پیش بینی عملکرد پسته با استفاده از رگرسیون چندمتغیره ی خطی و شبکه عصبی مصنوعی (مطالعه موردی: شهرستان های رفسنجان و انار استان کرمان)

نوع مقاله: مقالات تحلیلی-تفسیری

نویسندگان

1 دانشجوی دکتری گروه علوم خاک، دانشگاه شهرکرد، شهرکرد، ایران.

2 استاد گروه علوم خاک، دانشگاه شهرکرد، شهرکرد، ایران.

3 استادیار پژوهشکده پسته، مؤسسه تحقیقات علوم باغبانی، سازمان تحقیقات، آموزش و ترویج کشاورزی.

4 دانشیار گروه علوم خاک، دانشگاه ولی‌عصر (عج) رفسنجان، رفسنجان، ایران.

چکیده

امروزه، مدیریت اصولی اراضی به‏عنوان یک راهکار مهم برای رسیدن به عملکرد بیشتر در واحد سطح و استفاده بهینه از منابع خاک و آب، مورد توجه پژوهشگران، تولیدکنندگان و سیاست­گذاران عرصه کشاورزی قرار گرفته است. پژوهش حاضر با هدف بررسی ارتباط بین عملکرد پسته و عوامل مؤثر بر آن، صورت پذیرفت. بدین منظور، 129 قطعه باغ در مناطق مختلف شهرستآن‌های رفسنجان و انار شناسایی و انتخاب گردید. نمونه­برداری از آب آبیاری، برگ درختان و خاک همه باغ­ها انجام شد. همچنین برای هر باغ یک پرسشنامه به منظور جمع­آوری اطلاعات مدیریتی و تعیین مقدار عملکرد تهیه شد. در نهایت یک متغیر وابسته یعنی عملکرد محصول و 50 متغیر مستقل شامل ویژگی­های خاک، آب و گیاه برای انجام مدل­سازی به کمک مدل­های رگرسیون چند متغیره خطی و شبکه­های عصبی مصنوعی مورد استفاده قرار گرفت. نتایج نشان می­دهد که رگرسیون چند متغیره­ی خطی تنها 26 درصد تغییرات عملکرد را توجیه می­نماید اما وقتی با تقسیم منطقه به چهار بخش، داده­ها همگن­تر می‌شود، دقت این روش افزایش یافت. به طوری که ضریب تبیین اصلاح شده­ی مدل برای باغ­های منطقه نوق، انار، حومه شرقی و حومه غربی به ترتیب به حدود 4/92، 5/81، 95 و 6/53 درصد رسید. این مدل­ها، به ویژگی­های مربوط به آب آبیاری حساسیت زیادی نشان می­دهند. بنابراین، توجه ویژه به روش­های نوین آبیاری و اتخاذ رویکردهای صحیح مدیریتی به منظور افزایش بهره­وری آب ضروری به نظر می‌رسد. شبکه عصبی مصنوعی با 9 نرون در یک لایه پنهان، تابع فعال‌سازی تانژانت-سیگموئید و تابع آموزشی لونبرگ مارکوات دارای دقت 3/98 درصدی در پیش­بینی عملکرد محصول پسته در کل منطقه مورد مطالعه می­باشد.

کلیدواژه‌ها


عنوان مقاله [English]

Pistachio yield prediction using multiple linear regression and artificial neural network (A Case Study: Rafsanjan and Anar regions, Kerman Province)

نویسندگان [English]

  • Behrooz pourmohamadali 1
  • M.H. Salehi 2
  • S.J. Hosseinifard 3
  • H. Shirani 4
  • I. Esfandiarpour Borujeni 4
1 Ph.D. Student, Soil Science Department, Shahrekord University, Shahrekord, Iran.
2 Professor, Soil Science Department, Shahrekord University, Shahrekord, Iran.
3 Assistant Professor, Pistachio Research Center, Horticultural Sciences Research Institute Agricultural Research, Education and Extension Organization.
4 Associate Professor, Soil Science Department, Vali-e-Asr University of Rafsanjan, Rafsanjan, Iran.
چکیده [English]

