Evaluation of imputation techniques for infilling missing daily rainfall records on river basins in Ghana

Addi, Martin; Gyasi‐Agyei, Yeboah; Obuobie, Emmanuel; Amekudzi, Leonard K.

doi:10.1080/02626667.2022.2030868

Cited by 21 publications

(4 citation statements)

References 60 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For the methodology of eliminating the outlier data, Kulanuwat et al’s ( 2021 ), Azman et al’s ( 2021 ), and Baddoo et al’s ( 2021 ) study were reviewed with the aim of establishing a methodology that allows verification with statistical tests to determine the error and likewise establish the best method. And finally, the methodology of Chiu et al ( 2021 ) and Addi et al ( 2022 ) for the analysis of results under statistical models such as root mean square error and mean absolute error (Duarte et al . 2022 ) presents similar statistical indicators.…”

Section: Methodsmentioning

confidence: 99%

Time series outlier removal and imputing methods based on Colombian weather stations data

Parra-Plazas

Gaona-García

Plazas-Nossa

2023

Environ Sci Pollut Res

View full text Add to dashboard Cite

The time data series of weather stations are a source of information for floods. The study of the previous wintertime series allows knowing the behavior of the variables and the result that will be applied to analysis and simulation models that feed variables such as flow and level of a study area. One of the most common problems is the acquisition and transmission of data from weather stations due to atypical values and lost data; this generates difficulties in the simulation process. Consequently, it is necessary to propose a numerical strategy to solve this problem. The data source for this study is a real database where these problems are presented with different variables of weather. This study is based on comparing three methods of time series analysis to evaluate a multivariable process offline. For the development of the study, we applied a method based on the discrete Fourier transform (DFT), and we contrasted it with methods such as the average and linear regression without uncertainty parameters to complete missing data. The proposed methodology entails statistical values, outlier detection, and the application of the DFT. The application of DFT allows the time series completion, based on its ability to manage various gap sizes and replace missing values. In sum, DFT led to low error percentages for all the time series (1% average). This percentage reflects what would have likely been the shape or pattern of the time series behavior in the absence of misleading outliers and missing data.

show abstract

Section: Methodsmentioning

confidence: 99%

Time series outlier removal and imputing methods based on Colombian weather stations data

Parra-Plazas

Gaona-García

Plazas-Nossa

2023

Environ Sci Pollut Res

View full text Add to dashboard Cite

show abstract

“…For the methodology of eliminating the outlier data, (Kulanuwat et al 2021), (Azman et al 2021) and (Baddoo et al 2021), were reviewed with the aim of establishing a methodology that allows verification with statistical tests to determine the error and likewise establish the best method. And finally the methodology of (Chiu et al 2021) and (Addi et al 2022) for the analysis of results under statistical models such as root mean square error, mean absolute error, such as (Duarte et al 2022) that presents similar statistical indicators Finally, it was evaluated comparatively, therefore, the best behavior among the three methods was defined for its final implementation and thus completing the database for the next phase of the forecast project under a standard.…”

Section: A Methodologymentioning

confidence: 99%

Time Series Outlier Removal and Imputing Methods Based on Colombian Weather Stations Data

Parra-Plazas

Gaona-García

Plazas-Nossa

2022

Preprint

View full text Add to dashboard Cite

The time data series of weather stations are a source of information for floods. The study of the previous wintertime series allows knowing the behavior of the variables and the result that will be applied to analysis and simulation models that feed variables such as flow and level of a study area. One of the most common problems is the acquisition and transmission of data from weather stations due to atypical values and lost data, this generates difficulties in the simulation process. Consequently, it is necessary to propose a numerical strategy to solve this problem. The data source for this study is a real database where these problems are presented with different variables of weather. This study is based on comparing three methods of time series analysis to evaluate a multivariable process offline. For the development of the study, we applied a method based on the Discrete Fourier Transform (DFT) and we contrasted it with methods such as the average and linear regression without uncertainty parameters to complete missing data. The proposed methodology entails statistical values, outlier detection and the application of the DFT. The application of DFT allows the time series completion, based on its ability to manage various gap sizes and replace missing values. In sum, DFT led to low error percentages for all the time series (1% average). This percentage reflects what would have likely been the shape or pattern of the time series behavior in the absence of misleading outliers and missing data.

show abstract

“…The sparsity of rain gauges is a common problem worldwide, making modelling of complex rainfields with high spatial and temporal variabilities a major challenge 1 , 2 . Moreover, it is rare to find rain gauges with a complete record without missing values due to factors such as equipment malfunction and accessibility 3 – 5 .…”

Section: Introductionmentioning

confidence: 99%

Optimal selection of daily satellite precipitation product based on structural similarity index at 1 km resolution for the Pra catchment, Ghana

Gyasi-Agyei,

Obuobie,

et al. 2023

Sci Rep

Self Cite

View full text Add to dashboard Cite

Thirteen satellite precipitation products (SPPs), re-gridded to 1 km resolution, were evaluated in terms of the structural similarity index (SSI) over the Pra catchment in Ghana. Three SPP scenarios were considered: Scenario one (S1) was the original SPPs; Scenario two (S2) was bias-corrected SPPs; and Scenario three (S3) was the better of S1 and S2 for each wet day. For each scenario, the best SPP was selected to constitute the 14th SPP referred to as the BEST SPP. Each SPP was evaluated in terms of SSI against the rain gauge rainfield for each wet day. For S1, the top three SPPs were TMPA, GSMAP and CMORPH; for S2, CMORPH, PERCCS and MSWEP were the top three; and for S3, CMORPH, PERCCS and TMPA came out on top in order of decreasing performance. Bias correction led to improvement in the overall SSI measure (SSIM) for 73% of wet days. The BEST SPP increased the SSIM of the best individual SPP by over 50% for S1, and over 30% for both S2 and S3. Comparing the BEST SPP of the three scenarios, S2 increased the SSIM statistic by 20% over that for S1, and SSIM was further improved by 4% for S3. It is highly recommended to use BEST SPP (S3) to generate the required 1 km × 1 km rainfields for the Pra, or other catchments around the world with a sparse rain gauge network, through conditional merging with rain gauge data as demonstrated.

show abstract

Evaluation of imputation techniques for infilling missing daily rainfall records on river basins in Ghana

Cited by 21 publications

References 60 publications

Time series outlier removal and imputing methods based on Colombian weather stations data

Time series outlier removal and imputing methods based on Colombian weather stations data

Time Series Outlier Removal and Imputing Methods Based on Colombian Weather Stations Data

Optimal selection of daily satellite precipitation product based on structural similarity index at 1 km resolution for the Pra catchment, Ghana

Contact Info

Product

Resources

About