ارزیابی انواع الگوریتم های پنجره مجزاء برای محاسبه دمای سطح زمین جهت تعیین بهترین الگوریتم برای تصاویر سنجنده مودیس

نوع مقاله: مقاله پژوهشی

نویسندگان

1 کارشناسی ارشد سنجش از دور و سیستم اطلاعات جغرافیایی، دانشکده برنامه‌ریزی و علوم محیطی، دانشگاه تبریز، تبریز، ایران

2 دانشیار گروه سنجش از دور و سیستم اطلاعات جغرافیایی، دانشکده برنامه‌ریزی و علوم محیطی ، دانشگاه تبریز ، تبریز، ایران

چکیده

پیشینه و هدف در سال­ های اخیر مطالعه تغییرات اقلیمی و همچنین تاثیرات آن­ها تبدیل به یک موضوع ثابت در عرصه­ های علمی بسیاری از کشورها شده است. یکی ازویژگی ­های اصلی این تغییرات، افزایش دمای هوا در طی 5 دهه اخیر نسبت به 500 سال گذشته است. به طوری که آمارها بیانگر افزایش یک درجه سانتی ­گراد در دمای هوا در طی 5 دهه اخیر هستند. به دمای تابشی پوسته زمین و به مقدار خالص انرژی که تحت شرایط اقلیمی درسطح زمین به توازن رسیده و به مقدار انرژی رسیده، گسیلمندی سطح، رطوبت و جریان هوای اتمسفر بستگی دارد، دمای سطح زمین می­گویند. دمای سطح زمین به­عنوان یکی از متغیرهای کلیدی در مطالعات اقلیمی و محیطی سطح زمین محسوب می ­شود. همچنین از پارامترهای اساسی در خصوصیات فیزیک سطح زمین در همه مقیاس­ ها از محلی تا جهانی است. در حال حاضر مهم­ترین منبع داده ­های اقلیمی ایستگاه ­های هواشناسی می­باشند و این ایستگاه­ ها آمار اقلیمی نقاط خاصی را ارائه می­ دهند در حالی­ که دما ممکن است در فواصل مختلف از این ایستگاه ­ها متحرک بوده و نسبت به ایستگاه مورد نظر کاهش یا افزایش داشته باشد. از این رو نیاز به تکنولوژی­ای که بتواند کاستی­ های ایستگاه­ های هواشناسی را در محاسبه دما در فواصل نمونه ­برداری و در مکان­های صعب العبور که امکان احداث ایستگاه هواشناسی وجود ندارد برطرف کند ضروری است. در سال ­های اخیر علوم جدیدی مانند سنجش از دور روش­های جدیدی را برای نظارت بر محیط و کسب، ارزیابی و تجزیه و تحلیل داده­ های محیطی فراهم آورده است و قابلیت ارائه طیف وسیعی از پارمترهای مربوط به محیط را دارا می باشد. این تکنولوژی به عنوان یک منبع مهم و فزاینده از اطلاعات برای مطالعه تغییرات اقلیمی که بر میزان دمای سطح زمین تأثیر مستقیم دارد مطرح می ­شود. در