Accuracy of Various Interpolation Methods in Estimating Rainfall Values to Select the Most Optimal Algorithm (Case Study: Kurdistan Province)

Document Type : Full length article

Authors

1 MA in Remote Sensing, University of Tabriz, Tabriz, Iran

2 Associate professor of Geomorphology, University of Tabriz, Tabriz, Iran

Abstract

Introduction
Continuous spatial data of environmental variables are often required for environmental sciences and management. However, information for environmental variables is usually collected by point sampling. Thus, the methods generating such spatially continuous data from the point samples become essential tools for many environmental analyses. Spatial interpolation is the procedure of estimating the value of un-sampled points using existing observations. The methods for spatial interpolation can be classified into two main categories as deterministic and geostatistical. Deterministic interpolation techniques including Inverse Distance Weighting (IDW), Radial Basis Function (RBF), and so on calculate the values of un-sampled areas based on the known values of the sampled points and create surfaces from measured points. However, Geostatistical interpolation techniques, e.g. Kriging use statistical properties of the measured points to quantify the spatial autocorrelation among the measured points and account for the spatial configuration of the sample points around the estimation location.
Materials and methods
In this study, to assess the accuracy of the various interpolation methods to estimate the rainfall distribution of Kurdistan Province, we have used data from rain gauge stations, and synoptic and climatology data. After reviewing the statistical situation of the stations, statistical period of 2001-2013 has been selected. Among all stations in the basin, the stations which had 12 years of full or recyclable statistics until 2013 were selected for the study. It must be noted that the selection of the stations was according to their statistics rebuilt by the application of the highest correlation with the adjacent stations. Finally, normality of data quality and data sets were recorded and evaluated using Komogorov-Smirnov and Chi square X2 statistical tests. In addition, we have used the digital elevation model data collected by the SRTM satellite sensors with spatial resolution of 30 m. We have also employed analytic functions of the ArcGIS 10.2.2, Surfer 11, and IBM SPSS Statistics 22. After reviewing data on existing stations (77 Rain gauge stations, 22 synoptic and climatology stations of the Meteorology Organization, and 76 Rain gauge stations and evaporation stations of the Department of Energy), we have rebuilt the missing rainfall data and selected the minimum length of the statistical common period for less than 10 years. The normality of data has been evaluated to select 145 stations for the analysis of the interpolation methods and choosing the best method. The methods used in this study were IDW, Spline (with Tension, Thin Plate and Completely Regular functions), Ordinary Kriging (with Circular, Spherical, Exponential, and Gussian functions), and Universal Kriging (with Rational Quadratic and Liner functions).
Results and discussion
In order to conduct this statistical analysis, the data should be normal. Therefore, the average annual rainfall values of the Kurdistan Province were tested for normality. In this research, Komogorov - Smirnov and Chi square X2 tests in SPSS have revealed that the data have the normal distribution. In order to analyze the accuracy of the various interpolation methods, the models were implemented using the ArcGIS application. By applying each of these models on rainfall data, the maps were obtained. In order to evaluate and determine the most optimal model, the validity and the accuracy of the maps were evaluated. As it mentioned in the previous section, the more the Mean Absolute Error (MAE) and the closer Mean Bias Error (MBE) to zero, the accuracy of the model is higher. On the other hand, the less the Root Mean Square Error (RMSE) and the higher the correlation coefficient (R2), the less is the model error is. The results also show the error rate of implementation of the interpolation methods. Based on the findings of the study, the lowest error is related to the Ordinary Kriging Method with the circular function; and after that, it is related to the General Kriging Method with the Quadratic Variogram. In general, the Kriging method provides results with higher accuracy than those of other methods.
Conclusion
In this study, the models of the deterministic and geostatistical methods were compared with each other to find the most suitable spatial interpolation method in Kurdistan Province. To compare the actual results, the same conditions were used to assess the accuracy. Then, the most important methods of the validity were extracted and identified: Mean Absolute Error (MAE), Mean Bias Error (MBE), Root Mean Square Error (RMSE) and correlation coefficient (R2). Ordinary Kriging Method of interpolation with the circular function had the highest accuracy compared with the other methods. One of the most important factors to achieve high accuracy in this method is its ability to depict the non-bias linear estimation. However, other methods, especially the Universal Kriging with Quadratic function, due to the use of local procedures offers an acceptable accuracy.

