تعیین مهم ترین پارامترهای مؤثر خاک بر فراهمی فسفر در دشت سیستان به روش ارتباط وزنی در شبکه های عصبی

نوع مقاله : مقالات پژوهشی

نویسندگان

دانشگاه زابل

چکیده

فسفر به عنوان یک عنصر ضروری در تولید محصولات کشاورزی دارای اهمیت است. از سوی دیگر توانایی آن در القای کمبود عناصر کم-مصرف ضروری و اثرات منفی آن بر محیط زیست، سبب توجه بیشتر به این عنصر شده است. از آنجا که ویژگی های خاک از عوامل مهم در واکنش فسفر در خاک هستند، پژوهش حاضر جهت بررسی و تعیین مهم ترین ویژگی های خاک موثر بر فراهمی فسفر با استفاده از روش های رگرسیونی و شبکه‌های عصبی مصنوعی در دشت سیستان انجام شد. بدین منظور تعداد 200 نمونه خاک از اراضی دشت سیستان تهیه و مقادیر فسفر قابل جذب و سایر پارامترهای فیزیکو شیمیایی آن اندازه گیری گردید. نتایج بیانگر آن است که روش شبکه عصبی دارای دقت بیشتری در برآورد فسفر قابل جذب نسبت به روش رگرسیون چند متغیره خطی می‌باشد، به گونه‌ای که شبکه عصبی پرسپترون چند لایه با آرایش 1-6-4 نزدیک به 90 درصد از تغییرات فسفر قابل جذب را با استفاده از برخی ویژگی‌های خاک (درصد رس، ماده آلی، کربنات کلسیم و اسیدیته) پیش‌بینی نمود ولی معادله رگرسیون حاصله تنها توانست 43 درصد از تغییرات فسفر را توجیه کند. نتایج کمی کردن اهمیت متغیرها به روش وزن ارتباطی نشان داد عامل pH بیشترین مشارکت را در تغییرپذیری فسفر در منطقه مورد مطالعه دارد. به عبارت دیگر، مقادیر بالای pH مهم ترین عامل محدود کننده فراهمی فسفر در خاک های دشت سیستان است.

کلیدواژه‌ها


عنوان مقاله [English]

Determination of the Most Important Soil Parameters Affecting the Availability of Phosphorus in Sistan Plain, Using Connection Weight Method in Neural Networks

نویسندگان [English]

  • H. Mir
  • A. Gholamallzadeh Ahangar
  • A. Shabani
University of Zabol
چکیده [English]

Introduction: Phosphorus is important as an essential element in the production of agricultural products. On the other hand, its ability to induce essential micronutrient deficiency and its negative effects on the environment, have attracted more attention to this element. The knowledge of phosphorus availability conditions in the soil and consequently the accurate management of fertilizer consumption has a key role in the environmental protection. The degree of phosphorus absorption in the soil depends on the environmental factors, soil characteristics and compositions, and phosphorus fertilizer which have been used. The amount of available phosphorus in the soil has relationship with some of the physical and chemical properties of the soil. Since, the soil characteristics are important factors in the reaction of phosphorus in the soil, the present study aimed to investigate and determine the most important soil characteristics affecting the availability of phosphorus using regression and artificial neural network techniques, in the soils of Sistan plain.
Materials and Methods: Soil sampling was done in 1.5×1.5 km intervals, from 0-30 cm depth, and 200 soil samples were taken. The amounts of available phosphorus and the soil properties including the percentages of clay , organic matter, calcium carbonate and the amount of pH were measured. Then, stepwise multivariate linear regression analysis was performed to determine linear relation between available phosphorus and the soil properties. In order to model and validate the regression model, respectively 80 and 20% of data were selected and entered into SPSS software. To train the neural network, multilayer perceptron (MLP) network was used by MATLAB 7.6 package. In this type of network, 70% of data is selected for training, 15% for validation and 15% for testing the model. Levenberg-Marquardt algorithm and hyperbolic tangent (as a transfer function) were used to train the network. The numbers of neurons in the hidden layer were calculated based on the trial and error method and finally the best structure was selected according to the highest R2 and the lowest RMSE value. Moreover, quantifying the importance of variables in the neural network was done through employing connection weight approach. In this method, the connection weights of input-hidden and hidden-output neurons were used to indicate the significance of variables.
Results and Discussion: The values of the coefficient of variation for the soil properties were in the range of 5.66 for pH (the lowest) and 69.90 for available phosphorus (the highest). The high variation of the available phosphorus could be due to the different amounts of phosphorus fertilizers consumption and their diverse rate of conversion to less soluble forms. The validation results of regression and neural network methods showed that the latter technique was more accurate compared with the multivariate linear regression method, in the estimation of available phosphorus, as multi-layer perceptron neural network with 4-6-1 layout predicts nearly 90% of available phosphorus variability using soil properties (percentage of clay, organic matter, calcium carbonate and the amount of pH); however, the obtained regression equation could explain only 43% of phosphorus variances. The reasons for this could be: 1) considering nonlinear relations between the variables in the artificial neural network method, and 2) less sensitivity of this method to the existence of error in input data, comparing with the regression method. The values of R2 and RMSE were 0.43 and 11.23, respectively for training the multivariate linear regression method and 0.91 and 4.28, respectively for training the artificial neural network method. From the investigated soil properties in the current study, the percentage of organic matter and clay were entered in the regression model, and the values of standardized regression coefficient (beta) showed that the first variable is more important to explain the available phosphorus variability. The results of quantifying the importance of variables by the connection weight method showed that pH had the greatest contribution in the variability of phosphorus in the study area. In the other words, the high values of pH were the most important limiting factor for the availability of phosphorus in Sistan soils.
Conclusion: Considering nonlinear and complicated relations between variables, the artificial neural network model is an effective tool to assess the effect of soil properties on the availability of phosphorus in the study area. The results of quantifying the importance of variables by using the connection weight method showed that pH had the greatest contribution in the variability of phosphorus in the study area. In fact, the existence of lime in the soils of the study area, arid climate and lack of precipitation have resulted in the accumulation of basic cations in the soil and consequently increased pH values. Furthermore, the observed average values of pH that are more than 8.5 demonstrated the risk of soil sodicity in the study area. Thus, the management of this area by cultivating tolerant plants could be resulted in increasing organic matter content, which along with using chemical amendments such as sulfur will decrease pH values and increase the availability of phosphorus in Sistan plain. Applying such practices and through it modifying soil characteristics, decreasing the consumption of phosphate fertilizers and preventing their hazardous environmental effects would be expected in long run.

