摘要：With more than sixty free and publicly available high-quality datasets, including ecosystem variables, radiation budget variables, and land cover products, the MODIS instrument and the MODIS scientific team have contributed significantly to scientific investigations of ecosystems across the globe. The MODIS instrument, launched in December 1999, has 36 spectral bands, a viewing swath of 2330 km, and acquires data at 250 m, 500 m, and 1000 m spatial resolution every one to two days. Radiation budget variables include surface reflectance, skin temperature, emissivity, and albedo, to list a few. Ecosystem variables include several vegetation indices and productivity measures. Land cover characteristics encompass land cover classifications as well as model parameters and vegetation classifications. Many of these products are instrumental in constraining global climate models and climate change studies, as well as monitoring events such as the recent flooding in Pakistan, the unprecedented oil spill in the Gulf of Mexico, or phytoplankton bloom in the Barents Sea. While product validation efforts by the MODIS scientific team are both vigorous and continually improving, validation is unquestionably one of the most difficult tasks when dealing with remotely derived datasets, especially at the global scale. The quality and availability of MODIS data have led to widespread usage in the scientific community that has further contributed to validation and development of the MODIS products. In their recent paper entitled 'Land surface skin temperature climatology: benefitting from the strengths of satellite observations', Jin and Dickinson review the scientific theory behind, and demonstrate application of, a MODIS temperature product: surface skin temperature. Utilizing datasets from the Global Historical Climatological Network (GHCN), daily skin and air temperature from the Atmospheric Radiation Measurement (ARM) program, and MODIS products (skin temperature, albedo, land cover, water vapor, cloud cover), they show that skin temperature is clearly a different physical parameter from air temperature and varies from air temperature in magnitude, response to atmospheric conditions, and diurnal phase. Although the accuracy of skin temperature (Tskin) algorithms has improved to within 0.5–1°C for field measurements and clear-sky satellite observations (Becker and Li 1995, Goetz et al 1995, Wan and Dozier 1996), general confusion regarding the physical definition of 'surface temperature' and how it can be used for climate studies has persisted throughout the scientific community and limited the applications of these data (Jin and Dickinson 2010). For example, satellite sea surface temperature was used as evidence of global climate change instead of skin temperature in the IPCC 2001 and 2007 reports (Jin and Dickinson 2010). This work provides clarity in the theoretical definition of temperature variables, demonstrates the difference between air and skin temperature, and aids the understanding of the MODIS Tskin product, which could be very beneficial for future climate studies. As outlined by Jin and Dickinson, 'surface temperature' is a vague term commonly used in reference to air temperature, aerodynamic temperature, and skin temperature. Air temperature (Tair), or thermodynamic temperature, is measured by an in situ instrument usually 1.5–2 m above the ground. Aerodynamic temperature (Taero) refers to the temperature at the height of the roughness length of heat. Satellite derived skin temperature (Tskin) is the radiometric temperature derived from the inverse of Planck's function. While these different temperature variables are typically correlated, they differ as a result of environmental conditions (e.g. land cover and sky conditions; Jin and Dickinson 2010). With an extensive network of Tair measurements, some have questioned the benefits of using Tskin at all (Peterson et al 1997, 1998). Tskin and Tair can vary depending on land cover or sky conditions and variations may be large, e.g., for sparsely vegetated areas where net radiation is largely balanced by sensible heat flux (Hall et al 1992, Sun and Mahrt 1995, Jin et al 1997). Tskin can be higher than Taero at midday and lower at night (Sun and Mahrt 1995) and some models use Taero to approximate surface radiative temperature (Hubband and Monteith 1986). One of the strengths of the MODIS instrument is the simultaneous collection of surface and atmospheric conditions. By incorporating a range of MODIS variables in their comparison to Tskin, the authors examine the relationship of Tskin to atmospheric and surface conditions. Results from their global evaluation of Tskin highlight its variability on an inter-annual basis, its variation with solar zenith angle, and diurnal variations, which are not achievable with Tair measurements. Comparison with land cover type illustrates the seasonality of Tskin for different land covers. Comparison with the enhanced vegetation index (EVI) suggests more vegetation reduces skin temperature. Using the MODIS albedo, they demonstrate a clear relationship between yearly averaged Tskin and land surface albedo. Lastly, their examination of water vapor and cloud cover in comparison to Tskin suggests similar seasonality between these two variables. The MODIS Tskin product is not without uncertainty; retrieving Tskin requires a calculation of radiative transfer to account for atmospheric emission and molecular absorption, which is time and resource intensive (Jin and Dickinson 2010). Additionally, surface emissivity, instrument noise, and view angle geometry contribute to error in Tskin estimations (Jin and Dickinson 2010). The transparency of the scientific theory underlying this work, and the clear demonstration of the distinction between temperature measures on varying scales, demonstrates the usefulness of Tskin despite the uncertainties. Perhaps equally as important is the tone; in a time when the controversy surrounding climate change is peaking and the very ethics of the scientific community are being questioned, it is more critical than ever to be transparent in one's work and to assist the scientific community in understanding the tools we have available to us for investigating climate change. References Becker F and Li Z-L 1995 Surface temperature and emissivity at different scales: definition, measurement and related problems Remote Sensing Rev.12 225–53 Goetz S J, Halthore R, Hall F G and Markham B L 1995 Surface temperature retrieval in a temperate grassland with multi-resolution sensors J. Geophys. Res. Atmos.100 25397–410 Hall F G, Huemmrich K F, Goetz P J, Sellers P J and Nickeson J E 1992 Satellite remote sensing of the surface energy balance: success, failures and unresolved issues in FIFE J. Geophys. Res. Atmos.97 19061–90 Jin M and Dickinson R E 2010 Land surface skin temperature climatology: benefitting from the strengths of satellite observations Environ. Res. Lett.5 044004 Jin M, Dickinson R E and Vogelmann A M 1997 A comparison of CCM2/BATS skin temperature and surface-air temperature with satellite and surface observations J. Climate10 1505–24 Hubband N D S and Monteith J L 1986 Radiative surface temperature and energy balance of a wheat canopy Boundary Layer Meteorol.36 107–16 Peterson T C and Vose R S 1997 An overview of the Global Historical Climatology Network temperature data base Bull. Am. Meteorol. Soc.78 2837–49 Peterson T C, Karl T R, Jamason P F, Knight R and Easterling D R 1998 The first difference method: maximizing station density for the calculation of long-term global temperature change J. Geophys. Res. Atmos.103 25967–74 Sun J and Mahrt L 1995 Determination of surface fluxes from the surface radiative temperature Atmos. Sci.52 1096–106 Wan Z and Dozier J 1996 A generalized split-window algorithm for retrieving land-surface temperature from space IEEE Trans. Geosci. Remote Sensing34 892–905
