Skip to main content

Land Use / Land Cover (LULC)

The Land Use / Land Cover (LULC) map defines the spatial distribution of surface characteristics across the simulation domain.

In HydroPol2D, the LULC input is used to represent how different portions of the domain respond to rainfall and surface flow through parameter assignment.

Role in HydroPol2D​

The LULC map is used to define spatial variability in:

  • Surface roughness (Manning’s coefficient)
  • Interception storage
  • Impervious fraction
  • Surface-related evapotranspiration behavior

These properties directly influence runoff generation, flow velocity, and water retention at the surface.

LULC Representation​

HydroPol2D requires a categorical LULC map.

Each grid cell must be assigned a land use class identifier (integer or categorical value). These classes are not used directly in the governing equations. Instead, they serve as indices into parameter tables.

For each LULC class, a set of parameters is defined externally in the input tables. This includes:

  • Manning’s roughness coefficient
  • Interception capacity
  • Impervious fraction
  • Additional process-specific parameters

This approach separates spatial information from parameter definition and allows full flexibility in configuring model behavior.

Available LULC Datasets​

Several high-quality global and regional LULC datasets are freely available:

NLCD (National Land Cover Database)​

  • Region: United States
  • Resolution: 30 m
  • Includes detailed urban classes and impervious surfaces
  • Download: https://www.mrlc.gov/

ESA WorldCover​

Copernicus Land Monitoring Service​

  • Region: Europe and global products
  • Resolution: 10–100 m
  • Includes multiple thematic layers
  • Download: https://land.copernicus.eu/

MODIS Land Cover (MCD12Q1)​

Dynamic World​

  • Region: Global
  • Resolution: 10 m
  • Near real-time classification derived from Sentinel-2
  • Provides probabilistic class outputs
  • Particularly useful for dynamic or rapidly changing environments
  • Access: Google Earth Engine

Parameter Association​

LULC classes are mapped to model parameters through the input tables.

Each class must be associated with a consistent set of physical values. This mapping determines how the model represents:

  • Surface resistance to flow
  • Rainfall interception
  • Runoff generation behavior

Incorrect or inconsistent parameter assignment will directly affect simulation results.

Impervious Surfaces​

Urban classes typically include a significant fraction of impervious area (roads, rooftops, pavements).

These areas are characterized by:

  • Negligible infiltration
  • Rapid runoff generation
  • Increased flow velocities

The impervious fraction parameter is used to represent this behavior and must be carefully assigned for urban land use classes.

Fractional impervious raster override​

HydroPol2D can also ingest an optional ImperviousFraction_path raster. When present, it refines the cellwise impervious fraction beyond the class-based LULC assignment.

Current implementation details:

  • raster values may be given in [0,1] or [0,100] percent,
  • the raster is aligned to the DEM internally,
  • built-up / explicitly impervious LULC classes are still forced to remain fully impervious.

So the LULC table still defines the base impervious behavior, while the impervious-fraction raster provides a spatial refinement on top of that base.

Resolution and Alignment​

The LULC raster must:

  • Match or be resampled to the DEM resolution
  • Use the same projected coordinate reference system (CRS)
  • Align spatially with all other raster inputs

Misalignment leads to incorrect parameter assignment and inconsistent flow representation.

Reclassification​

Raw LULC datasets often contain a large number of classes.

For HydroPol2D applications, it is often preferable to:

  • Aggregate classes into a reduced set of representative categories
  • Group similar land use types
  • Simplify the parameterization scheme

This improves model robustness and reduces unnecessary complexity.

Interaction with Other Inputs​

The LULC map defines surface behavior and must be consistent with:

  • DEM → defines terrain and flow routing
  • Soil → controls infiltration capacity
  • LAI → influences vegetation-related processes
  • Albedo → affects surface energy balance

Each of these inputs contributes a different aspect of spatial variability in the model.

Summary​

In HydroPol2D, LULC is treated as a categorical map used to assign surface-related parameters through lookup tables.

The quality of the classification and the consistency of parameter mapping are critical for accurately representing surface flow resistance, runoff generation, and vegetation effects.