Introduction In recent decades, due to a significant increase of pistachio cultivation and uncontrolled exploitation of groundwater resources as well as reducing rainfall precipitation, groundwater level has dropped and the quantity and the quality of water has also been reduced. Therefore, agricultural producers, researchers and policy makers need to pay more attention to appropriate land management as an important strategy to achieve greater yield per unit area and optimal use of soil and water resources. Crop yield prediction regarding its temporal and spatial variations has an important role in developing proper management programs. However, few studies have been carried out in relation to pistachio yield prediction using an acceptable range of features on regional scale. In the present study, pistachio yield modeling was performed by multivariate linear regression and artificial neural networkbased on soil, water and management features.
Materials and Methods 129 orchard plots in different areas of Rafsanjan and Anar were identified and selected. The study area is located between 54° 56′ and 56° 41′ E, 29° 54′ and 31° 13′ N. Soil sampling, was performed from the areas under pistachio canopy and three soil depths of 0 to 40, 40 to 80 and 80 to 120 cm in each plot, fully expanded sub-terminal leaflets were randomly collected from non-fruiting branches, during the late July through August. Irrigation water of all orchards was also sampled. Moreover, for each orchard, a questionnaire was prepared to collect management and yield data. Soil quality indicators including particle size distribution, pH in saturated soil paste, electrical conductivity of saturated extract, soluble sodium, soluble calcium, soluble magnesium, available phosphorus and available potassium were determined for soil samples. The concentrations of phosphorus, potassium, iron, zinc, copper, manganese, calcium and magnesium in leaf samples and electrical conductivity in water samples, were also calculated. Finally, a dependent variable (pistachio yield) and 50 independent variables including soil, water and plant characteristics were used for modeling. For this purpose, stepwise multiple linear regression and artificial neural network technique were applied. Then, the study area was divided into 4 parts with the highest pistachio orchards densities and regression models were run for each part, separately. The ability of models to yield prediction was evaluated using the root mean square error (RMSE), relative root mean square error (% RMSE), adjusted coefficient of determination (adj - R2) and Durbin - Watson statistic (D – W).
Results and Discussion The average of yield in the study area is about 1,700 kilograms per hectare. Results indicated that multiple linear regression could explain only 26 percent of the pistachio yield variation, but its accuracy increased when data became more homogeneous via dividing the study area into four parts. The model adjusted-R2 for Noogh, Anar, eastern suburbs and western suburbs orchards rose to about 92.4, 81.5, 95 and 53.6 percent, respectively. In all regression models except the model of western suburbs, at least one of the characteristics associated with irrigation water was significant. Artificial neural network with 9 neurons in a hidden layer, Tangent - sigmoid activation function and Levenberg - Marquardt training function, has a 98.3 percent accuracy in predicting pistachio yield in the study area (% RMSE = 13.8).
Conclusion Multivariate linear regression model did not accurately predict the pistachio yield for the whole of study area whereas increasing data homogeneity and decreasing sources of variations, reduced complexity of relationships between features which  resulted in increasing of the efficiency of linear regression to modeling these relationships. These models were highly sensitive to irrigation water features. Therefore, special attention should be paid to modern irrigation techniques and proper management approaches in order to enhance water efficiency. Overall, artificial neural network had greater accuracy compared to multivariate linear regression for pistachio yield modeling. This indicates the existence of non-linear and complex relationships between pistachio yield and the factors affecting yield and also the necessity of using modern and robust data mining tools for crop yield estimating. It seems that artificial intelligence techniques can be used as an efficient tool for developing proper management programs.

کلیدواژه‌ها [English]

