Evaluation of Different Missing Data Reconstruction Methods for Daily Minimum Temperature in Elevated Stations of Iran: Comparison with New Proposed Approach

Document Type : Research Paper

Authors

1 Ph.D.Student/ Univ.ofTehran

2 Professor/University of Tehran

3 َAssociate Professor /Univ. of Tehran

Abstract

Daily minimum air temperature data are greatly needed in climatic studies of first-fall and last-spring frosts, frost periods, evaluation and improvement of crop production potentials, and eventually their effects upon food security. Despite the fact that climate stations, set up at high elevations play important roles in accurate estimate of temperature parameter gradients, and on their mappings, the number of such established stations in Iran is limited, causing many gaps to be served in their data time series. Hence, reconstruction of temperature data for elevated stations is considered to be essential, especially for studies requiring long-term homogeneous data items. This study was aimed at making a comparison of the different methods of readjustment of the daily minimum temperature data (obtained from highly elevated stations) and to determine the most suitable method for readjusting and lengthening of their record periods. To follow the purpose, a number of 12 stations at elevations exceeding 1900 m were selected. A number of 500 randomly sampled (minimum daily temperature) data were taken and reconstructed through 31 classic methods, and as well, through a new proposed approach, based on Cumulative Distribution Function (CDF) of minimum temperature data. Accuracies of these methods were tested using RMSE within 90 and 95 % of confidence interval of errors. Results revealed that Principle Component Analysis, proposed method based on CDF, and Artificial Neural Network stood in priority for reconstruction of daily minimum temperature data, with 95% of confidence intervals, reconstructed error of ±2.0, ±2.2 and ± 3.1 °c, respectively. This study led to completion of daily minimum temperature data series of highly elevated stations for the period of 1965-2010. This can be employed in climate change studies and as well in first-fall vs. last-spring frost risks, and reform of farming calendar depending upon climate change.

Keywords

Main Subjects

References

Ashraf, M., Loftis, J. C., & Hubbard, K. G. (1997). Application of geostatistics to evaluate partial weather station networks. Agricultural and forest meteorology, 84(3), 255-271.
Carrega, P. (1995). A method for the reconstruction of mountain air temperatures with automatic cartographic applications. Theoretical and applied climatology, 52(1-2), 69-84.
Coulibaly, P., & Evora, N. D. (2007). Comparison of neural network methods for infilling missing daily weather records. Journal of hydrology, 341(1), 27-41.
Demyanov, V., Kanevsky, M., Chernov, S., Savelieva, E., & Timonin, V. (1998). Neural network residual kriging application for climatic data. Journal of Geographic Information and Decision Analysis, 2(2), 215-232.
Di Piazza, A., Conti, F. L., Noto, L. V., Viola, F., & La Loggia, G. (2011). Comparative analysis of different techniques for spatial interpolation of rainfall data to create a serially complete monthly time series of precipitation for Sicily, Italy. International Journal of Applied Earth Observation and Geoinformation, 13(3), 396-408.
Khalil, M., Panu, U. S., & Lennox, W. C. (2001). Groups and neural networks based streamflow data infilling procedures.  Journal of Hydrology, 241(3), 153-176.
Khalili A (1991) Integrated Water Plan of Iran, Jamab Consulting Engineering Co., The Ministry of Energy, Tehran, 111-122. (In Farsi)
Henn, B., Raleigh, M. S., Fisher, A., & Lundquist, J. D. (2013). A comparison of methods for filling gaps in hourly near-surface air temperature data. Journal of Hydrometeorology, 14(3), 929-945.
Khorshiddoust, A. M., Nassaji, Z. M., and Ghermez, C. B. (2012). Time Series Reconstruction of Daily Maximum and Minimum Temperature using Nearest Neighborhood and Artificial Neural Network Techniques (Case Study: West of Tehran Province). Geographical Space, 12 (38), 197-214. (In Farsi)
Kim, J. W., & Pachepsky, Y. A. (2010). Reconstructing missing daily precipitation data using regression trees and artificial neural networks for SWAT streamflow simulation. Journal of hydrology, 394(3), 305-314.
Mileva-Boshkoska, B., & Stankovski, M. (2007). Prediction of missing data for ozone concentrations using support vector machines and radial basis neural networks. Informatica, 31(4).
Mwale, F. D., Adeloye, A. J., & Rustum, R. (2012). Infilling of missing rainfall and streamflow data in the Shire River basin, Malawi–A self organizing map approach. Physics and Chemistry of the Earth, Parts A/B/C, 50, 34-43.
Teegavarapu, R. S., & Chandramouli, V. (2005). Improved weighting methods, deterministic and stochastic data-driven models for estimation of missing precipitation records. Journal of Hydrology, 312(1), 191-206.
Wagner, P. D., Fiener, P., Wilken, F., Kumar, S., & Schneider, K. (2012). Comparison and evaluation of spatial interpolation schemes for daily rainfall in data scarce regions. Journal of Hydrology, 464, 388-400.
Xia, Y., Fabian, P., Stohl, A., & Winterhalter, M. (1999). Forest climatology: estimation of missing values for Bavaria, Germany. Agricultural and Forest Meteorology, 96(1), 131-144.
Yozgatligil, C., Aslan, S., Iyigun, C., & Batmaz, I. (2013). Comparison of missing value imputation methods in time series: the case of Turkish meteorological data. Theoretical and applied climatology, 112(1-2), 143-167.
You, J., Hubbard, K. G., & Goddard, S. (2008). Comparison of methods for spatially estimating station temperatures in a quality control system. International Journal of Climatology, 28(6), 777-787.
 
 
 
Iranian Journal of Soil and Water Research
Volume 48, Issue 2
August 2017
Pages 231-239

Files

Share

How to cite

Statistics