ارزیابی کاربرد مدل های شبکه عصبی مصنوعی، شبکه عصبی تطبیقی فازی و رگرسیون در پیش‌‌بینی کربن آلی ذره ای در مراتع خرابه سنجی ارومیه

نوع مقاله: مقاله پژوهشی

نویسندگان

1 دانشجوی کارشناسی ارشد مرتعداری، دانشکده منابع طبیعی، دانشگاه تربیت مدرس

2 دانشیار گروه مرتعداری، دانشکده منابع طبیعی، دانشگاه تربیت مدرس

3 دانشجوی کارشناسی ارشد آبخیزداری، دانشکده منابع طبیعی، دانشگاه تربیت مدرس

4 استادیار دانشکده منابع طبیعی، دانشگاه ارومیه

چکیده

کربن آلی خاک اثرات مفید­ی روی خواص شیمیایی­، فیزیکی و حرارتی خاک داشتهو همچنین روی فعالیت‌های بیولوژیکی خاک‌ها موثر است. کربن آلی ذره­اییکی از بخش های مهم ناپایدار مواد آلی می باشد و نقش قابل توجهی در کیفیت خاک و مدیریت سرزمینهای مرتعی دارد. در این تحقیق جهت برآورد دقیق کربن آلی ذره­ای خاک از مدل‌های شبکه عصبی مصنوعی (­­ANN)، شبکه عصبی تطبیقی- فازی(ANFIS)  و رگرسیون چند متغیره استفاده شد. جهت انجام تحقیق، 60 نمونه خاک از عمق 30- 0 سانتیمتری در میان 60 کوادرات یک متر مربعی که در طول 6 ترانسکت 100 متری در مراتع خرابه سنجی ارومیه مستقر شده بود، برداشت شد. خصوصیات خاک (­نیتروژن، رس، سیلت، کربن آلی، اسیدیته، هدایت الکتریکی و وزن مخصوص ظاهری خاک) اندازه­گیری شدند. شاخص های آماری RMSE و  CE جهت ارزیابی کارکرد مدل‌ها استفاده شدند. نتایج نشان داد  بر اساس معیارهای مجذور میانگین مربعات خطا و ضریب کارایی که در مدل رگرسیونی به ترتیب 16/0 و 41/0 و در مدل شبکه عصبی مصنوعی به ترتیب 11/0 و 65/0 و در مدل شبکه عصبی تطبیقی-فازی به ترتیب 06/0 و 79/0 می­باشند، مدل شبکه عصبی تطبیقی فازی (ANFIS) به عنوان ابزار قدرتمندتری در پیش­بینی کربن آلی ذره­ای خاک نسبت به آنالیز رگرسیون خطی چند­متغیره و شبکه عصبی مصنوعی عمل می­کند.

کلیدواژه‌ها


عنوان مقاله [English]

Evaluation of Artificial Neural Network (ANN), Adative Neuro-Fuzzy Inference System (ANFIS) and Regression Models in Prediction of Particulate Organic Matter-Carbon (POM-C) in the Rangelands Kharabe Sanji of Urmia

نویسندگان [English]

  • Behnam Bahrami 1
  • Ghasem Ali Dianati Tilaki 2
  • Saeid Khosro Beigi 3
  • Saeid Janizadeh 3
  • Javad Moetamedi 4
1 Graduate Student of Range Management, Faculty of Natural Resources, Tarbiat Modares University
2 Associate Professor, Faculty of Natural Resources, Tarbiat Modares University
3 M.sc. Student of Watershed Management, Faculty of Natural Resources, Tarbiat Modares University
4 Assistant Professor, Faculty of Natural Resources, Urmia University
چکیده [English]

Soil organic carbon has favorable effects on the chemical, physical and thermal properties of the soil as well as on the biological activities in the soil. Particulate organic matter-carbon (POM-C) is one of the important unstable elements in the soil organic matter has a considerable role in soil quality and rangeland management. In this research, in order to exact estimate of POM-C using ANN, ANFIS, Regression models were developed. Towards this attempt, 60 soil samples were taken from the depth of 0-30 cm of the soil within 60 quadrates of 1m2 of located along 6 transects of 100m in the rangelands Kharabeh Sangi of Urmia. Soil properties (Nitrogen, clay, silt, organic carbon, pH, EC, apparent specific weight of soil) were measured.  Statistic indicators RMSE, CE were used for performance evaluation of the models. The results showed RMSE and CE were calculated 0.16 and 0.41(in Regression Model), 0.11 and 0.65(in Artificial Neural Network Model), 0.06 and 0.79 (in Adative Neuro-Fuzzy Inference System Model), respectively. Also Adative Neuro-Fuzzy Inference System Model is considered as a strong tool in prediction of POM-C compared with Multivariate Linear Regression and Artificial Neural Network Models in the rangelands Kharabeh Sanji of Urmia.  

کلیدواژه‌ها [English]

  • soil properties
  • Organic carbon
  • modeling
  • efficiency coefficient
References

Allison LE. 1975. Organic carbon. In: Black CA, Evans DD, White JL, Ensminger LE, Clark FE. (Eds.), Methods of soil analysis, Part 2, Chemical and microbiological properties. American Society of Agronomy, Madison. 1367p.

Auobi Sh and Alizadeh MH. 2006. Soil surface attributes prediction using digital topographic model (Case Study: part of Mehr Watershed, Sabzevar, KhorasanProvince). Science and Technology of Agriculture and Natural Resources, 10 (2): 85-96.

Balan B, Mohaghegh S and Ameri S. 1995. State- of- Art- in permeability determination from well log data: Part 1- A comparative study, Model development. SPE. 30978: 17-25.

Bazartseren B, Hildebrandt G and Holz K. 2003. Short-term water level prediction using neural networks and neuro-fuzzy approach. Neuro computing, 55: 439-450.

