Method as a tool for estimating missing rainfall data. The data utilized are the rainfall data of the three neighboring station of Catarman, Northern Samar, Philippines. These stations are Catbalogan Station (Samar Province), Legazpi (Bicol Province) and Masbate (Masbate Province).
 The observed daily rainfall data for the Catarman (Northern Samar), Catbalogan, Legazpi, and Masbate were obtained from the Philippine Atmospheric Geographical Astronomical Services Administration. The monthly rainfall were computed for the three (3) neighboring stations (Catbalogan, Legazpi, Masbate).
 The evaluation used the T-test for correlated samples and the Pearson’s Correlation Coefficient for the monthly rainfall data computed of the three neighboring Station of Catarman, Northern Samar with the three neighboring stations. Based from the results, Normal Ratio Method performs better than Distance Power Method as applied to three neighboring stations.

Spatial Analysis of Climate variability and Change in the Great Ethiopian Rift Valley Basins

The generally high temporal and spatial climate variability and change in most parts of Ethiopia, where rainfed farming is the main form of crop production, has been the main cause of food insecurity in significant areas of the country. Spatial variations in selected climate variables were investigated for the great Rift Valley regions of Ethiopia during the baseline (1981-2010) and projected (2021-2100) periods. Baseline climate data from 16 stations that represent different agroecology were obtained from National Meteorology Agency of Ethiopia (NMA) ( http://www.ethiomet.gov.et ). Corresponding projected data grided over 0.5ᵒ X 0.5ᵒ were retrieved from eight GCM-RCM combinations under two Representative concentration pathways (RCPs) from CORDEX database ( http://www.cordex.org ). First order Markov chain model was used for missing rainfall data filling. Coefficient of variation and standardized anomaly index descriptors were used in the analysis of climate data for each station. Spatial maps were then generated from station values using the ordinary kriging method of interpolation. The result indicated that rainfall of the study basins showed both spatial and temporal variabilities. The total annual rainfall showed variation in the year-to-year variability ranging from low in the southern half to high in the northern half of the basins. Seasonal rainfall showed high to very high variability which is challenging to rainfed agriculture. For the projected periods, majority of the climate models projected a decline in annual rainfall and increase in temperature. HadGEM2-ES_RCA4 model simulation suggested precipitation change varying from +4.2 to -16% and +3.8 to -18% for near period under RCP4.5 and RCP8.5 emission scenarios, respectively. Mean temperature is projected to rise from +0.7 to +1.25 ºC under RCP4.5 to +0.9 to +1.6 ºC under RCP8.5 across the GRVB in the near future and further warming was projected in the mid and end centuries. Rainfed crop production in the region, which is already impacted by the current climate variability, is likely to be further challenged with future climate change. As a consequence, specific impact -based adaptation strategies are essential to reduce the vulnerability of rainfed crop production in the area. Keywords : Climate models, growing season, Rift Valley, seasonal water deficit, dry spell DOI: 10.7176/JEES/11-16-01 Publication date: June 30 th 2021

Long-Term Rainfall Information Forecast by Utilizing Constrained Amount of Observation through Artificial Neural Network Approach

Estimating models are becoming increasingly crucial in highlighting the nonlinear connections of the massive level of rough information and chaotic components. The study demonstrates a modern approach utilizing a created artificial neural network (ANN) method that may be an alternative strategy to conventional factual procedures for advancing rainfall estimate execution. A case study was presented for Turkey’s Düzce and Bolu neighboring territories located on the Black Sea’s southern coast. This study’s primary aim is to create an ANN model unique in the field to generate satisfactory results even with limited data. The proposed technique is being used to estimate rainfall and make predictions regarding future precipitation. Bolu daily average rainfall by month data and a limited number of Düzce rainfall data were used. Missing forecasts and potential rainfall projections will be examined in the fundamental research. This research further focuses on ANN computational concepts and develops a neural network for rainfall time series forecasting. The emphasis of this study was a feed-forward backpropagation network. The Levenberg–Marquardt algorithm (LMA) was implemented for training a two-layer feed-forward ANN for the missing rainfall data prediction part of this research. The inaccessible rainfall parameters for Düzce were determined for the years 1995 to 2009. From 2010 to 2020, a two-layer feed-forward ANN was trained using the gradient descent algorithm to forecast daily average rainfall data by month. The findings reported in this study guide researchers interested in implementing the ANN forecast model for an extended period of missing rainfall data.

Comparison of Missing Rainfall Data Treatment Analysis at Kenyir Lake

Rainfall is one of the frequent data used in weather-related studies. Sometimes the data have missing information that needs the treatment to make sure the data can be useful, complete and reliable. There are many methods in treating missing data suggested by previous studies. The best selected method to estimate missing rainfall data in different regions may vary depending on the rainfall pattern and spatial distribution. Therefore, this paper discussed and compared 3 different methods in missing data treatment. The selected methods are Expectation Maximization (EM), Inverse Distance Weighted (IDW) and Multiple Imputation (MI). After analysis, the best method is IDW based on root mean square error (RMSE), mean absolute error (MAE), correlation coefficient (r) and percentage of error (% of error) values. The IDW method has RMSE, MAE values and the lowest % of error values. In addition, the r value of IDW method is highest compared to EM and MI method. MI method recorded the highest values of RMSE, MAE and % of error with the lowest r value that proved MI method is the least accurate method to use in missing data treatment. After all methods were implemented, it proved that the IDW method is the best way to treat missing data because the analysis shows monthly rainfall distribution for 4 treatment stations in line to 3 missing data stations compared to EM and MI methods.

