برآورد دمای خاک از داده‌های هواشناسی با استفاده از مدل‌های یادگیری ماشین سریع، شبکه عصبی مصنوعی و رگرسیون خطی چندگانه

نوع مقاله : مقاله پژوهشی

نویسندگان

1 علوم و مهندسی خاک، دانشکده کشاورزی، دانشگاه شهید چمران اهواز، اهواز ، ایران

2 دانشیار، گروه علوم و مهندسی خاک، دانشکده کشاورزی، دانشگاه شهید چمران اهواز، ایران

چکیده

دمای خاک عامل کلیدی است که فرآیندها و خصوصیات فیزیکی، شیمیایی و بیولوژیکی خاک را کنترل می­کند؛ لذا بر کمیت و کیفیت تولید محصولات کشاورزی تأثیر می­گذارد. هدف از انجام این پژوهش برآورد دمای خاک با استفاده از پارامترهای هواشناسی به روش­های مختلف ماشین یادگیری بوده است. بدین منظور داده‌های هواشناسی و دمای خاک در عمق‌های 5، 10، 20، 30، 50 و 100 سانتی‌متری از 17 ایستگاه‌ سینوپتیک استان خوزستان مربوط به 25 سال (1994 تا 2018) گردآوری شد. مدل‌های شبکه عصبی پرسپترون چندلایه (MLPNN)، ماشین یادگیری سریع (ELM) و رگرسیون خطی چندگانه (MLR) برای برآورد دمای خاک استفاده شدند. داده‌های دمای هوا، سرعت باد، رطوبت نسبی هوا، فشار بخار، تبخیر و بارندگی به­عنوان ورودی برای آموزش مدل­ها استفاده شدند. نتایج نشان داد که تمام مدل­ها برآوردی بهتری از دمای لایه سطحی خاک (عمق 30-0 سانتی­متری) داشته و با افزایش عمق دقت­آن­ها کاهش می­یابد، به­طوری­که بهترین دقت برآورد دمای خاک مربوط به عمق 5 و کمترین آن مربوط به عمق 100 سانتی‌متری بود. نتایج نشان داد مدل­های MLR، MLPNN و ELM عملکردی مطلوب در مدل­سازی دمای خاک در تمام عمق­ها داشتند. برای مدل­های MLR، MLPNN، ELM به ترتیب مقادیر R2 از 864/0-700/0، 997/0-967/0 و 996/0-967/0، مقادیر RMSE از 823/2-557/2، 072/0-034/0 و 078/0-028/0 درجه سلسیوس و مقادیر MAE از 529/1-398/1، 063/0-023/0 و 065/0-023/0 درجه سلسیوس بود. نتایج نشان داد که دو مدل MLPNN و ELM دارای عملکردی تقریبا مشابه و بهتر از مدل رگرسیون خطی چندگانه بودند؛ اما به دلیل سرعت محاسبات بسیار بالای مدل ELM، پیشنهاد می­گردد از مدل MLPNN برای تخمین دمای نیمرخ خاک استفاده شود.

کلیدواژه‌ها

عنوان مقاله [English]

Estimating Soil Temperature from Metrological Data Using Extreme Learning Machine, Artificial Neural Network and Multiple Linear Regression Models

نویسندگان [English]

  • Hossein Zadmehr 1
  • Ahmad Farrokhian Firouzi 2
1 Soil Science and Engineering, Faculty of Agriculture, Shahid Chamran University of Ahvaz, Ahvaz, Iran
2 Associate Professor, Department of soil science, Faculty of Agriculture , Shahid Chamran University of Ahvaz, Iran
چکیده [English]