طی دو دهه گذشته برای محاسبه دمای سطح زمین 18 الگوریتم توسعه داده شده است که این الگوریتم­ ها در چهار دسته؛ مدل­ های وابسته به گسیلمندی، مدل­ های دو فاکتوره، مدل­ های پیچیده و مدل­ های بر مبنای رادیانس قرار دارند. بررسی نتایج مقایسه­ های انجام گرفته بین الگوریتم ­های مختلف نشان می ­دهد که الگوریتم­ های مختلف عملکرد متفاوتی را در موقعیت­ های مختلف با آب و هوای متفاوت جغرافیایی دارند. هدف از این تحقیق مقایسه انواع الگوریتم­ های محاسبه LST برای تصاویر سنجنده MODIS و تعیین بهترین الگوریتم برای استان آذربایجان شرقی می­باشد.
مواد و روش ها برای تبدیل ارزش های رقومی به تابش طیفی برای باندهای حرارتی تصاویر سنجنده MODIS استفاده قرار گرفت. تبدیل تابش طیفی به بازتاب طیفی با استفاده از رابطه پلانک، داده­ های حرارتی سنجنده MODIS، زمانی که توان تشعشعی آن­ ها حداکثر یک در نظر گرفته شوند، قابلیت تبدیل از تابش طیفی به بازتاب طیفی رادارند. در برآورد گسیلمندی سطحی از روش آستانه گذاری شاخص تفاضل نرمال شده گیاهی NDVI استفاده شد. جهت مشخص نمودن ویژگی ­های خاک در هر پیکسل و محاسبه میزان گسیلمندی و اختلاف گسیلمندی، توان تشعشعی به سه دسته تقسیم گردید؛ 0.2>NDVI به عنوان خاک خشک در نظر گرفته شده وتوان تشعشعی برای آن معادل 0.978 لحاظ می­گردد. 0.5NDVI نتایج و بحث در بین 18 الگوریتم محاسبه دمای سطح زمین برای تصاویر سنجنده MODIS به ترتیب؛ الگوریتم سوبرینو با مقدار RMSE، 1.79 بیشترین دقت، الگوریتم کول کاسلیس و پراتا با مقدار RMSE، 2.58 در جایگاه دوم و همچنین الگوریتم های سالیسبوری و سوبرینو با مقدار RMSE، 2.79 جایگاه سومی را برای محاسبه LST در بین سایر الگوریتم ها دارا می ­باشند. الگوریتم کیین با مقدار RMSE، 5.28 کم­ترین دقت را برای محاسبه LST به خود اختصاص داده است.
نتیجه ­گیری بررسی اطلاعات بدست آمده از مقایسه الگوریتم ­های پنجره مجزاء بیانگر تبعیت کلی دماهای محاسبه شده از شرایط توپوگرافی منطقه است، به طوری که تقریباً کمترین مقادیر درجه حرارت در تمام الگوریتم ­ها مربوط به قسمت ­های با ارتفاع بیشتر (کوهستانی) و پوشش سبز منطقه است و مقادیر دما در نواحی دارای ارتفاع پایین و فاقد پوشش گیاهی متراکم افزایش یافته است.

