Land Data Structure

The land structure is a NamedTuple that serves as the primary data container for SINDBAD models, facilitating information exchange between different model components.

Structure Organization

The land variables are organized hierarchically with exactly two levels:

1. Field: Groups variables by their functional category

2. Subfield: Contains the actual data values

Field Categories

Variables are grouped into the following categories based on their characteristics and usage:

Shared Fields

These fields contain variables used across multiple models:

• constants: Helper variables and structure-dependent parameters that remain constant throughout the simulation

• diagnostics: Derived variables indicating stressors, controllers, and rates based on forcing, pools, or states

• fluxes: Mass/area/time variables representing ecosystem fluxes

• models: Type-based instances used for soil property calculations and pool updates

• pools: Storage variables automatically generated from model_structure.json

• properties: Land surface characteristics (soil, vegetation) and their derivatives

• states: Ecosystem states and derived variables

Model-Specific Fields

Variables used exclusively by a single model are stored under the model name (e.g., cCycleBase).

Structure Validation

To ensure proper organization and prevent potential issues, use the following command to examine the land structure:

julia> tcPrint(land)

Important Considerations

• No automatic checks prevent variable overwriting

• Avoid duplicating fields across different groups

• Maintain consistent naming conventions

• Ensure proper variable grouping

Best Practices

1. Organization

• Keep variables in their appropriate categories

• Use consistent naming conventions

• Document any model-specific fields

2. Validation

• Regularly check structure using tcPrint

• Verify variable grouping

• Ensure no unintended overwrites

3. Maintenance

• Update documentation when adding new fields

• Review structure after model modifications

• Maintain clear variable categorization