Avaliação do produto MOD11A2 para o monitoramento da temperatura do dossel florestal em área de Mata Atlântica, Brasil

Abstract

The forest canopies are responsible for a significant portion of gas and water vapor exchange between plant and atmosphere, having an important influence on global climate balance. The canopy temperature analysis indicates the vegetation physiological behavior in relation to climatic variations. Remote sensing of Land Surface Temperature (LST), like the Moderate Resolution Imaging Spectroradiometer (MODIS) sensor products, may be used to measure the canopy temperature on regional and global scale. However, uncertainties related to the accuracy of orbital data make the validation of these products an essential step in establishing their applicability.Therefore, this work aims to validate the MOD11A2 V006 product from canopy surface temperature data obtained by infrared radiation sensor installed in an Atlantic Forest area in Itatiaia National Park (PNI). For the validation of the MOD11A2 product, a comparative analysis between the LST data, obtained by the MODIS sensor, and the canopy temperature data, obtained by the SI-111 infrared radiation sensor coupled to the PNI Micrometeorological Tower, was performed. The evaluation was based on Pearson's correlation coefficient (r), bias (VM), root mean square error (RMSE), and Willmott's index of agreement (d). Meteorological variables (air temperature, air relative humidity and wind speed and direction) and Land Surface Temperature collected from january to december 2018 in the PNI were also analyzed. The results show that the MOD11A2 product overestimates the canopy temperature in the daytime (VM ranging from 1.56°C to 3.57°C) and underestimates in the night time (VM ranging from -0.18°C to -4,22°C). During daytime, the months corresponding to the dry season presented a very high correlation (r = 0.74 and 0.86) and the highest values of the Willmott index of agreement (d = 0.70 and 0.64). At night time, the MOD11A2 product did not present a good perform for the TST estimation, especially in the rainy season, presenting a negative correlation and low agreement index from January to March (-0.51 and 0.11) and from October to December (-0.07 and 0.18). Therefore, it is possible to state that the MOD11A2 product has limitations to estimate the land surface temperature and that possible changes in the algorithm of this product can be performed for regions of high atmospheric humidity.