Premium
Estimation of Solar Radiation Data Missing from Long‐Term Meteorological Records
Author(s) -
Hook J. E.,
McClendon R. W.
Publication year - 1992
Publication title -
agronomy journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.752
H-Index - 131
eISSN - 1435-0645
pISSN - 0002-1962
DOI - 10.2134/agronj1992.00021962008400040036x
Subject(s) - environmental science , cloud cover , meteorology , precipitation , radiation , atmospheric sciences , climatology , geography , geology , cloud computing , physics , quantum mechanics , computer science , operating system
Solar radiation has a direct effect on plant growth and, thus, is used in most crop growth models. At most locations in the USA, though these measurements have been made only in recent years. At Tifton, GA, a 53‐yr (1938–1990) record was available with meteorological observations consisting of maximum/minimum temperature, precipitation, and pan evaporation. Only during the last 15 yr was solar radiation data available. To allow us to use these data to evaluate realistic scenarios for drought occurrence studies, we developed a method to estimate daily global solar radiation. Using a 13‐yr record of meteorological observations with solar radiation, a regression equation was developed to determine daily solar radiation from maximum and minimum air temperature, pan evaporation, precipitation, and calculated extraterrestrial radiation. A random component was added to express the variability associated with the unmeasured degree of cloudiness. The radiation estimate had the same variances and means as observed data and an overall correlation of 78%. The 53‐yr record with estimated solar radiation represents an intermediate step between computer generated radiation scenarios and fully measured data. The approach presented here should help to complete long‐term weather records so that crop growth and water balance models can be applied to historical data.