Spatial Interpolation for Missing Rainfall Data in Northern Region of Peninsular Malaysia

This study is aimed to estimate missing rainfall values for daily rainfall data from 30 selected rainfall stations. The daily rainfall data were obtained from the Department of Irrigation and Drainage Malaysia (DID) for the periods of 1999 to 2019. The missing values throughout the 20 years period were estimated using spatial interpolation methods. These methods include arithmetic average (AA), normal ratio (NR), inverse distance (ID) and coefficient of correlation (CC) weighting methods. The methods consider the distance between the target and the neighbourhood stations as well as the correlation between them. In determining the best spatial interpolation method, three tests for evaluating model performance have been used namely similarity index (S-index), mean absolute error (MAE) and root mean square error (RMSE). The homogeneity test using Standard normal homogeneity (SNHT), Buishand range (BR), Pettitt and Von Neumann (VNR) ratio are conducted to test the homogeneity of the rainfall data. The results show that the ID method is more efficient than the others method and 85% of the rainfall stations were homogenous based on this method. This study is important as it can be used to fill in the missing value rainfall data so that the conclusions that can be drawn from the data is valid.

Comparison of Methods to Estimate Missing Rainfall Data for Short Term Period at UMP Gambang

Hydrological analysis is crucial for planning or designing engineering structures and water resources system as stated by World Meteorological Organization (WMO). A long and complete data set of rainfall is required in order to carry out the hydrological study successfully. However, the most common problem that always come up in data collection is the lacking of data due to some occurrence involving the equipment installed that may cause the error to occur. For this study, another problem that has been a concern is the availability of the rainfall data is insufficient as the studied area is newly developed so the limitation of the existing data may affect to the relevance of this study. Therefore, this study is mainly conducted to compare three existing methods from previous scholars which are Arithmetic Mean Method, Normal Ratio Method and Inverse Distance Weighting Method to prove whether these methods are capable to fill the gaps of rainfall data for short term period of study. The existing data collected at the target station are then compared with the three methods whether the results gained are similar or otherwise. Analysis shows that the study for short term period is irrelevant due to the insufficiency of data collection.

Effect of Missing Rainfall Data on the Uncertainty of Design Floods

Hydrological data is a collection of information or facts about the hydrological phenomenon. Mistakes in the hydrological analysis have a fatal effect on the accuracy or accuracy of hydraulic building planning, whether in the form of planning for irrigation buildings, flood control structures, drainage systems, weirs. And others hydraulics structure design. As we already know, rainfall data is the most important data in the hydrological analysis. The minimum amount of rainfall data required in the hydrological statistical analysis is 10 years. To obtain this 10-year data, often encountered incompleteness in recording the rainfall data. Meanwhile, hydrological analysis always required for any design of hydraulic structures. Based on the above statement, an analysis has been carried out on how much influence the completeness of data and availability of rainfall data in an area of accuracy in predicting the discharge plan in different times and make a Rating Curve. In determining the unit hydrograph, the Nakayasu method is used. The results of this study indicate that the completeness of the data is not very influential. This is evidenced by obtaining the discharge value at a certain return period which is not much different. From the unit hydrograph curve, it can be seen that in the first condition with a return period of 100 years, a discharge of 801.775 m3 / s is obtained, for the second condition consisting of 11 rainfall data, a discharge of 833.848 m3 / s is obtained, and for the third condition, ignoring the rainfall data. The missing of rainfall data obtained a discharge of 816,489 m3 / s. Where the difference inaccuracy in the first condition is 3.846% against the second condition, and the difference inaccuracy in the third condition is 2.082% for the second condition. In the rating curve section of Batang Kuranji, the normal water level discharge is 169.948 m3 / s with an elevation of 2.89 m. At the same time, the maximum discharge at the cross-section of Batang Kuranji is 748,260 m3 / s with an elevation of 6.99 m.

Simulation Study of Adjusted Spatial Weighting Method to Estimate Missing Rainfall Data

Missing value especially in environmental study is a common problem including in rainfall modelling. Incomplete data will affect the accuracy and efficiency in any modelling process. In this study, simulation method is used to demonstrate the efficiency of the old normal ratio inverse distance correlation weighting method (ONRIDCWM) in solving missing rainfall data. The simulation study is used to identify the best parameters for correlation power of p, percentage of missing value and sample size, n of the ONRIDCWM through simulating for 10,000 times by varying the value of the parameters systematically. The results of the simulation are compared with other available weighting methods. The estimated complete rainfall data of the target station are compared and assessed with the observed data from the neighbouring station using mean, estimated bias (EB) and estimated root mean square error (ERMSE). The results show that ONRIDCWM is better than the other weighting methods for the correlation power of p at least four. For illustration of the weighting method, monthly rainfall data from Pahang is used to demonstrate the efficiency of the method using three error indices: S-Index, mean absolute error (MAE) and correlation, R.