Assessment of adaptive neuro-fuzzy inference system (ANFIS) to predict production and water productivity of lettuce in response to different light intensities and CO2 concentrations

Abstract

The accurate estimation of water productivity (WP) and plant production becomes imperative in planning and managing irrigation practices. Light intensity and CO2 concentration are among the most important determinants of growth and WP of crops. In this study, the adaptive neuro-fuzzy inference system (ANFIS) was used to model the changes in growth parameters, stomatal properties, and WP of lettuce due to various scenarios of light intensity and CO2 concentration. The lettuce plants were exposed to four levels of light intensity [75, 150, 300, and 600 µmol m−2 s−1 Photosynthetic Photon Flux Density (PPFD)] and CO2 concentration (400, 800, 1200, and 1600 ppm). The results showed that growth parameters such as fresh weight, dry weight, and leaf area improved by increasing the PPFD and CO2 concentration from 75 to 300 µmol m−2 s−1 and 400–1200 ppm, respectively. Maximum fresh weight was recorded at 300 µmol m−2 s−1 PPFD and 1600 ppm CO2 concentration while the highest dry weight was obtained at 600 µmol m−2 s−1 PPFD and 1600 ppm CO2 concentration. Stomatal pore width and length decreased by increasing PPFD and CO2 concentration. Moreover, evapotranspiration increased when plants were exposed to higher PPFDs and CO2 concentrations. ANFIS predicted all growth parameters, stomatal properties, and WP with acceptable performance (R2 > 0.99, RMSE < 0.8 ×10−2). The findings provide agricultural engineers with an artificial intelligence-based model to predict the WP and production of lettuce by having the light intensity and CO2 concentration.