Abstract
Abstract. While solar eclipses are known to greatly diminish the visible radiation reaching the surface of the Earth, less is known about the magnitude of the impact. We explore both the observed and modeled levels of change in surface radiation during the eclipse of 2017. We deployed a pyranometer and Pandora spectrometer instrument to Casper, Wyoming, and Columbia, Missouri, to measure surface broadband shortwave (SW) flux and atmospheric properties during the 21 August 2017 solar eclipse event. We performed detailed radiative transfer simulations to understand the role of clouds in spectral and broadband solar radiation transfer in the Earth's atmosphere for the normal (non-eclipse) spectrum and red-shift solar spectra for eclipse conditions. The theoretical calculations showed that the non-eclipse-to-eclipse surface flux ratio depends strongly on the obscuration of the solar disk and slightly on the cloud optical depth. These findings allowed us to estimate what the surface broadband SW flux would be for hypothetical non-eclipse conditions from observations during the eclipse and further to quantify the impact of the eclipse on the surface broadband SW radiation budget. We found that the eclipse caused local reductions of time-averaged surface flux of about 379 W m−2 (50 %) and 329 W m−2 (46 %) during the ∼3 h course of the eclipse at the Casper and Columbia sites, respectively. We estimated that the Moon's shadow caused a reduction of approximately 7 %–8 % in global average surface broadband SW radiation. The eclipse has a smaller impact on the absolute value of surface flux reduction for cloudy conditions than a clear atmosphere; the impact decreases with the increase in cloud optical depth. However, the relative time-averaged reduction of local surface SW flux during a solar eclipse is approximately 45 %, and it is not sensitive to cloud optical depth. The reduction of global average SW flux relative to climatology is proportional to the non-eclipse and eclipse flux difference in the penumbra area and depends on cloud optical depth in the Moon's shadow and geolocation due to the change in solar zenith angle. We also discuss the influence of cloud inhomogeneity on the observed SW flux. Our results not only quantify the reduction of the surface solar radiation budget, but also advance the understanding of broadband SW radiative transfer under solar eclipse conditions.
Highlights
On 21 August 2017, a total solar eclipse traversed the continental US from Oregon to South Carolina (Fig. 1)
For the Casper site (Fig. 7a), in the first period from 16:00 to 18:12 UTC before and during a large part of the eclipse, the observed surface SW flux varies rather smoothly with time, similar in behavior to that for modeled clear-sky flux except for a few tiny dips, which is likely due to fragments of thin cirrus passing through the field of view (FOV) of Pandora spectrometer instrument system (PSI), as indicated by small spikes in cloud optical depth observations (Fig. 2)
From 16:00 to 16:42 UTC, the observed flux exceeds the modeled one for clear atmospheric conditions by more than 20 W m−2 and by a much smaller amount as time proceeds after 16:42 UTC
Summary
On 21 August 2017, a total solar eclipse traversed the continental US from Oregon to South Carolina (Fig. 1) (https: //eclipse2017.nasa.gov/eclipse-maps, last access: 21 August 2020). The solar eclipse can cause large reductions in both temporally averaged surface broadband shortwave (SW) flux at a given site along the totality path and the spatially averaged global surface SW radiation budget at a given time during the eclipse. The eclipseinduced surface SW flux reduction can lead to a decrease in sensible heat flux and associated changes in wind speed G. Wen et al.: Changes in the surface broadband shortwave radiation budget during the 2017 eclipse
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
