Evaluating gridded BIOME-BGC for simulating LAI at Kasilian watershed-Iran

Leaf Area Index is a vegetation characteristic which affects water cycle in watersheds by intercepting water. Hence, we decided to simulate LAI with high spatial resolution for different land covers in Kasilian watershed located in Iran from 2004 to 2013. For this purpose, the gridded BIOME-BGC was developed with 500 m spatial resolution which includes 319 grids within Kasilian watershed domain. In this study, The particle swarm optimization (PSO) algorithm was applied to find a combination of 3 parameters including maximum stomatal conductance (Gs_max), new fine root C: new leaf C (FRC: LC), and canopy average specific leaf area which contribute to the best fit between BIOME-BGC LAI and MODIS LAI in each pixels. According to the PSO results, this algorithm indicated an appropriate performance during the calibration and the root mean square error (RMSE) between BIOME-BGC LAI and MODIS LAI converged to an optimum point during the primary iterations in all pixels. Moreover, the computed per cent error (PE) values in all pixels were less than approximately 30% demonstrating that LAI was simulated by BIOME-BGC with a reasonable accuracy for three-dominant land covers Deciduous Broadleaf Forest, Shrub, and C3 grasses in the case study.