Keywords

Main Subjects

References

تازه، م.؛ کـوثری، م.ر.؛ بخـشایی، م. و خـسروی، ی. (1387). پهنـه‏بنـدی خشکی براساس نمایة ترانسو با استفاده از زمین‏آمار و GIS (مطالعة مـوردی: بخـش غربـی استان اصـفهان)،کنفرانس بین‏المللی گیاه‏شناسی درختی و تغییر اقلیم در اکوسیستم‏های خزری، ساری، پژوهشکدة اکوسیستم‏های خزری، ص 1-12.
ثقفیان، ب.؛ دانش‏کار آراسته، پ.؛ رحیمی بندرآبادی، س.؛ فتاحی، ا. و محمدزاده، م. (1391). راهنمای روش‏های توزیع مکانی عوامل اقلیمی با استفاده از داده‏های نقطه‏ای، معاونت برنامه‏ریزی و نظارت راهبردی رئیس‏جمهور، 585: 1-133.
جلالی، ق.؛ طهرانی، م.؛ برومند، ن. و سنجری، ص. (1392). مقایسة روش‏های زمین‏آمار در تهیة نقشة پراکنش مکانی برخی عناصر غذایی در شرق استان مازندران، فصل‏نامة پژوهش‏های خاک (علوم خاک و آب)، 27: 196-204.
رحیمی بندرآبادی، س. و مهدیان، م.ح. (1382). بررسی تغییرات مکانی بارنـدگی ماهانه در مناطق خشک و نیمه‏خشک جنوب شرق ایران، سـومین کنفـرانس منطقـه‌ای و اولین کنفرانس ملی تغییر اقلیم، 29 مهر تا اول آبان، دانشگاه اصفهان. ص 1-8.
رستمی، م.؛ سالجقه، ع.؛ صانعی، م. و مهدوی، م. (1393). ارزیابی برخی روش‏های درون‏یابی در مطالعة مورفولوژی بستر رودخانه‏‏ها و کانال‏های آبی، مجلة علوم و مهندسی آبخیزداری ایران، 8(25): 57-67.
رنگزن، ک.؛ مختاری، م. و شایگان، م. (1384). ارزیابی دقت مدل‏های IDW و Kriging جهت درون‏یابی داده‏های سطح آب زیرزمینی دشت میان آب شوشتر، همایش ژئوماتیک 1384، ص 1-15.
سلطانی، س. و مدرس، ر. (1385). تحلیل فراوانی و شـدت خـشک‏سـالی هواشناسـی استان اصفهان، مجلة منابع طبیعی ایران، 1: 15-26.
شمس‏الدینی، ع. (1379). تغییرات منطقه‏ای بارندگی با استفاده از روش کریجینگ در استان‏های شمالی، پایان‏نامة کارشناسی ارشد آبیاری و زهکشی، دانشکدة کشاورزی، دانشگاه شیراز.
صفری، ه. (1386). مقایسة دو روش درون‏یابی Kriging و IDW، مجلة شهرنگار، 7(40): 33-39.
صفری، م. (1381). تعیین شبکة بهینة اندازه‏گیـری سـطح آب زیرزمینـی بـا کمـک روش‏های زمین‏آماری، مطالعة موردی: دشت چمچال، پایان‏نامة کارشناسی ارشد آبیاری و زهکشی، دانشکدة کشاورزی، دانشگاه تربیت مدرس.
صفرراد، ط.؛ فرجی سبکبار، ح.؛ عزیزی، ق. و عباسپور، ر.ع. (1391). تحلیل مکانی تغییرات بارش در زاگرس میانی از طریق روش‏های زمین‏آمار (1995ـ2004)، جغرافیا و توسعه، 31: 149-164.
عباسی جندانی، ش. و ملکی‏نژاد، ح. (1393). ارزیابی روش‏های زمین‏آماری مبتنی بر GIS برای پهنه‏بندی مکانی بارش (مطالعة موردی: دشت کوهپایه- سگزی)، دومین همایش ملی بیابان با رویکرد مدیریت مناطق خشک و کویری، دانشکدة کویرشناسی دانشگاه سمنان، ص 1-10.
فرجی سبکبار، ح.ع. و عزیزی، ق. (1385). ارزیابی میزان دقت روش‏های درون‏یابی فضایی مطالعة موردی: الگوسازی بارندگی حوضه کارده مشهد، فصل‏نامة پژوهش‏های جغرافیایی، 58: 1-15.
گل‏محمدی، گ.؛ معروفی، ص. و محمدی، ک. (1386). منطقه‏ای‏نمودن ضریب رواناب در استان همدان با استفاده از روش‏های زمین‏آماری و GIS، فصل‏نامة علوم و فنون کشاورزی و منابع طبیعی، 12(46): 501-514.
مسعودیان، س.ا. (1382). تحلیل ساختار دمای ماهانة ایران، مجلة پژوهشی علوم‏انسانی دانشگاه اصفهان، 15: 87-96.
محمودی، پ. و علیجانی، ب. (1392). مدل‏بندی رابطة بارش‏های سالانه و فصلی با عوامل زمین‏اقلیم در کردستان، نشریة تحقیقات کاربردی علوم جغرافیایی، 13(31): 93-112.
محمدی، ب. (1391). تحلیل روند بارش استان کردستان، پانزدهمین کنفرانس ژئوفیزیک ایران، ص 1-11.
محمدی، ج. (1385). پدومتری (آمار مکانی)، تهران: انتشارات پلک.
مهدی‏زاده، م. (1381). ارزیابی زمین‏آماری برای برآورد دما و بارنـدگی در حوضـة آبریز دریاچـة ارومیـه، پایـان‏نامة کارشناسـی ارشـد هواشناسـی کـشاورزی، دانـشکدة کشاورزی، دانشگاه تهران.