کلیدواژه‌ها [English]

  • Available phosphorus
  • Connection Weight Method
  • Multivariate Regression
  • Neural Network
Agyare W.A., and Park S.J. 2007. Artificial neural network estimation of saturated hydraulic conductivity. Vadose Zone Journal, 6: 423-431.
2- Ayoubi S., Mehnatkesh A., Jalalian A., Sahrawat K.L., and Gheysari M. 2014. Relationships between grain protein, Zn, Cu, Fe and Mn contents in wheat and soil and topographic attributes. Archives of Agronomy and Soil Science, 60 (5): 625-638.
3- Bertrand I., Holloway R.E., Armstrong R.D., and Mclaughlin M.J. 2003. Chemical characteristics of phosphorus in alkaline soils from southern Australia. Australian Journal of Soil Research, 41: 61-76.
4- Besalatpour A.A., Ayoubi S., Hajabbasi M.A., Mosaddeghi M.R., and Schulin R. 2013. Estimating wet soil aggregate stability from easily available properties in a highly mountainous watershed. Catena, 111: 72-79
5- Bocco, M., Willington, E. and Arias M., 2010. Comparison of regression and neural networks models to estimate solar radiation. Chilean Journal of Agricultural Research, 70: 428-435.
6- Carreira J.A., Vinegla B., and Lajtha K. 2006. Secondary CaCo3 and precipitation of P-Ca compounds control the retention of soil P in and ecosystems. Journal of Arid Environments, 64(3): 460-473.
7- Chahooki M.A.Z. 2010. Data Analysis in Natural Resources Research using SPSS Software. Jihad e Daneshgahi, Tehran. (in Persian)
8- Dadgar M., Aliha M., and Faramarzi E. 2011. Relationship between available phosphorus and some soil physical and chemical characteristics in Absard Plain (Damavand Province). Iranian journal of Range and Desert Reseach, 18 (3): 498-504. (in Persian with English abstract)
9- Dahiya I.S., Richter J., and Malik R.S. 1984. Soil spatial variability: A review. International Journal of Tropical Agriculture, 11:1-102.
10- Delgado A., Madrid A., Kassem S., Andreu L., and Campillo M. C. 2002. Phosphorus fertilizer recovery from calcareous soils amended with humic and fluvic acids. Journal of Plant and Soil, 245: 277-286.
11- Dia X., Huo Z., and Wang H. 2011. Simulation for response of crop yield to soil moisture and salinity with artificial neural network. Field Crops Research, 121:441-449.
12- Hallajnia A., Fotovat A., and Khorasani R. 2006. Availability of soil phosphorus resulting from different amounts of phosphorus fertilizer in soils of Hamedan province. Journal of Science and Technology of Agriculture and Natural Resources, 4(10): 121-132. (in Persian)
13- Hattab N., Hambli R., Motelica-Heino M., Bourrat X., and Mench M. 2013. Application of neural network model for the prediction of chromium concentration in phytoremediated contaminated soils. Journal of Geochemical Exploration, 128: 25-34.
14- Jalali M., and Kolahchi Z. 2005. Availability of soil phosphorus resulting from different amounts of phosphorus fertilizer in soils of Hamedan province. Soil and Water Science, 19(1): 53-60. (in Persian)
15- Kemp S.J., Zaradic P., and Hansen F. 2007. An approach for determining relative input parameter importance and significance in artificial neural networks. Ecological Modelling, 204: 326-334.
16- Keshavarzi A., Sarmadian F., Sadeghnejad M., and Pezeshki P. 2010. Developing pedotransfer functions for estimating some soil properties using artificial neural network and multivariate regression approaches. ProEnvironment, 3: 322-330.
17- Kumar M., Raghuwanshi N.S., Singh R., Wallender W.W., and Pruitt W.O. 2002. Estimating evapotranspiration using artificial neural network. Journal of Irrigation and Drainage Engineering-ASCE, 128: 224-233.