  • Pistachio yield
  • modeling
  • Multiple linear regression
  • Artificial neural network
  1. Adams, M.L., Cook, S.E., Caccetta, P.A. and Pringle, M.J. 1999. Machine learning methods in site-specific management research: An Australian case study. Proceedings of the fourth international conference on precision agriculture, Madison, USA. pp: 1321–1333.
  2. Alipour, H., and Hosseinifard, S.J. 2003. Detection and alleviation nutrient deficiency in pistachio. Ministry of Agriculture, Agricultural Research Organization. 53p. (In Persian).
  3. Anonymous. 2015. Agricultural Statistics in 2013-2014, Vol. 3: Horticultural crops. Ministry of Agriculture, Deputy of Planning and Economy, ICT Center. 138p. (In Persian).
  4. Bagheri, S., Gheysari, M., Ayoubi, Sh. and Lavaee, N. 2012. Silage maize yield prediction using artificial neural networks. Journal of Plant Production, 19(4): 77-95. (in Persian with English abstract).
  5. Barrows, H.L. and Elmer, C.S. 1962. An EDTA Method for the Direct Routine Determination of Calcium and Magnesium in Soils and Plant Tissue.Soil Science Society of America Journal, 26(5): 443-445.
  6. Beede, R.H., Brown, P., Kallsen, C. and Weinbaum, S.A. 2005. Diagnosing and correcting nutrient deficiencies. In: Ferguson, L. (ed.) Pistachio Production Manual, 4th Edition. pp. 147-157.
  7. Brye, K.R., Slaton, N.A., Stavin, M.C., Norman, R.J. and Miller, D.M. 2003. Short-torm effects of land leveling on soil physical properties and Microbial Biomass. Soil Science Society of America Journal, 67: 1405- 1417.
  8. Chapman, H.D. and Pratt, P.F. 1961. Methods of analysis for soils, plants and waters. Division of Agricultural Sciences, University of California, Riverside, USA.
  9. Chitsazan , M., Rahmani, Gh. and Neyamadpour, A. 2015. Forecasting groundwater level by artificial neural networks as an alternative approach to groundwater modeling. Journal of the Geological Society of India, 85(1): 98-106.
    1. Dahiya, I.S., Richter, J. and Malik, R.S. 1984. Soil spatial variability: A review. International Journal of Tropical Agriculture, 11(1): 1-102.
    2. Davarynejad, G.H., Azizi, M. and Akheratee, M. 2009. Effect of foliar nutrition on quality, quantity and of alternate bearing of Pistachio (Pistacia vera L.). Journal of Horticultural Sciences, 23(2): 1-10. (In Persian with English abstract).
    3. Durbin, J. 1969. Tests for serial correlation in regression analysis based on the periodogram of least-squares residuals. Biometrika, 56 (1): 1-15.
    4. Eskandari, S., Mozaffari, V., and Tajabadi Pour, A. 2014. Effects of salinity and copper on growth and chemical composition of pistachio seedlings. Journal of Plant Nutrition, 37(7): 1063-1079.
    5. Ferguson, L., Polito, V. and Kallsen, C. 2005.  The pistachio tree; botany and physiology and factors that affect yield. In Ferguson, L. (ed.), Pistachio Production Manual, 4th Edition. pp: 31-39.
    6. Foody, G.M. 2000. Mapping Land Cover from Remotely Sensed Data with a Softened Feedforward Neural Network Classification. Journal of Intelligent and Robotic Systems, 29: 433-449.
    7. Gee, G.W. and Bauder, J.W. 1986. Particle size analysis. In: Klute A. (Ed), Methods of soil analysis. Part 1. 2nd ed. Agron. Monogr. 9. ASA. Madison. WI, USA. pp: 383-411.
    8. Goldhamer, D.A. 1987. Effects of substained defict irrigation on pistachio tree performance. California Pistachio Industry. Annual Report-Crop year. pp: 61-66.
    9. Hajizad, Sh., Samadi, A., Movahedi Naeini, S.A.R. and Khormali, F. 2010. Quantity-intensity parameters and forms of potassium in relation to clay mineralogy in some vineyard growing calcareous soils in Western Azarbaijan Province. Journal of Water and Soil Conservation, 17(3): 65-83. (in Persian with English abstract)
    10. Hasheminasab, H. and Assad, M.T. 2015. A predictive model for breeding of pistachio yield stability under water stress condition. Biological Forum, 7(1): 212-217.
    11. Heydari, M. 2006. Identification of dominant soils and the effect of their properties on leaf concentration, quantity and quality of pistachio in Anar region, Rafsanjan. MSc. Thesis, Soil Science Department, Agricultural College, University of Shahrekord, Shahrekord, Iran. (In Persian with English abstract).
    12. Hosseinifard, S.J, Khademi, H., and Kalbasi, M. 2010. Different forms of soil potassium as affected by the age of pistachio (Pistacia vera L.) trees in Rafsanjan, Iran. Geoderma. 155: 289-297.
