ارزیابی دقت روش شبکه عصبی در برآورد سرمای دیررس بهاره در مناطق پسته‌کاری شهر کرمان

نوع مقاله : مقالات پژوهشی

نویسندگان

• مبینا عبدالهی فوزی ¹
• بهرام بختیاری ¹
• کورش قادری ²

¹ بخش علوم و مهندسی آب، دانشکده کشاورزی، دانشگاه شهید باهنر کرمان، کرمان، ایران

² بخش علوم و مهندسی آب، دانشکده کشاورزی، دانشگاه شهید باهنر کرمان، کرمان، ایرانن

https://doi.org/10.22067/jsw.2024.87064.1392

چکیده

سرمای دیررس بهاره تأثیر قابل‌توجهی بر اندام­های آسیب­پذیر گیاهان می­گذارد. این رویداد در آسیا، آمریکای شمالی و اروپا بیش از سایر مخاطرات مرتبط با آب‌وهوا باعث زیان اقتصادی به کشاورزی شده است. همچنین این پدیده باعث کاهش عملکرد محصول در ایران شده است. آخرین آمار منتشر شده از سوی سازمان خواربار و کشاورزی ملل متحد (فائو) نشان می‌دهد که ایران یکی از بزرگ‌ترین تولیدکنندگان محصول پسته در جهان می­باشد. استان کرمان سهم زیادی از سطح زیر کشت محصول پسته را به خود اختصاص داده است. خسارت سرمازدگی بهاره در پسته باعث کاهش عملکرد محصول در چند سال اخیر شده است. یک اصل مهم در مطالعه سرمازدگی، برآورد این پدیده است. در این تحقیق از روش شبکه عصبی مصنوعی مدل FFBP برای برآورد سرمای دیررس بهاره در محصول پسته شهرستان کرمان استفاده شد. بدین‌منظور داده‌های روزانه ایستگاه سینوپتیک شهر کرمان از سازمان هواشناسی کشور در بازه زمانی 2000-2020 اخذ شد. این داده­ها شامل میانگین، بیشینه و کمینه دما، رطوبت نسبی، سرعت باد، فشار بخار اشباع و ساعات آفتابی می­باشد. پنج ترکیب مختلف از این متغیرها به‌عنوان ورودی در روش شبکه عصبی برای مدل‌سازی دماهای کمینه در نظر گرفته شد. در نهایت ترکیب 8 متغیره­ای از بین مدل­ها انتخاب گردید و شبیه‌سازی مقادیر دمای کمینه و محاسبه ویژگی­های سرمای دیررس بهاره با آن انجام شد. عملکرد این روش با استفاده از شاخص‌های آماری ضریب تعیین، ریشه میانگین مربعات خطا، میانگین خطای انحراف و ضریب نش­ساتکلیف ارزیابی شد. بررسی نتایج مدل­سازی نشان داد با کاهش تعداد متغیرها دقت مدل­ها کاهش می­یابد. مدل M₁ با کمترین مقدار RMSE و بیشترین مقدار R² در بین سایر مدل­ها عملکرد بهتری داشت. پس از شبیه­سازی با روش شبکه عصبی مقادیر شاخص‌های 963/0R²= و صفر=MBE حاصل شد که نشان‌دهنده ارتباط قوی با داده‌های واقعی بود. علاوه بر آن، مقدار شاخص­های 027/0= RMSE و 966/0NSE= کارایی بالای مدل را در برآورد نشان داد. بررسی میانگین دمای سالانه نشان داد نوسانات دما در بازه زمانی 10-31 مارس در مقایسه با ماه­های آوریل و می زیاد می­باشد. کاهش محسوس میانگین دمای سالانه در سال­های 2000، 2006 و 2020 در این بازه نسبت به دیگر سال­ها بیشتر بود. در ماه آوریل نیز سال­های 2001، 2005، 2006، 2009، 2016 و 2019 کاهش دمای محسوسی داشتند. در ماه می باتوجه به میانگین دمای کمینه بین 10 تا 14 درجه سلسیوس احتمال سرمازدگی کمتری نسبت به ماه مارس و آوریل وجود داشت. نتایج نشان داد تعداد روزهای یخبندان بهاره مشاهداتی و برآوردی حاصل از روش شبکه عصبی انطباق خوبی با یکدیگر داشتند. این روش در برآورد تعداد روزهای بحرانی (دماهای کمینه  کمتر و مساوی 2 درجه سلسیوس) نیز دقت قابل قبولی داشت. همچنین سال­های 2000، 2004، 2005، 2012، 2015، 2019 و 2020 بیشترین تعداد روزهای یخبندان بهاره و سال­های 2006، 2016 و 2019 بیشترین تعداد روزهای بحرانی را در دو دهه­ی اخیر دارا بودند. با بررسی نتایج می­توان گفت روش شبکه عصبی مصنوعی در برآورد دمای کمینه و ویژگی­های مرتبط با سرمای دیررس بهاره از دقت بالایی برخوردار است.