18- Kuo S. 1996. Total organic phosphorus. p. 869-919. In D.L. Sparks (ed.) Methods of Soil Analysis. Part 3. Chemical Methods. SSSA. Madison, WI.
19- Malakouti M.J., and Homaee M. 2003. Soil Fertility in Arid and Semiarid Regions “Problems and Solutions”. 2nd Ed. Tarbiat Modarres University, Tehran. (in Persian)
20- Menhaj M.B. 2012. Fundamentals of Neural Networks. No. 1. Amirkabir University, Tehran. (in Persian)
21- Minasny B., Hopman J., Harter W.T., Eching S.O., Toli A., and Denton M.A. 2004. Neural networks prediction of soil hydraulic functions for alluvial soils using multistep outflow data. Soil Science Society of American Journal, 68: 417-429.
22- Mohebbi S.M.J. 2014. Investigation of relationships between available phosphorus, potassium and some soil properties in agricultural lands of Varamin- Iran. International Journal of Agriculture and Biosciences, 3(1): 7-12.
23- Mostashari M., Ardalan M., Karimian N., Rezaei H., and Mirhoseini H. 2009. Distribution of organic forms of phosphorus and its relation with soil properties in some calcareous soils of Qazvin province. Journal of Soil Research (Soil and Water Science), 23(1): 11-22. (in Persian)
24- Olden J.D., and Jackson D.A. 2002. Illuminating the black box approach for understanding variable contributions in artificial neural networks randomization. Ecological Modelling, 154: 135-150.
25- Olden J.D., Joy M.K., and Death R.G. 2004. An accurate comparison of methods for quantifying variable importance in artificial neural networks using simulated data. Ecological Modelling, 178: 389-397.
26- Oskoie P.A., Takrimi Nia S.M., and Kahneh E. 2014. Correlations between content and some soil properties in groundnut cultivation, north of Iran. International Journal of Plant, Animal and Environmental Sciences, 4(2): 288-290.
27- Panday S., Thapa K.B., and Oli I.B. 2012. Correlations of available phosphorus and potassium with soil pH and organic matter content at different soil reactions categories in soils of western development region, Nepal. Journal of Chemical, Biological and Physical Sciences, 3:128-133.
28- Piri H., and Ansari H. 2013. Investigating drought in Sistan plain and its effect on Hamoun international wetland. Talab, 4(15): 63-73. (in Persian)
29- Rezaee Pazhand H. 2001. Application of Probability and Statistics in Water Resources. 1st Edition. Sokhan Gostar, Mashhad. (in Persian)
30- Salaridini A.A. 2008. Soil Fertility. University of Tehran, Tehran. (in Persian)
31- Singh R.P., and Mishra S.K. 2012. Available macro nutrients (N, P, K and S) in the soils of Chiraigaon block of district Varanasi (U.P.) in relation to soil characteristics. Indian Journal of Scientific Research, 3(1): 97-100.
32- Soil Survey Staff. 1996. Soil Survey Laboratory Methods Manual. Soil Survey Investigations Rep. 42. Version 3.0. U.S. Gov. Print, Washington DC.
33- Soltani S.M., and Samadi A. 2003. Different forms of phosphorus in some soils of Fars province and their relations with soil physic-chemical properties. Journal of Science and Technology of Agriculture and Natural Resources, 3: 119-127. (in Persian)
34- Taghizadeh Asl Z., Dordsipour E., Gholizadeh A.L., and Kiani F. 2009. Investigating the relation between plant available phosphorus and some of soil properties in soils of south of Gorganroud. Proceedings of the 11th Iranian Soil Science Congress, 12–13 Jul. 2009. Gorgan, Iran. (in Persian)
35- Tajik S., Ayoubi S., and Nourbakhsh F. 2012. Prediction of soil enzymes activity by digital terrain analysis: comparing artificial neural network and multiple linear regression models. Environmental Engineering Science, 29(8): 798-806.