Soil temperature (Ts) is a key factor controlling the soil physical, chemical and biological properties and processes and consequently affects agricultural crop productions. The objective of this study was to estimate Ts from meteorological data using different machine learning methods. For this purpose, meteorological data and soil temperature at different depths (5, 10, 20, 30, 50 and 100 cm from the soil surface) for 25 years (1994-2018) were collected from 17 synoptic stations in Khuzestan province, Iran. Air temperature, wind speed, relative humidity, evaporation, precipitation, and vapor pressure were used as inputs to train the models. Multiple Linear Regression (MLR), Multilayer Perceptron Neural Network (MLPNN) and Extreme Learning Machine (ELM) models were used to predict soil temperature from metrological data. The results indicated that all models predicted temperature of the top layer (0-30 cm) better than the ones in sublayers. on the other hand, by increasing soil depth the accuracy of the models diminished; so that, the best and worst Ts predictions were belong to 5 cm and 100 cm depth, respectively. The results revealed that MLR, MLPNN and ELM models provided desirable performance in modeling Ts at all depths, with R2 values ranging 0.700-0.864, 0.967-0.997, and 0.967-99, RMSE values ranging 2.557–2.873, 0.034–0.072, and 0.028–0.078 °C, and MAE values ranging 1.398–1.529, 0.023–0.063, and 0.023–0.065 °C, respectively. Overall, the results showed that MLPNN and ELM models had approximately similar performance and better accuracy than MLR model. However, because of the high computational speed of the ELM model, it is recommended to use MLPNN model for estimation of soil profile Ts.

کلیدواژه‌ها [English]

  • Khuzestan province
  • Meteorological parameters
  • Multiple Linear Regression
  • Multilayer Perceptron Neural Network
  • Extreme Learning Machine