کلیدواژه‌ها

• پسته
• دمای بحرانی
• سرمازدگی بهاره
• شبکه عصبی مصنوعی
• فراوانی یخبندان

موضوعات

• هواشناسی کشاورزی

عنوان مقاله [English]

Evaluation of Accuracy of Neural Network Method for Late Spring Frost Estimating in Pistachio Growing Areas of Kermann

نویسندگان [English]

• M. Abdollahi Fuzi ¹
• B. Bakhtiari ¹
• K. Qaderi ²

¹ Water Structural Engineering and Associate Professor, Water Engineering Department, Faculty of Agriculture, Shahid Bahonar University of Kerman, Kerman, Iran, respectively.

² Water Structural Engineering and Associate Professor, Water Engineering Department, Faculty of Agriculture, Shahid Bahonar University of Kerman, Kerman, Iran, respectively.

چکیده [English]

Introduction
Spring frost is considered an important threat to agricultural products in high and middle latitudes. The damage caused by Late Spring Frosts (LSFs) significantly impacts vulnerable plant organs. This event has caused more economic losses to agriculture than any other climatic hazard in Asia, North America, and Europe. Also, these phenomena have contributed to low crop yields in Iran. The latest statistics released by the Food and Agriculture Organization of the United Nations (FAO) show that Iran is one of the largest producers of agricultural products and the world’s second-biggest producer of pistachios. Kerman province is one of the significant areas of pistachio production. This province has a large share of the pistachio word area plantation. Spring frost damage to pistachio crops has led to low yields in recent years. A key aspect of studying frost is the ability to accurately estimate its occurrence. In this study, artificial neural network methods have been used to estimate late spring frost in the pistachio crop of Kerman city.
 
Materials and Methods
In this study, the efficiency of this method was investigated in the estimation of minimum temperature. For this purpose, the daily data of the synoptic station of Kerman city were obtained from Iran Meteorological Organization from 2000 to 2020. Meteorological data including mean, maximum, and minimum temperatures, relative humidity, wind speed, saturated vapor pressure, and sunshine hours were used. Five different combinations of these variables was considered as input variables in artificial neural network method for minimum temperatures modeling. After entering data into network and modeling with each combination, RMSE and R² values were calculated. Finally, the combination of 8 variables including average and maximum temperature, the minimum temperature the previous day and two days prior, relative humidity, wind speed, saturated vapor pressure, and sunny hours were selected as the most suitable combination of variables. Subsequently, a simulation of minimum temperature values was conducted using 10% of the data. The performance of the methods was evaluated using statistical indices of coefficient of determination (R²), mean square of error (RMSE), Mean Bias Error (MBE), and Coefficient of Nash–Sutcliffe (NSE).
 
Results and Discussion
The accuracy of an analytical method is the degree of agreement between the test results generated by the method and the true value. Upon examining the models, the M1 model was identified as the best due to its lowest RMSE and higher R². ANN model results were evaluated using various performance measure indicators. The simulated outcome of the model indicated a strong association with actual data, where the correlation coefficient was above 0.95, and the MBE index was zero. Also, the RMSE value was positive and close to zero, and the N_SE value was above 0.75. Therefore artificial neural network method had high accuracy. In this study, mean annual minimum temperature was estimated using artificial neural network models (from March 10 to May 20). Comparison between the observed and calculated data showed that these data were in good agreement. Also, the results showed that temperature fluctuations were high between March 10 and March 31. From 2011 to 2017, an almost uniform temperature trend has been observed between March 10 and March 31. However, the years 2000, 2006, and 2020 showed a noticeable decrease in temperature. From 2018 to 2020, this trend of temperature reduction continued. In April, the temperature values were between 7 and 10 degrees Celsius. The years 2001, 2005, 2006, 2009, 2016, and 2019 had a noticeable decrease in temperature. In May, the mean minimum temperature was between 10 and 14 degrees Celsius. Therefore, the probability of frost occurrence in early-flowering cultivars was higher in late March than in April and May. The years 2000, 2004, 2005, 2012, 2015, 2019 and 2020 had the highest number of frost days in the last two decades.
 