مهرشاهی، د. و خسروی، ی. (1387). ارزیابی روش‏های میان‏یابی کریجینگ و رگرسیون خطی بر پایة مدل ارتفاعی رقومی جهت تعیین توزیع مکانی بارش سالانه (مطالعة موردی استان اصفهان)، فصل‏نامة برنامه‏ریزی و آمایش فضا، 14(4): 233-249.
میرموسوی، س.ح؛ مزیدی، ا. و خسروی، ی. (1388). تعیین بهترین روش زمین‏آمار جهت تخمین توزیع بارندگی با استفاده از GIS  (مطالعة موردی: استان اصفهان)، مجلة علمی- پژوهشی فضای جغرافیایی، 10(30): 105-120.
نادی، م؛ جامعی، م.؛ بذرافشان، ج، و جنت رستمی، س. (1391). ارزیابی روش‏های مختلف درون‏یابی داده‏‏های بارندگی ماهانه و سالانه، پژوهش‏های جغرافیای طبیعی، 44(4): 117-130.
نورزاده حداد، م.؛ مهدیان، م.ح. و ملکوتی، م.ج. (1392). مقایسة کارایی برخی روش‏های زمین‏آماری به منظور بررسی پراکنش مکانی عناصر ریزمغذی در اراضی کشاورزی، مطالعة موردی: استان همدان، نشریة دانش آب و خاک، 23(1): 71-81.
واقفی، م.؛ احمدآبادی، ع.؛ فتح‏نیا، ا.ا. و قدسیان، م. (1388). مقایسة روش‏های درون‏یابی در بررسی توپوگرافی بستر کانال‏های قوسی، فصل‏نامة علوم و مهندسی آبخیزداری ایران، 3(6): 17-26.
Abasi Jondani, Sh. and Malikinejad, H. (2014). Evaluate geostatistical methods based on GIS for spatial zoning of precipitation (Case Study: Plain Kvhpayh- Segzi). The second national conference wilderness on approach of management Regions Arid and desert, Faculty Desert Studies Semnan University, pp. 1-10.
Biau, G.; Zorita, E.; Storch, H.V. and Wackernagel, H. (1999). Estimation of Precipitation by Kriging in the EOF Space of the Sea Level Pressure Field, Journal of Climate,12: 1070-1085.
Bianchini, S.; Fabio, P.; Teresa, N. and Casagli, N. (2015). Building Deformation Assessment by Means of PersistentScatterer Interferometry Analysis on a Landslide- AffectedArea: The Volterra (Italy) Case Study, Remote Sens, 7: 4678-4701.
Buhmann, M.D. (2003). Radial Basis Functions, Theory and Implementations, Cambridge University Press.
Burrough, P.A. and McDonnell, R.A. (1998). Principles of Geographical Information Systems, Oxford University Press, Oxford.
Carratala, A.; Gomez, A. and Bellot, J. (1998). Mapping Rain Composition in the East of SPAIN by Applying Kriging, Water, Air, and Soil Pollution, 104(1-2): 9-27.
Chang, K.T. (2004). Introduction to Geographic Information System, 2nd edition, Boston; London: McGraw-Hill Higher Education, Book, pp. 1-400.
Cressie, N. (1993). Statistics for Spatial Data (revised edition), John Wiley & Sons, Inc., New York. pp. 1-928.
Cressie, N. (1985). Fitting variogram models by weighted least squares, Mathematical Geology, 22(3): 239-252.
Cressie, N. and Zimmerman, L. (1992). On the Stability of the Geostatistical Method, Mathematical Geology, 24(1): 45-58.
Eldeiry, A. and Garcia, L. (2011). Using Deterministic and Geostatistical Techniques to Estimate Soil Salinity at the Sub-Basin Scale and the Field Scale. Remote Sensing. 6(1): 1137-1157.
Faraje Sabokbar, H.A. and Azizi, Gh. (2006). To compare the accuracy of spatial interpolation methods Case study: modeling rainfall catchment Kardh Mashhad, Journal of Geographical Research, 58: 1-15.
Fasshauer, G.M. (2007). Approximation Methods with MATLAB, ser. Interdisciplinary mathematical sciences. Vol. 6. World Scientific Publishers, Singapore, pp. 1-500.
Friedel, M.J. and Iwashita, F. (2013). Hybrid modelling of spatial continuity for application to numerical inverse problems Environ, Model. Softw., 43: 60-79.
