Photosynthesis prediction and light spectra optimization of greenhouse tomato based on response of red–blue ratio

Abstract

Light is a key environmental variable regulating the growth and development of vegetables in controlled environment agriculture, and the red–blue ratio (R:B) plays an important role in photosynthesis, where the red and blue regions are considered as the main energy sources for photosynthetic CO2 assimilation. The photosynthesis response to the R:B of greenhouse tomatoes was investigated through the photosynthesis data corresponding to the R:B of 0.5, 1, 3, 5, 7, 9, pure blue, and pure red at various CO2 concentrations. The light response curves of the R:B were studied using the exponential model to identify the photosynthetic parameters of different R:B. The photosynthesis prediction model with light spectra based on incremental constructive extreme learning machine optimized by the sparrow search algorithm was established and used with evolutionary computation to obtain the R:B response curves and optimal R:B. Experimental results show that the determination coefficient of the prediction model in the testing set is 0.986, and the root mean square error is 0.833. At 25 °C, the R:B with the maximum net photosynthetic rate at 400, 700, and 1000 μmol CO2·mol−1 are 5.6, 5.8, and 6.6, respectively. The optimization of light spectra could improve the photosynthetic activity in plants, and the optimal light spectra could reduce the greenhouse energy consumption and light supplement in the environment.