Conclusion
The results showed that the artificial neural network method had a high performance in estimating the minimum temperature. The values of the statistical indicators were R²=0.963, RMSE=0.027^oC, MBE= 0 and NSE=0.966 respectively. In addition, the ANN method performed well in estimating the number of critical frost days for pistachio crops. The results showed that, although reducing the amount of input data in models decreases their output precision, data-driven methods can still be useful tools for minimum temperature estimation.

کلیدواژه‌ها [English]

• Abundance of frost
• ANN
• Critical temperature
• Pistachio
• Spring frost

©2024 The author(s). This is an open access article distributed under Creative Commons Attribution 4.0 International License (CC BY 4.0)

مراجع

1. Abdollahi Fuzi, M., Bakhtiari, B., & Qaderi, K. (2023). A systematic review of research on late spring frost and its estimation methods. Nivar, 47(120-121), 105-126. (In Persian with English abstract). https://doi.org/10.30467/ 2023.379316.1235
2. Abdollahi Fuzi, M., Bakhtiari, B., & Qaderi, K. (2023). A review of research on frost and chilling from 1954 to 2020 (Bibliometric Analysis). Water Management in Agriculture, 10(1), 131-146. (In Persian with English abstract). https://doi.org/1001.1.24764531.1402.10.1.9.7
3. Allen, R.G., Pereira, L.S., & Raes, D. (1998). Crop evapotranspiration-guidelines for computing crop water requirements-FAO Irrigation and Drainage Paper 56. FAO, Rome, 300(9),
4. Anandhi, A., Zion, M.S., Gowda, P.H., Pierson, D.C., Lounsbury, D., & Frei, A. (2013). Past and future changes in frost day indices in Catskill Mountain region of New York. Hydrological Processes, 27(21), 3094-3104. https://doi.org/10.1002/hyp.9937
5. Augspurger, C.K. (2009). Spring 2007 warmth and frost: Phenology, damage and refoliation in a temperate deciduous forest. Functional Ecology, 23, 1031–1039. https://doi.org/10.1111/j.1365-2435.2009.01587.x
6. Bascietto, M., Bajocco, S., Mazzenga, F., & Matteucci, G. (2018). Assessing spring frost effects on beech forests in Central Apennines from remotely-sensed data. Agricultural and Forest Meteorology, 250, 240-248. https://doi.org/10.1016/j.agrformet.2017.10.007
7. Buishand, T.A. (1982). Some methods for testing the homogeneity of rainfall records. Journal of Hydrology, 58, 11–27. https://doi.org/10.1016/0022-1694(82)90066-X
8. Chen, W., Liu, W., Huang, W., & Liu, H. (2017). Prediction of salinity variations in a Tidal Estuary using artificial neural network and three-dimensional hydrodynamic models. Computational Water, Energy, and Environmental Engineering, 6, 107-128. https://doi.org/10.4236/cweee.2017.61009
9. Cybenko, G. (1989). Approximation by superpositions of a sigmoidal function. Mathematics of Control, Signals, and Systems, 2(4), 303–314. https://doi.org/10.1007/BF02551274
10. Drepper, B., Bamps, B., & Gobin, A. (2022). Strategies for managing spring frost risks in orchards: effectiveness and conditionality-a systematic review. Environmental Evidence, 11(1), 29. org/10.1186/s13750-022-00281-z.
11. Entezari, A., Haddad Niya, J., Jafar Zade, M., & Korvandi, E. (2011). A MLP neural network which predict frosts in kermanshah province. Journal of Arid Regions Geographic Studies, 2(3), 59-71.
12. Fuentes, M., Campos, C., & García-Loyola, S. (2018). Application of artificial neural networks to frost detection in central Chile using the next day minimum air temperature forecast. Chilean Journal of Agricultural Research, 77, 327-338. https://doi.org/10.4067/S0718-58392018000300327
13. Gholipour, Y. (2007). Low temperature threshold and grow th degree day (GDD) for tw o pistachio cultivars. Journal of Applied Horticulture, 9(2), 150-152. https://doi.org/10.37855/jah.2007.v09i02.33
