Mastering root distribution is essential for optimizing the root zone environment and for improving water use efficiency, especially for crops cultivated in greenhouses. Here, we set up two irrigation amount levels based on measurements of the cumulative 20 cm pan evaporation (Ep) (i.e., K0.9: 0.9 Ep; K0.5: 0.5 Ep), and three ventilation modes through opening the greenhouse vents at different locations (TR: open the roof vents only; TRS: open both the roof and south vents; TS: open the south vents only) to reveal the effects of the ventilation mode and irrigation amount on the root distribution of greenhouse tomato. Six treatments were designed in blocks with the ventilation mode as the main treatment and the irrigation amount as the vice treatment. On this basis, the normalized root length density (NRLD) model of six treatments was developed by considering air environment, soil water and temperature conditions, root length density (RLD) and yield. The results showed that air speed of the TRS was significantly higher than TR and TS (p < 0.01), and the air temperature and relative humidity under different ventilation showed the rule: TR > TS > TRS. There was a significant third-order polynomial function relationship between NRLD and soil depth, and the coefficient of the cubic term (R0) had a bivariate quadratic polynomial function relationship with irrigation amount and air speed (determination coefficient, R2 = 0.86). Root mean square errors of the simulated and measured value of NRLD under TR, TRS and TS were 0.20, 0.23 and 0.27 in 2020, and 0.31, 0.23 and 0.28 in 2021, respectively, normalized root mean squared errors were 15%, 17%, 20% in 2020, and 23%, 18% and 21% in 2021. The RLD distribution ratio from the ground surface to a one-quarter relative root depth was 74.1%, and 88.0% from the surface to a one-half relative root depth. The results of the yield showed that a better combination of ventilation and irrigation was recommended as TRS combined with K0.9.