    13. Hosseinifard, S.J., Salehi, M.H., Esfandiarpour, I. and Mohammadi, J. 2008. Spatial variability of groundwater quality and its relationship with pistachio yield in Anar region, Iran. Journal of Applied Sciences, 8 (20): 3697-3702.
    14. Husnjak, S. and Hengle, T. 2006. Evaluation adequacy and usability of soil maps in Croatia. Soil Science Society of America Journal, 70: 920-929.
    15. Kaul, M., Hill, R.L. and Walthall, C. 2005. Artificial neural networks for corn and soybean yield prediction. Agricultural Systems, 85: 1-18.
    16. Khakural, B.R., Robert, P.C., and Huggins, D.R. 1999. Variability of corn/soybean yield and soil/landscape properties across a southwestern Minnesota landscape. In Proceedings of the fourth international conference on precision agriculture, pp: 573-579.
    17. Khoshgoftarmanesh, A.H. and Siadat, H. 2002. Mineral Nutrition of vegetables and horticultural crops in saline conditions. Ministry of Agriculture, Deputy of Horticulture, Iran. 87p. (In Persian)
    18. Knudsen, D., Peterson, G.A. and Pratt, P.F. 1982. Lithium, sodium and potassium. In: Page, A.L. Miller R.H. and Keeny D.R. (eds.), Methods of Soil Analysis. Part 2: Chemical and Microbiological Properties. Agron. Monogr. 9, ASA, Madison, WI, USA. pp: 225-246.
    19. Kravchenko, A.N., and Bullock, D.G. 2000. Correlation of corn and soybean grain yield with topography and soil properties. Agronomy Journal, 92(1): 75–83.
    20. Landeras, G., Ortiz-Barredo, A. and López, J.J. 2009. Forecasting weekly evapotranspiration with ARIMA and artificial neural network models. Irrigation and Drainage Engineering, ASCE, 135: 323-334.
    21. Malakouti, M.J., Tabatabaei, S.J., 1999. Proper nutrition of fruit trees for improving the yield and quality of horticultural crops in Iran. Agricultural education publications. Karaj, Iran. 266 p. (In Persian)
    22. Miao, Y., Mulla, D.J. and Rober, P.C. 2006. Identifying important factors influencing corn yield and grain quality variability using artificial neural networks. Precision Agriculture, 7: 117–135.
    23. Moazenpour, M. 1994. Evaluation of drought resistance of pistachio trees and determining the irrigation depth and interval. The report of Pistachio Research Institute of Iran, pp: 233-240. (In Persian).
    24. Mohajer, R., Salehi, M., and Beigi Herchegani, H. 2009. Estimating Soil cation exchange capacity (in View of Pedotransfer Functions) Using Regression and Artificial Neural Networks and the Effect of Data Partitioning on Accuracy and Precision of Functions. Water and Soil Science, 13(49): 83-97. (in Persian with English abstract)
    25. Mohammadi Mohammadabadi, A., Salehi, F. and Nekouie, M. 2004. Determining reference crop potential evapotranspiration ETo (Grass). The final report of Pistachio Research Institute of Iran. 21p. (In Persian)
    26. Mohammadi Mohammadabadi, A. and  Sedaghati, N. 2010. Water. In: Mehrnejad, M.R. and Javanshah, A. (eds.), The strategic framework for developing and promoting pistachio research in Iran. Pistachio Research Institute of Iran, Jomhouri publication. pp: 71-110. (In Persian)
    27. Mohsenian, Y., Roosta, H.R., Karimi, H.R. and Esmaeilizade, M. 2012. Investigation of the ameliorating effects of eggplant, datura, orange nightshade, local Iranian tobacco, and field tomato as rootstocks on alkali stress in tomato plants. Photosynthetica, 50(3):411-421.
    28. Mozaffari, V. 2005. The role of potassium, calcium and zinc in controlling Pistachio dieback. PhD. Thesis, Soil Science Department, Agricultural College, Tarbiat Modarres University. Tehran, Iran. (in Persian).
    29. Norouzi, M., Ayoubi, S., Jalalian, A., Khademi, H. and Dehghani, A.A. 2010. Predicting rainfed wheat quality and quantity by artificial neural network using terrain and soil characteristics. Acta Agric Scandinavica, Section B-Plant and Soil Sciences, 60: 341-352.
    30. Olsen, S.R. and Sommers, L.E. 1982. phosphorus. In: Page A.L. (Eds.), Methods of Soil Analysis. Part 2. 2nd ed. Argon. Mongr. 9. ASA and SSSA, Madison, WI, USA. pp. 403-430.
    31. Pachepsky, Y., Timlin, D.J. and Varallyay, G. 1996. Artificial neural networks to estimate soil water retention from easily measurable data. Soil Science Society of America Journal, 60(3): 727-733.
    32. Papageorgiou, E.I., Markinos, A. and Gemptos, T. 2009. Application of fuzzy cognitive maps for cotton yield management in precision farming. Expert Systems with Applications, 36: 12399-12413.