Gang, M. (2014). Evaluating the Power of GPU Acceleration for IDW Interpolation Algorithm. Hindawi Publishing Corporation, e Scientific World Journal, Article ID 171574, pp.1-8.
Gholmohamadi, G.; Marofi, S. and Mohammadi, K. (2007). Regional of runoff coefficient in Hamedan province using geostatistical methods and GIS, Journal of Science and Technology Agriculture and Natural Resources, 12(46): 501-514.
Goovaerts, P. (1997). Geostatistics for Natural Resources Evaluation, Oxford University Press, New York. 0-19-511-538-4. pp. 1-477.
Hengl, T.; Minasny, B. and Gould, M. (2009). A geostatistical analysis of geostatistics, Scientomettric, 80(2): 491-514.
Isaaks, E.H. and Srivastava, R.M. (1989). Applied Geostatistics, Oxford University Press, New York.
Jalali, Gh.; Tehrani, M.; Boromand, N. and Sanjeri, S. (2013). Comparison of the Geostatistics method to map the spatial distribution of some nutrient elements in the East Mazandaran province, Journal of Soil Research (Soil and Water), 27: 196-204.
Journel, A.G. and Huijbregts, C.J. (1978). Mining Geostatistics, Academic Press, London.
Li, J. and Andrew, D. (2014). Heap Spatial interpolation methods applied in the environmental sciences: A review, Environmental Modelling & Software, 53: 173-189.
Losser, T.; Li, L. and Piltner, R. (2014). A Spatiotemporal Interpolation Method Using Radial Basis Functions for Geospatiotemporal Big Data, Fifth International Conference on Computing for Geospatial Research and Application, pp. 17-24.
Masodian, S.A. (2003). Analysis of monthly temperature Iran, The research Journal of Human sciences, Isfahan University, 15: 87-96.
Meyer, P.A. (1982). Interpolation entre espaces d’Orlicz, Seminar on Probability, XVI, Lecture Notes in Mathematics, New York, 920: 153-158.
Mehdizadeh, M. (2002). Geostatistical analysis to estimate the temperature and precipitation in the drainage basin of Lake Urmia, Thesis Master of Agricultural meteorology, Faculty of Agriculture. Tehran University, Supervisor: Dr. Mohammad Hossein Mahdian.
Mehrshahi, D. and Khosravi, E. (2008). Evaluation of Methods Kriging interpolation and Linear regression based on Digital Elevation Model to determine the spatial distribution of rainfall (Case Study: Isfahan Province), Journal of Scheduled And Spatial Planning, 14(4):233-249.
Mirmosavi, S.H; Mazidi, A. and Khosravi, Y. (2009). Determine the best geostatistics method for estimating the distribution of rainfall using GIS (Case Study: Isfahan Province), Journal of geographic space, 10(30): 105-120.
Mohammadi, J. (2006). Pedometer (Spatial Statistics), Published Pelk, Tehran, Iran.
Nadi, M.; Jamie, M.; Bazrafshan, J. and Janatrostami, S. (2012). Evaluation Various methods Interpolation Monthly and annual rainfall data, Journal of natural geographical research, 44(4): 117- 130.
Naghefi, M.; Ahmadabadi, A.; Fathnia, A.A. and Ghodsian, M. (2009). Comparison of interpolation methods in the study Topographic base Curved channels, Journal of Management and Engineering Science Iranian, 3(6): 17-26.
Norzadeh Hadad, M.; Mehdian, M.H. and Malekoti, M.J. (2013). Compare the performance of some Geostatistical methods To investigate the spatial distribution of micronutrients in Agricultural land, Case Study: Hamedan Province, Journal of Soil and Water, 23(1): 71-81.