کلیدواژه‌ها


عنوان مقاله [English]

Evaluating the types of split window algorithms for calculating the land surface temperature to determine the best algorithm for MODIS sensor images

نویسندگان [English]

  • Mohammad Kazemi Garajeh 1
  • Behnam Salmani 1
  • Bakhtiar Feizizadeh 2
1 MSc of Remote Sensing and Geographical Information System, Faculty of Planning and Environmental Sciences, University of Tabriz, Tabriz, Iran
2 Associate Professor, Department of Remote Sensing and Geographical Information System, Faculty of Planning and Environmental Sciences, University of Tabriz, Tabriz, Iran
چکیده [English]

Background and ObjectiveIn recent years, the study of climate changes as well as their effects, has become a constant topic in the scientific fields of many countries. One of the main features of these changes is the increase in air temperature over the last 5 decades compared to the last 500 years. Statistics show an increase of one degree centigrade in air temperature over the last 5 decades. The land surface temperature means the radiant temperature of the earth's crust and the amount of pure energy that is balanced on the earth's surface under climatic conditions and depends on the reached the amount of energy, surface emissivity, humidity and atmospheric airflow. Land surface temperature is considered as one of the key variables in climate and environmental studies of the Earth’s surface. It is also one of the basic parameters in the physical features of  the earth's surface at all scales from local to global. Currently, the most important sources of climatic data are meteorological stations, and these stations provide climatic statistics for certain points, while the temperature may alter at different intervals stations and decrease or increase compared to the desired station. Therefore, it is necessary to have a technology that can eliminate the shortcomings of meteorological stations in calculating the temperature at sampling intervals and in impassable places where it is not possible to build a meteorological station. In recent years, new sciences such as remote sensing have provided new ways to monitor the environment and acquire, evaluate, and analyze environmental data, and can provide a wide range of parameters relating to the environment. This technology is considered as an important and increasing source of information for studying climate change that has a direct impact on global warming. Over the past two decades, 18 algorithms have been developed to calculate the land surface temperature. These algorithms fall into four categories: emissivity-dependent models, two-factor models, complex models, and radio-based models. The results of the comparisons between different algorithms shows that different algorithms perform differently in different situations with different geographical climates. Therefore, the present study aims to compare the types of LST calculation algorithms for MODIS sensor images and determine the best algorithm for East Azarbaijan province.
Materials and Methods Convert digital numbers (DN) to spectral radiation. The following equation was used to convert the numerical values to spectral radiation for thermal bands of MODIS sensor images. Planck's equation was used to convert spectural radation to spectral reflection when the radiant power of thermak data of MODIS sensor is considered to be a maximum of one. In order to estimate the surface emissivity, the Normalized Difference Vegetation Index (NDVI) thresholding method is used. The radiant power is divided into three categories to determine the soil characteristics in each pixel and to calculate the emissivity rate and emissivity difference; 0.2>NDVI, it is considered as dry soil and its radiant power is considered to be equal to 0.978. 0.5 NDVI, it is related to pixels with higher vegetation density and its radiant power is considered 0.985. 0.5>NDVI<0.2, it is based on a combination of pixels relating to vegetation and soil and the radiant power for them can be calculated. The vegetation ratio, that its value can be calculated. The value of each scientific finding depends on its accuracy. To compare the obtained results from the algorithms used to calculate the land surface temperature with the recorded temperature in meteorological station.
Results and DiscussionThe results of the present study show that among the 18 algorithms for the land surface temperature estimation for MODIS sensor images, the Sobrino algorithm with RMSE value of 1.79 has the highest accuracy, Cole Casillas and Prata algorithm with RMSE value of 2.85 is in the second position, and also the Salisbury and Sobrino algorithms with RMSE values of 2.39 have the third place for LST calculation among the other algorithms. The Qin algorithm with a RMSE value of 5.28 has the lowest accuracy for LST estimation.
Conclusion A review of the data obtained from comparing split-window algorithms shows the overall compliance of the calculated temperatures with the topographic conditions of the region, so that almost the lowest temperature values in all algorithms are related to the parts having more height (mountainous) and green cover of the region and also, temperature values have risen in low-lying areas lacking dense vegetation.

کلیدواژه‌ها [English]

  • Land surface temperature (LST)
  • Split window algorithms (SW)
  • MODIS sensor
  • East Azarbaijan Province
­Alsdorf DE, Rodríguez E, Lettenmaier DP. 2007. Measuring surface water from space. Reviews of Geophysics, 45(2): 1-24. doi:https://doi.org/10.1029/2006RG000197.

Asadzadeh A, Faith H, Shawl M. 2015. Spatial Inequalities in the Development of the Agricultural Sector of East Azerbaijan Province. Journal of Space Economics and Rural Development, 4(2): 41-45. (In Persion)

Alavi Panah SK. 2016. Thermal Remote Sensing and its Application in Earth Sciences, Third Edition, University of Tehran Press, 666 p. (In Persion)

Bakhtiari B, Delgarm S, Rezazadeh M. 2016. Selecting the most appropriate split-window algorithm for land surface temperature estimation using MODIS sensor (Case study: Kerman plain). Journal of Water and Soil Conservation, 23(2): 81-98. (In Persion)

Benali A, Carvalho AC, Nunes JP, Carvalhais N, Santos A. 2012. Estimating air surface temperature in Portugal using MODIS LST data. Remote Sensing of Environment, 124: 108-121. doi:https://doi.org/10.1016/j.rse.2012.04.024.

Bulivoury RE, Hartford RA, Eidenshink JC. 1993. Using NDVI to assess departure from average greenness and its relation to the fire business. Intermountain Research Station: USDA Forest Service, 8: 121-137.

