Photosynthetically active radiation (PAR) was measured at two times of day (8:00 am and noon Central Standard Time) in a 915 × 915-cm area of a 1006 × 915-cm gable roof greenhouse. PAR measurements were taken across a grid at 40-cm intervals, a total of 529 data points. Spatial variation of PAR in the greenhouse was evaluated through contour plots and the geostatistical technique of semivariogram construction. Semivariograms provide a visual guide to the degree of spatial correlation of a variable, allowing a quantification of the distance at which variables cease to be spatially correlated (the range) Measured PAR contained distinct zones of lowered values, a function of overhead greenhouse structures, wall-hung electrical boxes, and tall plants in adjacent greenhouses. Although the amount of PAR changed over time, zones of high and low PAR remained relatively constant, except at the sides of the greenhouse. Constructed semivariograms revealed that PAR contained strong spatial correlation (up to a 350-cm separation) as measured in the north-south direction, moving parallel to greenhouse bench placement. When PAR measurements perpendicular to benches (east-west) were used in directional semivariograms PAR was found to be completely random, plotting as a horizontal line called a nugget effect. Thus, plants placed perpendicular to the greenhouse benches (east-west) would not be affected by the spatial correlation of PAR.
If the inline PDF is not rendering correctly, you can download the PDF file here.