An efficient hybrid method for estimating clear‐sky surface downward longwave radiation from MODIS data

Abstract

This paper proposes an efficient hybrid method for estimating 1 km instantaneous clear‐sky surface downward longwave radiation (LWDN) from Moderate Resolution Imaging Spectroradiometer (MODIS) thermal infrared observations and the MODIS near‐infrared column water vapor (CWV) data product. The LWDN was formulated as a nonlinear function of surface upwelling longwave radiation estimated from the MODIS top‐of‐atmosphere (TOA) radiance of channels 29, 31, and 32, as well as CWV and the MODIS TOA radiance of channel 29. Ground measurements collected at 62 globally distributed sites from six networks were used to develop and validate the proposed hybrid method. The validation results showed that the bias and root‐mean‐square error (RMSE) were 0.0597 W/m2 and 21.008 W/m2. These results demonstrate that the performance of our method is superior to that of other studies reported in the literature. The drawback of our method is that LWDN is overestimated over high‐elevation areas with extremely low CWV (<0.5 g/cm2) and underestimated over regions with tropical climates that have extremely high CWV. A power function relating LWDN to CWV was derived and used as a complementary method to address these circumstances. The overestimation was overcome, and the bias and RMSE decreased from 9.407 W/m2 and 23.919 W/m2 to −0.924 W/m2 and 19.895 W/m2. The underestimation was also alleviated.

New Methods for Deriving Clear‐Sky Surface Longwave Downward Radiation Based on Remotely Sensed Data and Ground Measurements

“…According to the validation results of MODIS LST products, the accuracy of LST is about 1 K under clear sky (Wan, 2014;Wan & Li, 2008). The relatively high 10.1029/2019EA000754 Earth and Space Science retrieval accuracy of MODIS PWC and LST among other remote sensingbased atmosphere and land surface parameters ensure the applicability of those parameters in LWDR estimation (Cheng et al, 2017;Yu et al, 2013). Land cover types (LUCC) product (MCD12C1) defined by the International Geosphere-Biosphere Programme was also used to identify the land covers of the in-situ sites.…”

“…The amount of each component and its change has significant influences on global climate and hydrological variables including the temperature, precipitation, evapotranspiration, and runoff and thus cannot be ignored especially the LWDR. LWDR plays a vital role in the interaction of atmospheric and Earth surface, and describes the energy emitted by H 2 O, CO 2 , and O 3 (Yamada et al, 2012) from the atmosphere to the surface in the form of thermal radiation at wavelength from 4 to 100 μm (Cheng et al, 2017). It is also considered to be the most poorly estimated radiation component, and its uncertainty is regarded as substantially larger than other terms of surface energy budget (Stephens et al, 2012;Trenberth et al, 2009;Trenberth & Fasullo, 2012;Wang et al, 2018;Wang & Dickinson, 2013).…”

“…During the past decades, a number of ground-based radiation observation networks have been set up to monitor the surface energy balance, including the Surface Radiation Budget Network (SURFRAD; Augustine et al, 2000), the Baseline Surface Radiation Network (BSRN; Ohmura et al, 1998), the Atmospheric Radiation Measurements (Stokes & Schwartz, 1994), the Coordinated Energy and Water Cycle Observations Project (CEOP; https://www.eol.ucar.edu/field_projects/ceop), and the FLUXNET (Baldocchi et al, 2001). Comparing to other approaches, the ground-based measurement is more accurate and is frequently used to develop the LWDR estimation algorithms and validate LWDR derived from other methods (Cheng et al, 2017;Zhou & Cess, 2001). Parameterization schemes usually calculate LWDR basing on Stefan-Boltzmann's law using near-surface meteorological parameters such as air temperature, vapor pressure, and other variables as inputs.…”

“…With the development of remote sensing technology, the global mapping of LWDR at kilometer scales becomes possible, and it is also deemed as an indispensable way for studying the global radiation budget (Wang & Dickinson, 2013). For example, a global 1-km instantaneous clear-sky longwave radiation product (including LWDR, LWUR, and net longwave radiation) has been produced from MODIS data using the hybrid method (Cheng, & Liang, 2016;Cheng et al, 2017) and freely released to public (http://glassproduct.bnu.edu.cn/). Darnell et al (1983Darnell et al ( , 1986 first presented a scheme to derive LWDR using satellite-based meteorological data and a radiative transfer model, which unfold the estimation of LWDR using remote sensing data.…”

“…Zhou et al (2007), Zhou and Cess (2001) found that LWDR has a strong relationship with LWUR and column precipitable water vapor (PWC) and applied this relationship to estimate LWDR globally using remote sensing products. Cheng et al (2017) built a LWDR estimation model using the MODIS-based LWUR (Cheng & Liang, 2016), PWC, MODIS TOA radiance of band 29, and ground-based LWDR observations. The advantage of the hybrid method is the high computational efficiency, but the model's accuracy depends on the representativeness of the simulated database as well as the regression method.…”