Becker F, Li Z-L. 1990. Towards a local split window method over land surfaces. Remote Sensing, 11(3): 369-393. doi:https://doi.org/10.1080/01431169008955028.

Cao L, Li P, Zhang L, Chen T. 2008. Remote sensing image-based analysis of the relationship between urban heat island and vegetation fraction. Paper presented at the The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Vol. XXXVII. Part B7. Beijing 2008, 1379-1384.

Carlson TN, Ripley DA. 1997. On the relation between NDVI, fractional vegetation cover, and leaf area index. Remote Sensing of Environment, 62(3): 241-252. doi:https://doi.org/10.1016/S0034-4257(97)00104-1.

Chehbouni A, Lo Seen D, Njoku EG, Monteny BM. 1996. Examination of the difference between radiative and aerodynamic surface temperatures over sparsely vegetated surfaces. Remote Sensing of Environment, 58(2): 177-186. doi:https://doi.org/10.1016/S0034-4257(96)00037-5.

Cole A, Menenti M, Feddes R, Holtslag A. 1994. A remote sensing surface energy balance algorithm for land (SEBAL) 1 Formulation. Journal of Hydrology, 212(3): 198-212.

Coll C, Caselles V, Sobrino JA, Valor E. 1994. On the atmospheric dependence of the split-window equation for land surface temperature. Remote Sensing, 15(1): 105-122. doi:https://doi.org/10.1080/01431169408954054.

Eleftheriou D, Kiachidis K, Kalmintzis G, Kalea A, Bantasis C, Koumadoraki P, Spathara ME, Tsolaki A, Tzampazidou MI, Gemitzi A. 2018. Determination of annual and seasonal daytime and nighttime trends of MODIS LST over Greece - climate change implications. Science of The Total Environment, 616-617: 937-947. doi:https://doi.org/10.1016/j.scitotenv.2017.10.22.

Eskandari S. 2019. Comparison of different algorithms for preparing land cover map in sensitive habitats of Zagros using Sentinel 2 satellite image (Case study: part of Ilam province). Journal of RS and GIS for Natural Resources, 10(1): 72-87. (In Persion)

Emami H, Mojarradi B, Safari A. 2016. Presenting a method for assessing the accuracy and validation of land surface temperature from remote sensing data: a case study of Fars province. Journal of Mapping Science and Technology, 6(1): 1-17. (In Persion)

Feizizadeh B, Blaschke T, Nazmfar H, Akbari E, Kohbanani HR. 2013. Monitoring land surface temperature relationship to land use/land cover from satellite imagery in Maraqeh County, Iran. Journal of Environmental Planning and Management, 56(9): 1290-1315. doi:https://doi.org/10.1080/09640568.2012.71788.

Faizizadeh B, Dideban Kh, Gholamnia Kh. 2016. Estimation of land surface temperature using Landsat 8 satellite images and a split-window algorithm (Case study: Mahabad basin). Journal of Sepehr, 25(98): 171-181. (In Persion)

Franc G, Cracknell A. 1994. Retrieval of land and sea surface temperature using NOAA-11 AVHRR· data in north-eastern Brazil. International Journal of Remote Sensing, 15(8): 1695-1712. doi:https://doi.org/10.1080/01431169408954201.

Gillies RR, Carlson TN. 1995. Thermal remote sensing of surface soil water content with partial vegetation cover for incorporation into climate models. Journal of Applied Meteorology, 34(4): 745-756. doi:https://doi.org/10.1175/1520-0450.

Ghaffarian Malmiri H,  Zareh Khormizi H. 2016. Highlighting the time series of satellite data Earth surface temperature using time series harmonic analysis algorithm (HANTS) algorithm. RS and GIS for Natural Resources, 8(3): 37-55. (In Persion)

Hashemi Darreh Badami S, Nouraei Sefat A, Karimi S, Nazari S. 2015. Analysis of the development process of urban thermal islands in relation to land use /cover change using Landsat image time series. RS and GIS for Natural Resources, 6(3): 15-28. (In Persion)

Jin M, Dickinson RE. 2010. Land surface skin temperature climatology: Benefitting from the strengths of satellite observations. Environmental Research Letters, 5(4): 044004.