Rahimi Bandarabadi, S. and Mahdian, M.H. (2003). Spatial analysis of monthly rainfall in arid and semi-arid South East Iran, Third Regional Conference and first National Conference on Climate Change. 29 September to 1 October, Isfahan University, pp. 1-8.
Rangzan, K.; Mokhtari, M. and Shaigan, M. (2005). Assess the accuracy IDW and Kriging models of interpolation to data on Plain Groundwater levels between Shushtar water, Geomatics Conference, pp. 1-15.
Rostami, M.; Saljagheh, A.; Saneei, M. and Mahdavi, M. (2014). Evaluation of interpolation in the study of morphology the river and water channel, Jornal of Management and Engineering Science Iranian, 8(25): 57-67.
Safari, H. (2007). Comparison of two interpolation method Kriging and IDW, Shhrngar Journal, 30(40): 33-39.
Safari, M. (2002). Determine the optimum network of measuring groundwater levels using geostatistical methods. Case study: Plain Chmchal, Irrigation and Drainage Master's thesis. Faculty of Agriculture, Tarbiat Modarres University. Supervisor: Dr. Syed Majid Merlatifi.
Sistena, S. and Tumbobc, M. (2015). Interpolation of daily rain gauge data for hydrological modeling in data sparse regions using pattern information from satellite data, Hydrological Sciences Journal, pp. 1-27.
Shamsadin, A. (2000). Changes in regional rainfall using kriging method in northern provinces. The Irrigation and Drainage Master's thesis, College of Agriculture, Shiraz University, Supervisor: Dr. Syed Ali Abtahi.
Shepard, D. (1968). A two-dimensional interpolation function for irregularly-spaced data. In: Proceedings of the 1968 23rd ACM National Conference, pp. 517-524.
Soltani, S. and Modares, R. (2006). Analysis of frequency and intensity Meteorological drought Isfahan Province, Journal of Natural Resources Iran, 1: 15-26.
Sun, B. and Petreson, T.C. (2006). Estimating precipitation normal for USCRN stations, Journal of Geophysical Research, 111(D9): 1984-2012.
Tazeh, M.; Kavsari, M.R.; and Khosravi, E. (2008). Zoning, drought based index of Transv using geostatistical and GIS (Case Study: the western part of the province), International Conference on botany trees and climate change on the ecosystems of Caspian, Sari, Caspian ecosystems Research Center, pp. 1-12.
Wackernagel, H. (2003). Multivariate Geostatistics: An Introduction with Applications, third ed., Springer, Berlin.
Webster, R. and Oliver, M.A. (1992). Sample adequately to estimate variograms of soil propert--ies, J.Soil Sci., 43: 177-192.
 Wendland, H. and Rieger, CH. (2005). Wendland. H Approximate Interpolation with Applications to Selecting Smoothing Parameters, Numerische Mathematik, 101(4):729-748.
Zhu, P.; Zhang, L.W. and Liew, K.M. (2013). Geometrically nonlinear thermomechanical analysis of moderately thick functionally graded plates using a local Petrov-Galerkin approach with moving Kriging interpolation, Composite Structures, 107: 298-314.
Zimmerman, L. (1993). Another look at anisotropy in geostatistics. Mathematical Geology, 25, (4): 453–470.
Physical Geography Research
Volume 50, Issue 2
July 2018
Pages 323-338

Files

History

  • Receive Date: 05 June 2015
  • Revise Date: 20 December 2017
  • Accept Date: 09 January 2018

Share

How to cite

Statistics