مراجع

Abyaneh, H.Z. Varkeshi, M.B., Golmohammadi, G., and Mohammadi, K. (2016). Soil temperature estimation using an artificial neural network and co-active neuro-fuzzy inference system in two different climates, Arabian Journal of Geosciences, 9(5), 377.
Alavipanah, S. K. (2006). Thermal remote sensing and its application in the earth sciences. Universality of Tehran Press, 523. ‏
Alizadeh, A. (2014). Soil Physics (7th ed).  Publishing Ferdowsi University of Mashhad.
Asadi, L., Hezarjaribi, A., Ghorbani, Kh., Zakernia, M. and AghaShariatmadari, Z. (2014). Estimating Soil Temperature Using Modern Methods of Data Analysis. Iranian Journal of Irrigation and Drainage, 8(1), 145-152
Bachmann, J., Horton, R., Grant, S. A., and van der ploeg R. R. (2002). Temperature dependence of water retention curves for wettable and water repellent soils. Soil Science Society of America Journal, 66(1), 44-52.
Chow, T. T., Long, H., Mok, H. Y., and Li, K. W. (2011). Estimation of soil temperature profile in Hong Kong from climatic variables. Energy and Buildings, 43(12), 3568-3575. ‏
Citakoglu, H. (2017). Comparison of artificial intelligence techniques for prediction of soil temperatures in Turkey. Theoretical and Applied Climatology, 130 (1-2), 545–556.
Deng, W. Y., Zheng, Q. H., Chen, L., and Xu, X. B. (2010). Research on extreme learning of neural networks. Chinese Journal of Computers, 33(2), 279-287.
Ding, S., Zhao, H., Zhang, Y., Xu, X., and Nie, R. (2015). Extreme learning machine: algorithm, theory and applications. Artificial Intelligence Review, 44(1), 103-115. ‏
Feng, Y., Cui, N., Hao, W., Gao, L., and Gong, D. (2019). Estimation of soil temperature from meteorological data using different machine learning models. Geoderma, 338, 67-77. ‏
Gill, J., and Singh, S. (2015). An efficient neural networks based genetic algorithm model for soil temperature prediction. International Journal of Emerging Technologies in Engineering Research (IJETER), 3, 1-5. ‏
Ismail Beekey, F. Zare haghi, D. Ghorbani, M. and Neyshabouri, M. (2017). Soil Temperature Estimation Based on Climatic Parameters Using Artificial Neural Network Models and Hybrid Firefly Algorithm (Case Study, East Azarbaijan Province). Fourth International Conference on Environmental Planning and Management, Tehran, Faculty of Environment, University of Tehran.
Jacobs, A. F., Heusinkveld, B. G., and Holtslag, A.A. (2011). Long-term record and analysis of soil temperatures and soil heat fluxes in a grassland area, The Netherlands. Agricultural and Forest Meteorology, 151(7), 774-780. ‏
Jaffari Golestan, M., Raeini-Sarjaz, M. and Ahmadi, M.Z. (2008). Estimation of soil depth temperatures using curve analysis and regression models for City of Sari (Iran). Journal of Agricultural Sciences and Natural Resources, 14(5), 112-123. (In Farsi)
Janbaz Ghobadi, GH., Yaghoubi, M., and Gholi Zade, B. (2017). Investigation of the Relationship between Meteorological Shelter Temperature with The temperature of different soil depths in air Humidity conditions. (Case Study: Sari City). Earth Science Research, 8(30), 80-94. (In Farsi)
Kim, S. and Singh, V.P. (2014). Modeling daily soil temperature using data-driven models and spatial distribution. Theoretical and Applied Climatology, 118 (3), 465–479.
Kisi, O., Tombul, M. and Kermani, M.Z. (2015). Modeling soil temperatures at different depths by using three different neural computing techniques. Theoretical and applied climatology, 121(1-2), 377-387.
Mazidi, A. and Fallah Zadeh F. (2011). Study the process of annual soil temperature in Yazd station. Geography and Development, 9(24), 39-50.
Mazidi, A., Mozafari, G.A., and Jahani, F. (2018). Determination of the Relationship between Temperature and Depth of Soil at Abade Station. The Second National Iranian Meteorological Conference, May 28, 2018, Ferdusi University, Mashhad, Iran, pp. 1-6.
Momeni, M., and Faal Qayyumi, A. (2010). Statistical analysis using SPSS. University of Tehran Press.
Nahvi, B., Habibi, J., Mohammadi, K., Shamshirband, S. and Razgan, O.S.A. (2016). Using selfadaptive evolutionary algorithm to improve the performance of an extreme learning machine for estimating soil temperature. Computers and Electronics in Agriculture, 124, 150-160.
Parsafar, N., and Marofi, S. (2011). Estimation of Soil Temperature from Air Temperature Using Regression Models, Artificial Neural Network and Adaptive Neuro-Fuzzy Inference System (Case Study: Kermanshah Region). Soil and Water Science, 3(21), 139-152. (In Farsi)
Plauborg, F. (2002). Simple model for 10 cm soil temperature in different soils with short grass. European Journal of Agronomy, 17, 173-179.
Sabziparvar, A. A., and Khoshhal, J. F. (2018). Comparison of Multi-Layer Perceptron artificial neural network and Linacre regression model performance for predicting daily minimum temperature (Case study: Kerman, Shiraz, Rasht and Hamedan). Iranian Journal of Geophysics, 12(3), 107-121. (In Farsi)
Samadianfard, S., Asadi, E., Jarhan, S., Kazemi, H., Kheshtgar, S., Kisi, O., and Manaf, A.A. (2018). Wavelet neural networks and gene expression programming models to predict short-term soil temperature at different depths. Soil and Tillage Research, 175, 37-50.
Tabari, H., Hosseinzadeh Talaee, P., and Willems, P. (2015). Short‐term forecasting of soil temperature using artificial neural network. Meteorological Applications, 22(3), 576-585.
‏ Yazdani, V., Ghahreman, B., Farhadi, G., and Nori, H. (2012). Modeling soil depth temperature by using methrological parameters. JournalofWaterandSoilConservation. 19(4) 1-15, (In Farsi)
Yusefi, A., Farrokhian Firouzi, A., and Aminzadeh, M. (2020). Effect of mulch on soil moisture, temperature and heat flux variation in the presence of shallow groundwater. Iranian Journal of Soil and Water Research, 50(9), 2213-2225. (In Farsi)
تحقیقات آب و خاک ایران
دوره 51، شماره 4
تیر 1399
صفحه 895-906

فایل ها

هم رسانی

ارجاع به این مقاله

آمار