Kou X, Jiang L, Bo Y, Yan S, Chai L. 2016. Estimation of land surface temperature through blending MODIS and AMSR-E data with the Bayesian maximum entropy method. Remote Sensing, 8(2): 105. doi:https://doi.org/10.3390/rs8020105.

Luterbacher J, Dietrich D, Xoplaki E, Grosjean M, Wanner H. 2004. European seasonal and annual temperature variability, trends, and extremes since 1500. Science, 303(5663): 1499-1503. doi:https://doi.org/10.1126/science.1093877.

Liu Y, Yamaguchi Y, Ke C. 2007. Reducing the discrepancy between ASTER and MODIS land surface temperature products. Sensors, 7(12): 3043-3057. doi:https://doi.org/10.3390/s7123043.

Latif MS. 2014. Land Surface Temperature Retrival of Landsat-8 Data Using Split Window Algorithm-A Case Study of Ranchi District. International Journal of Engineering Development and Research, 2(4): 2840-3849.

Khorchani M, Vicente-Serrano SM, Azorin-Molina C, Garcia M, Martin-Hernandez N, Peña-Gallardo M, El Kenawy A, Domínguez-Castro F. 2018. Trends in LST over the peninsular Spain as derived from the AVHRR imagery data. Global and Planetary Change, 166: 75-93. doi:https://doi.org/10.1016/j.gloplacha.2018.04.006.

Kerr YH, Lagouarde JP, Imbernon J. 1992. Accurate land surface temperature retrieval from AVHRR data with use of an improved split window algorithm. Remote Sensing of Environment, 41(2): 197-209. doi:https://doi.org/10.1016/0034-4257(92)90078-X.

Mildrexler DJ, Zhao M, Running SW. 2011. A global comparison between station air temperatures and MODIS land surface temperatures reveals the cooling role of forests. Journal of Geophysical Research: Biogeosciences, 116(G3). doi:https://doi.org/10.1029/2010JG001486.

Mao K, Qin Z, Shi J, Gong P. 2005. A practical split‐window algorithm for retrieving land‐surface temperature from MODIS data. International Journal of Remote Sensing, 26(15): 3181-3204. doi:https://doi.org/10.1080/01431160500044713.

Neteler M. 2010. Estimating daily land surface temperatures in mountainous environments by reconstructed MODIS LST data. Remote sensing, 2(1): 333-351. doi:https://doi.org/10.3390/rs1020333.

Ottlé C, Vidal-Madjar D. 1992. Estimation of land surface temperature with NOAA 9 data. Remote Sensing of Environment, 40(1): 27-41.

Prata AJ. 1993. Land surface temperature from the advanced very high resolution radiometer and the along-track scanning radiometer. Journal of Geophysical Research. 98: 16689-16702.

Price JC. 1984. Land surface temperature measurements from the split window channels of the NOAA 7 advanced very high resolution radiometer. Journal of Geophysical Research Atmosphere. 89 (D5): 7231-7237.

Qin Z, Li W, Chen Z, Tang H. 2004. Land surface emissivity estimation for LST retrieval from Landsat TM6 data. Remote Sensing for Land and Resources, 3: 28-32.

Qin Z, Dall'Olmo G, Karnieli A, Berliner P. 2001. Derivation of split window algorithm and its sensitivity analysis for retrieving land surface temperature from NOAA‐advanced very high resolution radiometer data. Journal of Geophysical Research: Atmospheres, 106(D19): 22655-22670. doi:https://doi.org/10.1029/2000JD900452.

Rott H. 2000. Physical principles and technical aspects of remote sensing. In:  Remote sensing in hydrology and water management. Springer, pp 15-39. https://doi.org/10.1007/1978-1003-1642-59583-59587_59582.

