Day 1: mHM Configuration Tasks

[MuellerSeb]

All task will be done on the test domain 1 from the mHM repository.

To download this (assuming mhm was installed with conda), you can execute this command:

mhm-download -v

Afterwards you will see a new folder test_domain/ with all input data and configuration files inside.
You can run the test domain with

mhm test_domain

A simple tool to check NetCDF files is ncdump, that can be used to get header information (-h) about a file with:

ncdump -h test_domain/output_b1/mHM_Fluxes_States.nc

For documentation on the namelist files see the reference namelists.

First Block

Task 1: Output hourly and only runoff

• Set model output timestep to 1 hour (timeStep_model_outputs in mhm_outputs.nml)
  • What happens for different output frequencies?
• Adjust output settings in mhm_outputs.nml / mrm_outputs.nml
  • Enable only discharge/runoff variables
  • Disable all other variables
• Run a short test and check that only hourly runoff files are created

---

Task 2: Change hydrological resolution

• Check the resolutions of latlon_1.nc and latlon_2.nc in input/latlon/
• Change the path of file_LatLon to latlon_2.nc in &directories_general (mhm.nml)
• Update resolution_Hydrology in &mainconfig (mhm.nml) to match the L1 info from the new latlon file
• Update resolution_Routing in &mainconfig_mhm_mrm (mhm.nml) to match the L11 info from the new latlon file
• Run a short test to verify that the grid and outputs match the new resolution

---

Task 3: Change time frame

• Edit the &time_periods namelist in mhm.nml
  • Set new start and end dates (eval_Per%yStart, %mStart, %dStart, etc.)
  • Optionally change warm-up days (warming_Days)
• Ensure forcing and land cover data cover the new simulation period
• Run the model for a short time range to confirm correct setup

---

Second Block

Task 4: Restart files (create one and restart your model from it)

First run:

• Set write_restart = .TRUE. in mhm.nml
• Run the model and check that restart files (mHM_restart_*.nc) are written

Second run:

• Set read_restart = .TRUE. in mhm.nml
• Ensure paths to restart files are correct
• Adjust simulation period to continue from the previous run
• Run the model again and confirm that it starts from the restart state (no spin-up)

---

Optional Tasks

Optional Task A: Change infiltration / soil-moisture process

• In &processSelection in change processCase(3) (soil moisture) to another option, e.g.:
  • 1 = Feddes + infiltration capacity (default)
  • 2 = Jarvis equation (alternative ET reduction)
  • 4 = Feddes + global FC dependency
• Run a short test and compare resulting soil moisture or runoff behavior

---

Optional Task B: Change PET method

• In &processSelection, add processCase(5) (PET process) and add required input directories in &directories_general:
  • 0 = PET as input (aspect correction)
  • 1 = Hargreaves–Samani (needs dir_MinTemperature, dir_MaxTemperature)
  • 2 = Priestley–Taylor (needs dir_NetRadiation)
  • 3 = Penman-Monteith method (needs dir_NetRadiation, dir_absVapPressure, dir_windspeed)
• Make sure the corresponding input folders and files exist (e.g., <dir_MinTemperature>/tmin.nc)
  • if required files are not found, mHM will raise an error
• Run and observe changes in evapotranspiration and discharge

---

Optional Task C: Check parameter ranges for soil moisture process

(Jarvis equation for ET reduction and global FC dependency on root fraction coefficient)

Requires that you already looked at mhm_parameter.nml

• Open mhm_parameter.nml and find parameters related to soil moisture:
  • infiltrationShapeFactor
  • rootFractionCoefficient_forest
• Check which value ranges (minimum, maximum, and initial value) are defined for these parameters
• These parameters are active when the soil moisture process is set to Jarvis equation for ET reduction and global FC dependency on root fraction coefficient (processCase(3) = 3)
• (Optional) Slightly modify one of the parameters within its valid range and re-run a short test to see how soil moisture or discharge respond