Solar Radiation Weighting for Weather and Corn Growth Models1

Abstract

AbstractAlthough agriculture is the business of converting solar energy into food and fiber, there has been little use of solar radiation data in regression models of weather and crop growth. A major problem in these studies is the rapidly increasing plant size and growth potential under similar environmental conditions. This has led to the use of standardized measures of Crop Growth Rate (CGR), such as Relative Growth Rate (RGR) and Net Assimilation Rate (NAR), which often overcorrect for crop size. Rather than correct CGR for crop size in the present work solar radiation is weighted to estimate the amount of solar energy intercepted by a corn canopy. Another challenge in solar radiation‐crop growth studies to assess the interaction of solar radiation and plant moisture stress, and this factor is also examined. Weather and corn (zea mays L.) growth measurements in the field at West Lafayette, In., from 1969 to 1971, were used to assess different methods of handling solar radiation in corn growth regression models. Daily solar radiation (S) was weighted with a leaf area function, f(LAI), using Boguer's law and an extinction coefficient determined by K. R. Stevenson and J. W. Tanner to obtain the intercepted solar radiation, SI = S(1 — exp(−0.79 LAI)). This variable was better correlated with total aboveground corn plant growth rates than was either f(LAI) or S. SI was divided by pan (Epan) or Piche evaporation (EPiche), or their square roots, to obtain a derived variable which identified not only the solar energy intercepted by the corn canopy, but also a consideration of atmospheric conditions causing plant moisture stress limitations upon energy conversion within the corn plant. The single variable, SI‐E‐−Piche, was associated with about 45% of the variance in above ground plant growth rates and about 25% of the variability in grain growth rates during the 1969–1971 seasons. The derived variables were tested with 1972 weather and corn growth data and also by pooling data for the eight seasons into three other periods, 1969–1970, 1971–1972, and 1969–1972. The single derived variables, SI‐E½Piche, was associated with 72% of the short period plant growth rate variance in the 1969–1970 period and about 27% in the 1971–1972 period, but it explained practically none of the variability in 1971–1972 grain growth.