Sabziparvar A, Fakharizadeh Shirazi A, Nazem Sadat S, Rezaei Y. 2016. Land surface temperature validation obtained from satellite images of MODIS and Landsat-5 (Case study: wheat fields of Marvdasht plain). Journal of Water and Soil Conservation, 23(2): 21-43. (In Persion)

Salehi N, Ekhtesasi MR, Talebi A. 2019. Predicting the trend of land use change using the Markov chain model (Case study: Ramsar Saffarude). Journal of RS and GIS for Natural Resources, 10(1): 106-121. (In Persion)

Santamouris M. 2013. Using cool pavements as a mitigation strategy to fight urban heat island-A review of the actual developments. Renewable and Sustainable Energy Reviews, 26: 224-240. doi:https://doi.org/10.1016/j.rser.2013.05.047.

Sun YJ, Wang JF, Zhang RH, Gillies RR, Xue Y, Bo YC. 2005. Air temperature retrieval from remote sensing data based on thermodynamics. Theoretical and Applied Climatology, 80(1): 37-48. doi:10.1007/s00704-004-0079-y.

Sun AY. 2013. Predicting groundwater level changes using GRACE data. Water Resources Research, 49(9): 5900-5912. doi:https://doi.org/10.1002/wrcr.20421.

Sobrino JA, Raissouni N, Li Z-L. 2001. A Comparative Study of Land Surface Emissivity Retrieval from NOAA Data. Remote Sensing of Environment, 75(2): 256-266. doi:https://doi.org/10.1016/S0034-4257(00)00171-1.

Sobrino J, Caselles V. 1991. A methodology for obtaining the crop temperature from NOAA-9 AVHRR data. International Journal of Remote Sensing, 12(12): 2461-2475. doi:https://doi.org/10.1080/01431169108955280.

Sobrino J, Coll C, Caselles V. 1991. Atmospheric correction for land surface temperature using NOAA-11 AVHRR channels 4 and 5. Remote Sensing of Environment, 38(1): 19-34. doi:https://doi.org/10.1016/0034-4257(91)90069-I.

Salisbury CM. 1997. Retrieving land-surface temperature from satellites. UCSB MODIS LST Group mom page. Retrieved from: http://www.icess.ucsb.edu/esrg/sum97/ studentEss.1997/cleo_Salisbury/cleo_final_ paper.html.

Tang B-H, Shao K, Li Z-L, Wu H, Tang R. 2015. An improved NDVI-based threshold method for estimating land surface emissivity using MODIS satellite data. International Journal of Remote Sensing, 36(19-20): 4864-4878. doi:https://doi.org/10.1080/01431161.2015.1040132.

Ulivieri C, Castronuovo MM, Francioni R, Cardillo A. 1994. A split window algorithm for estimating land surface temperature from satellites. Advances in Space Research, 14(3): 59-65. doi:https://doi.org/10.1016/0273-1177(94)90193-7.

Williamson SN, Hik DS, Gamon JA, Jarosch AH, Anslow FS, Clarke GKC, Scott Rupp T. 2017. Spring and summer monthly MODIS LST is inherently biased compared to air temperature in snow covered sub-Arctic mountains. Remote Sensing of Environment, 189: 14-24. doi:https://doi.org/10.1016/j.rse.2016.11.009.

Wan Z, Zhang Y, Zhang Q, Li Z-l. 2002. Validation of the land-surface temperature products retrieved from Terra Moderate Resolution Imaging Spectroradiometer data. Remote Sensing of Environment, 83(1): 163-180. doi:https://doi.org/10.1016/S0034-4257(02)00093-7.

Valizadeh Kh, Gholamnia Kh, Einali G, Mousavi M. 2016. Estimation of land surface temperature and extraction of thermal islands using split-window algorithm and multivariate regression analysis (case study of Zanjan city). Journal of Urban Research and Planning, 8(31): 35-50. (In Persion)

Zhao S, Qin Q, Yang Y, Xiong Y, Qiu G. 2009. Comparison of two split-window methods for retrieving land surface temperature from MODIS data. Journal of Earth System Science, 118(4): 345. doi:10.1007/s12040-009-0027-4.