14. Gu, L., Hanson, P.J., Post, W.M., Kaiser, D.P., Yang, B., Nemani, R., Pallardy, S.G., & Meyers, T. (2008). The 2007 Eastern US spring freeze: Increased cold damage in a warming world? BioScience, 58, 253–262. https://doi.org/ 10.1641/b580311
15. Graczyk, D., & Szwed, M., (2020). Changes in the occurrence of late spring frost in Poland. Agronomy, 10(11), 1835. https://doi.org/10.3390/agronomy10111835
16. Gunhan, T., Demir, V., Hancioglu, E., & Hepbasli, A. (2005). Mathematical modeling of drying of bay leaves. Energy Conversion and Management, 46, 1667–1679.
17. Hopfield, J.J. (1982). Neural networks and physical systems with emergent collective computational abilities. Proceedings of the National Academy of Sciences, 79(8), 2554–2558. https://doi.org/10.1073/pnas.79.8.2554
18. Hernandez, G., Müller, G.,Villacampa, Y., Navarro-Gonzalez, F., & Luis, A. (2019). Predictive models of minimum temperatures for the south of Buenos Aires province. Science of The Total Environment, 699, 134280. https://doi.org/ 10.1016/j.scitotenv.2019.134280
19. Khalili, A. (2014). Quantitative evaluation of spring frost risk to agriculturial crops in Iran and modeling. Jornal of Agricultural Meteorology, 2(1), 17-31.
20. Khalili, A., Bazrafshan, J., & Cheraghalizadeh, M. (2022). A Comparative study on climate maps of Iran in extended de Martonne classification and application of the method for world climate zoning. Journal of Agricultural Meteorology, 10(1), 3-16. (In Persian with English abstract). https://doi.org/10.22125/agmj. 156309
21. Khatami Firouzabadi, S.M.A., Dehdashti Shahrokh, Z., Roshani, A., & Akhgari, A. (2016). A hybrid fuzzy approach using AHP and Topsis methods to prioritize the export target markets of pistachio in Iran, Yazd. Iranian Journal of Trade Studies, 20(79), 121-154.
22. Hufkens, K., Friedl, M.A., Keenan, T.F., Sonnentag, O., Bailey, A., O’Keefe, J., & Richardson, A.D. (2012). Ecological impacts of a widespread frost event following early spring leaf-out. Global Change Biology, 18, 2365–2377. https://doi.org/10.1111/j.1365-2486.2012.02712.x
23. Kreyling, J., Thiel, D., Nagy, L., Jentsch, A., Huber, G., Konnert, M., Beierkuhnlein, C. (2012). Late frost sensitivity of juvenile Fagus sylvatica differs between Southern Germany and Bulgaria and depends on preceding air temperature. European Journal of Forest Research, 131, 717–725. https://doi.org/10.1007/s10342-011-0544-y
24. Krause, P., Boyle, D.P., & Bäsel, F. (2005). Comparison of different efficiency criteria for hydrological model assessment. Advances in Geosciences, 5, 89–97. https://doi.org/10.5194/adgeo-5-89-2005
25. Larcher, W., & Bauer, H. (1981). Ecological significance of resistance to low temperature. In Physiological Plant Ecology I: Responses to the Physical Environment; Lange, O.L., Nobel, P.S., Osmond, C.B., Ziegler, H., Eds.; Springer: Berlin/Heidelberg, Germany, pp. 403–437.
26. Larcher, (1981). Effects of low temperature stress and frost injury on plant productivity. In Physiological Processes Limiting Plant Productivity; Johnson, C.B., Ed.; Butterworth: London, UK, pp. 253–269.
27. Littlewood, L.G., Clarke, R.T., Collischonn, W., Croke, B.F.W. (2007). Predicting daily streamflow using rainfall forecasts, a simple loss module and unit hydrographs: Two Brazilian catchments. Environmental Modelling and Software, 22, 1229-1239. https://doi.org/10.1016/j.envsoft.2006.07.004
28. Mesgari, E., Tavousi, T., Mahmoudi, P., & Jahanshahi, S.M.A. (2020). Evaluation of selected transfer functions of artificial neural network model for prediction of minimum temperature (Case Study: Sanandaj Station). Journal of Agricultural Meteorology, 8(1), 40-50.) In Persian with English abstract.( https://doi.org/10.22125/agmj.2020. 191817.1066
29. Moustris, K.P., Ziomas, I.C., & Paliatsos, A.G. (2010). 3-day-ahead forecasting of regional pollution index for the pollutants NO₂, CO, SO₂ and O₃ using artificial neural networks in Athens, Greece. Water Air Soil Pollut, 209, 29–43. https://doi.org/10.1007/s11270-009-0179-5