    33. Qadir, M., Steffens, D., Yan, F. and Schubert, S. 2003. Sodium removal from a calcareous saline–sodic soil through leaching and plant uptake during phytoremediation. Land Degradation and Development, 14: 301-307.
    34. Sajadi, M., Zeinedini, A. and Mahmudi, Sh. 2012. Quality impacts of irrigation on soil properties and yield of pistachio in plain Robat Shahrbabak. Journal of Irrigation and Water Engineering, 7:36-45. (In Persian with English abstract)
    35. Sajadian, H., Hokmabadi, H. 2011. Investigation on Effects of Calcium Nitrate, Calcium Sulfate (soil application) and Calcium chelate as Foliar spray on Endocarp Lesion Disorder and Some Qualitative Characteristics of pistachio nut Cv. Akbari. International Journal of Nuts and Related Sciences, 2(3): 23-28.
    36. Salehi, M.H. and Hosseinifard, S.J. 2012. Soil and groundwater relationships with pistachio yield in the Rafsanjan area, Iran. Communications in Soil Science and Plant Analysis, 43: 660-671.
    37. Salehi, M.H., Hosseinifard, S.J., Heydari, M. 2010. The effect of soil and leaf nutrients on pistachio yield in the Rafsanjan area, Iran.OptionsMéditerranéennes: Série A. Séminaires Méditerranéens, 94: 79 -84.
    38. Sanden B., Fulton, A. and Ferguson, L. 2005. Managing salinity, soil and water amendments. In: Ferguson, L. (ed.) Pistachio Production Manual, 4th Edition. pp. 129-146.
    39. Sarcheshmepour, M. and Malakouti, M.J. 2005. Necessity for potassium fertilization in pistachio (higher yield with better quality). Technical Bulletin No. 442. Soil and Water Research Institute, Sana Publications. 17p. (In Persian)
    40. Schultz, A., Wieland, R., and Lutze, G. 2000. Neural networks in agroecological modeling-stylish application or helpful tool?. Computers and Electronics in Agriculture, 29, 73–97.
    41. Sedaghati, N., Hosseinifard, S.J., Hokmabadi, H. and Mohammadi Mohammadabadi, A. 2012. Study the possibility of alternate irrigation borders in order to reducing irrigation intervals in pistachio orchards. The final report of Pistachio Research Institute of Iran. 22p. (In Persian)
    42. Sedaghati, N., Hosseinifard, S.J. and Mohammadi Mohammadabadi, A. 2009a.  The study of possibility to decrease of irrigation borders width in mature pistachio (Pistacia vera L.) orchards. Pajouhesh And Sazandegi, 21: 153-162. (in Persian with English abstract)
    43. Sedaghati, N., Mohammadi Mohammadabadi, A. and Hosseinifard, S.J. 2009b. The effect of irrigation regimes on occurrence of early split of pistachio (Pistacia vera L.) CV. ohadi. Pajouhesh and Sazandegi, 21(78): 149-158. (in Persian with English abstract)
    44. Seybold, C.A., Grossman, R.B. and Reinsch, T.G. 2005. Preicting Cation Exchange Capacity for Soil Survey Using Linear Models. Soil Science Society of America Journal, 69: 856-86.
    45. Shabani, A., Haghnia, Gh., Karimi, A. and Ahmadi, M.M. 2011. Wheat yield prediction using artificial neural networks in Sisab Region, Northeastern Iran. 12th Iranian Soil Science Congress, Tabriz, Iran. (In Persian)
    46. Soares, J.D.R., Pasqual, M., Lacerda, W.S., Silva, S.O. and Donato, S.L.R. 2013. Utilization of artificial neural networks in the prediction of the bunches’ weight in banana plants. Scientia Horticulturae, 155: 24–29.
    47. Tekin, H., Guzel, N. and Ibrikci, H. 1995. Influence of manure and inorganic fertilizer on yield and quality of pistachio. Journal of Plant Nutrition, 18(6): 1263-1272.
    48. Wilding, L.P. and Dress, L.R. 1983. Spatial variability and pedology. In Wilding, L.P., Smeckand, N.E. and Hall G.F. (eds.), Pedogenesis and Soil Taxonomy. I. Concepts and Interactions. Elsvier Science Pub. pp. 83-116.
    49. Yu, H., Liu, D., Chen, G., Wan, B., Wang, Sh. and Yang, B. 2010. A neural network ensemble method for precision fertilization modeling. Mathematical and Computer Modelling, 51:1375-1382.
    50. Zeinadini, M., Shirani, H., Mozafari, V. and Esfandiarpour, I. 2013. Variability of soil electrical conductivity and sodium adsorption ratio and their impact on growth Pistachio. Journal of Water and Soil Conservation, 20(6): 165-181. (In Persian with English abstract)
    51. Zeng, D.Q., Brown, P.H. and Holtz, B.A. 2001. Potassium fertilization affects soil K, leaf K concentration, and nut yield and quality of mature pistachio trees. Horticultural Science, 36(1):85–89.
    52. Zhang, X.C. and Norton, L.D. 2002. Effect of exchangeable Mg on saturated hydraulic conductivity, disaggregation and clay dispersion of disturbed soils. Journal of Hydrology, 260: 194-205.