Abstract: Intercropping is relatively common and important in tropical and temperate areas because of the effective utilization of resources such as solar energy, water and nutrients, and reduced incidence of weeds, insect pests, and diseases.Among resources, light partitioning is a crucial issue in multi-species canopies because light is involved in most plant responses(e.g.photosynthesis, transpiration, morphogenesis).Numerous experiments have been conducted to investigate radiation interception and utilization by sole and intercropped crops.The hypothesis in these models is that the slowly developing subordinate canopy does not intermingle with the dominant canopy.Therefore, separate strata are used for the whole intercropping period.Interaction among crops in strip intercropping occurs primarily at the edge rows of the strips.It may not be possible to separate border/inner row differences on light interception in these models.Light interception within 3D plant canopies were also implemented with different levels of detail, ranging from plant or facet projections to ray-tracing and radiosity techniques.However, such kinds of calculations need high performance computing system and computing time, and are difficult to be widely used.Munz et al.developed a light partitioning model to calculate the available amount of photosynthetically active radiation(PAR) at the top of the canopy for a given point within a strip of a smaller, subordinate crop.The model captured reasonably the influence of strip design and maize canopy architecture.However, the simulation accuracy will decrease for narrow stripped intercropping systems.Therefore, the objectives of this study were to: 1) introducing maize leaf area density distribution within the canopy and recalculating extinction coefficient(Kd) of scattered radiation based on the experimental data into the model of Munz et al.; 2) evaluating this modified instantaneous light partitioning model for narrow strip intercropping systems and calculating the available amount of PAR at the top of the canopy of a given row within the strip of a subordinate crop; Field experiment was conducted in 2014 at the Shang zhuang experimental farm(40°08’ N, 116°10’ E) of the China Agricultural University.Maize(ZD958) and soybean(ZH30) were planted in a north-south orientation on 1 May and harvested on 4 October.The field experiment comprised two treatments: 1) three rows of intercropped maize with six rows of intercropped soybean(3∶6), this is normally planted in east-north of China; 2) two rows of intercropped maize with three rows of intercropped soybean(2∶3).At silking stage, plant height of 20 plants of maize and soybean were measured.Leaf area was recorded with LI-COR area Meter at the interval of 10 cm canopy height.Photosynthetically active radiation(PAR) was measured with QSO-S PAR sensors.The data was recorded continuously in ten-minute intervals with a EM50 data logger.The results indicated that high accuracy of simulations, under both clear and overcast conditions were found for the comparison between ten-minute averaged simulated and observed values of PAR across the bush bean strip for both treatments.Overall, simulations of ten-minute values of PAR across the bean strip showed that the results is better for 3:6 treatments compared with 2∶3 treatments, and inner rows were better than the border rows: the range of a root mean square error(RMSE) for 3∶6 intercropping system were 59.1-76.4 μmol/(m2·s) for the middle row, 76.8-99.7 μmol/(m2·s) for the inner row, 97.2-157.8 μmol/(m2·s) for the border row and the range of determination coefficient(R2) is 0.76-0.99; The range of RMSE for 2∶3 intercropping system were: 94.9-129.5 μmol/(m2·s) for middle row and 125.8~181.0 μmol/(m2·s) for border row and the range of R2 is 0.83-0.97.The modified model proved to be a helpful tool for understanding the characteristics of light availability across the strip of the subordinate species and their contributions to yields in the strip-intercropping systems.