30. Moriasi, D.N., Arnold, J.G., Van Liew, M.W., Bingner, R.L., Harmel R.D., & Veith, T.L. (2007). Model evaluation guidelines for systematic quantification of accuracy in watershed simulations. Transactions of the ASABE, 50, 885-900. https://doi.org/10.13031/2013.23153
31. Motovilov, Y.G., Gottschalk, L., England, K., & Rodhe, A. (1999). Validation of distributed hydrological model against observations. Agricultural and Forest Meteorology, 99, 257-277. https://doi.org/10.1016/S0168-1923%2899%2900102-1
32. Nash, J.E., & Sutcliffe, J.V. (1970). River flow forecasting through conceptual models; Part I – a discussion of principles. Journal of Hydrology, 10, 282–290. https://doi.org/10.1016/0022-1694(70)90255-6
33. Ningre, F., & Colin, F. (2007). Frost damage on the terminal shoot as a risk factor of fork incidence on common beech (Fagus sylvatica ). Annals of Forest Science, 64, 79–86. https://doi.org/10.1051/forest:2006091
34. Noh, I., Doh, H.W., Kim, S.O.S.H., Kim, S., & Shin, Lee, S.J. (2021). Machine learning-based hourly frost-prediction system optimized for orchards using automatic weather station and digital camera image data. Atmosphere, 12(7), 846. https://doi.org/10.3390/atmos12070846
35. Sadorsky, P. (2006). Modeling and forecasting petroleum futures volatility. Energy Economics, 28, 467-488. https://doi.org/10.1016/j.eneco.2006.04.005
36. Shcherbakov, M.V., Brebels, A., Shcherbakov, N.L., Tuukov, A.P., Janovsky, T.A., & Kamaev, V.A. (2013). A survey of forecast error measures. World Applied Sciences Journal, 24, 171-176. https://doi.org/10.5829/idosi. wasj.2013.24.itmies.80032
37. Shonwiese, C.D. (1997). Statistical analysis of observed climate trends and statistical signal detection analysis, WMO Publ., No. 834, Geneva.
38. Snyder, R.L., & de Melo-Abreu, J.P. (2005). Frost Protection: Fundamentals, Practice and Economics.
39. Taghavi, S., & Omidzade, H. (2015). Spring frost prediction using neural network multilayer prediction (MLP) and it^,s impact on the public transporation in Khorramabad City. Environmental Based Territorial Planning (Amayesh), 8(28), 111-124. (In Persian)
40. Tajabadipour, A., Fattahi Moghadam, M.R., Zamani, Z., Nasibi, F., & Hokmabadi, H. (2018). Evaluation of physiological and biochemical changes of pistachio (Pistacia vera cv. Ahmad-Aghaii) on cold tolerant and sensitive rootstocks under freezing stress conditions. Journal of Horticultural Science, 32(3), 471-484. (In Persian)
41. Tran, T., Bateni, M., Jin Ki, S., & Vosoughifar, H. (2021). A review of neural networks for air temperature forecasting. Water, 13(9), 1294. https://doi.org/10.3390/w13091294
42. Vitasse, , Schneider, L., Rixen, C., Christen, D., & Rebetez, M. (2018). Increase in the risk of exposure of forest and fruit trees to spring frosts at higher elevations in Switzerland over the last four decades. Agricultural and Forest Meteorology, 248, 60–69. https://doi.org/10.1016/j.agrformet.2017.09.005
43. Woodward, F.I., & Williams, B.G. (1987). Climate and plant distribution at global and local scales. Vegetatio, 69, 189–197. https://doi.org/10.1007/BF00038700
44. Zohner, C.M., Mo, L., Renner, S.S., Svenning, J.C., Vitasse, Y., Benito, B.M., Ordonez, A., Baumgarten, F., Bastin, J.F., Sebald, V., Reich, P.B., Liang, J., Nabuurs, G.J., De-Migueln, S., Alberti, G., Antón-Fernández, C., Balazy, R., Brändli, U.B., Chen, H.Y.H., & Crowther, T.W. (2020). Late-spring frost risk between 1959 and 2017 decreased in North America but increased in Europe and Asia. Proceedings of the National Academy of Sciences of the United States of America, 117(22). https://doi.org/10.1111/geb.13088
45. Zolfaghari, H., Zahedi, GH., & Sajjadifar, T. (2012). Predicting last spring freezing day in West Northwest of Iran. Geography and Enviromental Sustainability, 2(4), 59-74. (In Persian with English abstract)