Baiat varkesh M, Zare abiane H, Marofi S, Sabziparvar A and Soltani F. 2009. Simulates daily crop reference evapotranspiration using artificial intelligence methods and compared with experimental measurements of having cold semi-arid climate, Hamedan. Journal of Soil and Water Conservation Research, 16 (4): 79-83.

Blake GR and Hartge KH. 1986. Bulk density. P 363-375, In: Klute, A. Methods of soil analysis. Part 1. 2nd Ed. Agron. Monogr. 9. ASA. Madison. WI.

CambardellaCA and  Elliott ET. 1992. Particulate soil organic matter changes across a grassland cultivation sequence. American Journal of Soil Science, 56: 777-783.

Caudill M. 1987.  Neural networks primer: Part I, AI Expert.

Fajri F. 2009. The report on rangeland & vegetation cover feasibility studies in the kharabeh-sanji basin. Faculty of Natural Resources. University of Urmia.

Gee GW and Bauder JW. 1986. Particle size analysis. 383-411p, In: Methods of soil analysis. Part 1. 2nd  Ed. Klute, A. Agron. Monogr. 9. ASA. Madison. WI.

           

   
   
104

   
   
Handayani IP, Coyne MS, Barton C and Workman S. 2008. Soil carbon pools and aggregation follwing land restoration:BernheimForest. Ken. J. Enveron. Monitor. Restor. 4: 11-28.

 

Haynes RJ. 2005. Labile organic matter fraction as central components of the quality of agricultural soils: An overview. Adv. Agron. 85: 221-268.

Hecht R. 1987. Kolmogorov mapping, neural network existence theorem. 1st IEEE ICNN, 3. Sandiego.

Holmberg M, Forsius M, Starr M and Huttunen M. 2006. An application of artificial neural networks to carbon, nitrogen and phosphorus concentration in three boreal streams and impacts of climate change. International Society for Ecological Information 3rd Conference. Grottaferrata, Roma, Italy. 195: 51-60. 

Ingleby HR and Crowe TG. 2001. Neural network models for predicting organic matter content in Saskatchewan soils. Canadian Biosystems Engineering, 43:71-75.

Khanna T. 1990. Foundations of neural networks, Addison-Wesley Pub. Co.USA .

Merdun H, Ozer T, Meral R and Apan M. 2006. Comparison of artificial neural network and regression pedotransfer functions for prediction of soil water retention and saturated hydraulic conductivity. Soil and Tillage Research, 90: 108-116.

Memarian Fard M and Beigi Hrchgany H. 2009. Comparison of artificial neural network models and regression transfer functions to predict soil exchange capacity in Chaharmahal and BakhtiariProvince. Journal of Soil and Water, 4: 90-99.

Moghaddamnia A, Ghafari Gousheh M, Piri J, Amin S and Han D. 2009. Evaporation estimation using artificial neural networks and adaptive neurofuzzy inference system techniques. Advances in Water Resources, 32: 89-97.

Norani V and Salehi K. 2008. Rainfall-runoff modeling using adaptive fuzzy neural network and comparison with neural network and fuzzy inference. National Congress on Civil Engineering, TehranUniversity. Proceedings of the Fourth National Congress of Civil Engineering.

Parasurman K, Elshorbagy A and Si B. 2006. Estimating saturated hydraulic conductivity in spatially variable fields using neural network in Ensembles. SSSA. J. 70: 1851-1859.

Parton WJ, Schmel DS, Cole CV and Ojima DS. 1987. Analysis of factors controlling soil organic matter levels in Great Plains grasslands. A. J. Soil Sci. 51: 1173-1179.

Parsafar NA and Marofi S. 2011. Estimated temperatures at depths using network neural networks Fuzzy (case study: Kermanshah region). J. Soil and Water Sci. 21(3): 21-22.

Pilevari A, Auobi Sh and  Khademi H. 2010. Comparison of artificial neural network and multiple linear regression analysis to predict soil organic carbon data to the ground. J. Soil and Water, 24 (6): 1151-1163.

Sabzi parvar A and Beiatorkeshi M. 2010. Assess the accuracy of fuzzy artificial neural network, neurotropic solar radiation simulation. Iranian J. Physic. Res. 4(10): 347-536.

Sinowski W and Auerswald K. 1999. Using relief parameters in a discriminate analysis to stratify geological areas with different spatial variability of soil properties. Geoderma, 89: 113-128.

SkullbergU. 1991. Seasonal Variation of pH h2o and pH cacl2 in centimeter- layers of Mor Humus in a Picea Abies (L.) Karst stand. SwedenUniversity of Agri Sci, Dep.Forest Site Res.

Somaratne S, Seneviratne G and CoomaraswamyU. 2005. Prediction of soil organic carbon  across different landuse patterns: A neural  network approach. SSSA. J. 69: 1580-1589.

Spaccini R, Mbagwu JC, Igwe CA, Conte P and Piccolo A. 2004. Carbohydrate and aggregation  in lowland soil of Nigeria as influenced by organic input. Soil and Tillage Res. 75: 161-172.

Sumfleth K and Duttmann R. 2008. Prediction of soil property distribution in paddy soil landscapes using terrain data and satellite information as indicators. Eco. Indicators, 8: 485–501.

Thompson JA and Kolka RK. 2005. Soil carbon storage estimation in a forested watershed using quantitive soil-landscape modeling. SSSA. J. 69: 1086-1093.

Zahedi Gh. 1998. Relation between vegetation and soil characteristics in a mixed hard wood stand. Academic press, GhentUniversity (Belgium), 319p.

Zevebergen LW and Thorne CR. 1987. Quantitative analysis of land surface topography. Earth Surface Processes Landforms, 12: 47-56.