Estimating air temperature of an alpine meadow on the Northern Tibetan Plateau using MODIS land surface temperature Gang Fu a,b,1 , Zhenxi Shen a,⇑ , Xianzhou Zhang a , Peili Shi a , Yangjian Zhang a , Jianshuang Wu a,b a Lhasa Plateau Ecosystem Research Station, Key Laboratory of Ecosystem Network Observation and Modeling, Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences, Beijing 100101, China b Graduate University of Chinese Academy of Sciences, Beijing 100049, China article info Article history: Received 26 April 2010 Revised 18 October 2010 Accepted 5 November 2010 Keywords: Air temperature Land surface temperature MODIS Alpine meadow Northern Tibetan Plateau abstract Meteorological data are scarce due to lack of meteorology stations in the Qinghai–Tibet Plateau. This often results in an imprecise estimation of air temperature. A linear estimation of air temperature of an alpine meadow on Northern Tibetan Plateau at heights of 1.5 m–2.1 m by using MODIS land surface temperature (LST) data was conducted in this study. The results showed that linear estimation of daily maximum and daytime mean air temperatures from MODIS LST data were not accurate enough (P > 0.01, R 2 < 0.10) during the growing season. In contrast, the linear relationships between daily maxi- mum and daytime mean air temperature and MODIS LST during the non-growing season were both sig- nificant (P < 0.01, R 2 > 0.40). MODIS LST data were accurate enough to linearly estimate daily minimum and nighttime mean air temperatures (P < 0.01, R 2 > 0.55). Moreover, derived LST from MODIS/Terra plat- form (MOD11A2) had higher accuracies than derived LST from MODIS/Aqua platform (MYD11A2) in lin- early estimating air temperatures mentioned above. Ó 2010 Ecological Society of China. Published by Elsevier B.V. All rights reserved. , , MODIS , (P > 0.01, R 2 < 0.10) ° , (P < 0.01, R 2 > 0.40) ° , MODIS (P < 0.01, R 2 > 0.55) ° MYD11A2 , MOD11A2 . Ó 2010 Ecological Society of China. Published by Elsevier B.V. All rights reserved. 1. Introduction Air temperature is commonly measured at standard meteoro- logical shelter height (i.e., approximately 2.0 m height above the ground) through meteorology observing station with high accuracy and temporal resolution [1,2]. Air temperature has been widely used in forests, grasslands and agriculture ecosystems [3–7], in order to estimate animal population dynamics [8], soil tempera- ture [9], frost [10], solar radiation [11], gross and net primary pro- duction [12–14], ecosystem respiration [15], and climate change [16–19]. Routine air temperature measurement from meteorological sta- tion is just in situ point data. Hence, different geographical interpo- lation methods, e.g. inverse distance weighting, kriging and spline methods, have been used to estimate air temperature between points in producing air temperature [20–25]. Besides, a variety of simulation approaches are used to model air temperature [1,26– 29]. You et al. [30] confirmed that spatial regression test method was better than inverse distance weighting method in estimating maximum and minimum daily air temperature across the USA. Chronopoulos et al. [31] found that artificial neural network mod- els had high accuracy with low mean absolute error values ranging from 0.82 to 1.72 °C in estimating air temperature. Cristóbal et al. [32] showed a good estimation of air temperature with an R 2 of 0.60 and an RMSE of 1.75 °C for daily temperatures and with an R 2 of 0.86 and an RMSE of 1.00 °C for monthly and annual temper- atures by combining remote sensing and GIS data. Recent studies have confirmed that LST is highly correlated with air temperature and can be used to linearly estimate daily maxi- mum and minimum air temperatures [33–36]. Mostovoy et al. [35] even indicated that daily maximum and minimum air temper- atures could be estimated from MODIS LST by using simple linear 1872-2032/$ - see front matter Ó 2010 Ecological Society of China. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.chnaes.2010.11.002 ⇑ Corresponding author. Tel.: +86 10 64888176; mobile: +86 13552550826. E-mail address: shenzx@igsnrr.ac.cn (Z.X. Shen). 1 Mobile: +86 13401139763. Acta Ecologica Sinica 31 (2011) 8–13 Contents lists available at ScienceDirect Acta Ecologica Sinica journal homepage: www.elsevier.com/locate/chnaes