problems with perennial plants

[marianna leone]

1. Primary Issue Location

• Component:  SWAT+ Executable / Model Functionality  
• Area of impact: Plant growth, perennial vegetation phenology, evapotranspiration output  

2. Software Versions

• GIS Interface:  QSWAT+ 
• SWAT+ Editor:  4.0.1 
• SWAT+ Revision:  62.0.0

 The same perennial-growth behaviour was also reproduced using a custom build from the fix/perennial-root-depth-init branch (Rev 61.0.2.5-998-g8d1ec1c).  

3. Project Characteristics

• Watershed Details:  Lake Leado watershed, Ethiopia, approximately 546 km². The project is discretized into LSUs and HRUs using land use, soil and slope classes. HRU thresholds were reduced from 3% to 2% during model testing.  
• Modeling Aim:  Water balance, evapotranspiration, vegetation growth and soil erosion assessment in a data-limited basin.  

4. Description of Issue

• The Problem: Several perennial vegetation classes do not develop normally during the simulation. Their LAI remains fixed at the minimum LAI value for the entire simulation period, while biomass growth remains zero.

For example, the original RNGE plant remained at: LAI = 0.05 lai_max = 0 bm_grow = 0 c_gro = 0. 

I then replaced it with rnge_sudrf. The model correctly reads the new plant definition, because LAI changes from 0.05 to the new lai_min = 0.50, but it remains fixed at 0.50 for the entire simulation period and biomass growth is still zero.

The same type of behaviour is also observed for other perennial land covers such as FRST, COFF and RNGB, while annual crops such as corn and wheat show normal seasonal LAI development and biomass growth.

The perennial vegetation is initialized as already established vegetation, for example:  lc_status = y lai_init = 2.0 bm_init = 20000 kg/ha yrs_init = 1.0.  Therefore, these plants are intended to represent already-established perennial vegetation rather than newly planted vegetation.

• Step to Reproduce: 
• Run the model with perennial vegetation initialized as already established.
• Check hru_pw_mon output.
• Perennial vegetation remains at lai_min throughout the simulation, with lai_max = 0 and bm_grow = 0.
• Replace the original RNGE definition with rnge_sudrf.
• The model reads the new minimum LAI correctly (0.50 instead of 0.05), but LAI still does not develop and biomass growth remains zero.
• Both temp_gro and moisture_gro growth triggers were tested without solving the issue.
• Changes to dormancy-related settings were also tested without effect.
• The fix/perennial-root-depth-init branch was tested. It produced a significant increase in plant transpiration and total ET, confirming that the perennial root-depth correction affects the hydrological response, but it did not solve the LAI/growth problem.
• As an additional diagnostic test, perennial plants initialized with igro = "y" were forced to start with idorm = "n" in plant_init.f90. The resulting plant-growth and water-balance outputs were unchanged.

 Has anyone experienced the same behaviour with established perennial vegetation? Is there an additional plant-community, management or phenology setting required to activate seasonal growth, or could this be related to the current perennial growth logic in the SWAT+ executable? 
thank you very much