Terrain Analysis Pipeline¶

Generate slope, aspect, hillshade, and other terrain derivatives from Digital Elevation Models (DEM).

Overview¶

graph TB
    subgraph "Input"
        A[Digital Elevation Model]
        B[Resolution 10-30m]
        C[Region Boundary]
    end

    subgraph "Analysis"
        D[Slope Analysis]
        E[Aspect Analysis]
        F[Hillshade]
        G[Roughness Index]
        H[Topographic Indices]
    end

    subgraph "Outputs"
        I[Slope Map]
        J[Aspect Map]
        K[Hillshade Map]
        L[Roughness Map]
        M[Statistics]
    end

    A --> D
    A --> E
    A --> F
    A --> G
    A --> H

    D --> I
    E --> J
    F --> K
    G --> L
    H --> M

Terrain analysis provides foundational data for hydrology, agriculture, infrastructure planning, and environmental assessments.

Available Products¶

graph LR
    A[DEM Input] --> B[Basic Products]
    A --> C[Derived Products]
    A --> D[Indices]

    B --> E[Slope]
    B --> F[Aspect]
    B --> G[Hillshade]

    C --> H[Curvature]
    C --> I[Flow Direction]
    C --> J[Flow Accumulation]

    D --> K[TPI]
    D --> L[TRI]
    D --> M[Roughness]

Algorithm Details¶

Slope Calculation¶

graph LR
    A[3x3 Window] --> B[dz/dx]
    A --> C[dz/dy]

    B --> D[Slope = atan√(dz/dx² + dz/dy²)]
    C --> D

    D --> E[Degrees]

Using Horn's method for improved accuracy:

def calculate_slope(dem):
    """Calculate slope in degrees from DEM."""
    # Calculate gradients
    dz_dx = ((dem[0,0] + 2*dem[0,1] + dem[0,2]) -
             (dem[2,0] + 2*dem[2,1] + dem[2,2])) / (8 * cell_size)

    dz_dy = ((dem[2,0] + 2*dem[1,0] + dem[0,0]) -
             (dem[2,2] + 2*dem[1,2] + dem[0,2])) / (8 * cell_size)

    # Calculate slope
    slope_rad = np.arctan(np.sqrt(dz_dx**2 + dz_dy**2))
    slope_deg = np.degrees(slope_rad)

    return slope_deg

Aspect Calculation¶

graph LR
    A[dz/dx] --> B[Aspect = atan2(dz/dy, -dz/dx)]
    A --> C[dz/dy]
    C --> B
    B --> D[Convert to 0-360]
    D --> E[North = 0°]

Aspect represents the compass direction that a slope faces:

Hillshade¶

graph LR
    A[Slope] --> D[Hillshade = 255 × (cos(Z) × cos(S) + sin(Z) × sin(S) × cos(Az-As))]
    B[Aspect] --> D
    C[Sun Azimuth] --> D
    C2[Sun Zenith] --> D

Default sun position: Azimuth = 315°, Zenith = 45°

Input Requirements¶

Required Parameters¶

{
  "state": "karnataka",
  "district": "raichur",
  "block": "devadurga",
  "products": ["slope", "aspect", "hillshade"],
  "resolution": "10m",
  "azimuth": 315,
  "altitude": 45
}

DEM Sources¶

graph TB
    A[DEM Sources] --> B[Cartosat-1]
    A --> C[ALOS World 3D]
    A --> D[SRTM]
    A --> E[Custom]

    B --> F[2.5m Resolution<br/>India Coverage]
    C --> G[30m Resolution<br/>Global]
    D --> H[30m Resolution<br/>Global]
    E --> I[Any Resolution<br/>User Provided]

Processing Workflow¶

sequenceDiagram
    participant U as User
    participant A as API
    participant D as DEM Processor
    participant T as Terrain Tools
    participant S as Storage

    U->>A: Submit Request
    A->>D: Fetch/Generate DEM
    D->>A: DEM Data

    loop For Each Product
        A->>T: Calculate Product
        T->>T: Apply Algorithm
        T->>A: Result Raster
    end

    A->>S: Store Results
    S->>U: Download URLs

Step 1: DEM Preparation¶

• Reproject to EPSG:4326
• Resample to target resolution
• Fill sinks (hydro-enforcement)
• Clip to region boundary

Step 2: Derivative Calculation¶

graph TB
    A[Prepared DEM] --> B[Slope Calc]
    A --> C[Aspect Calc]
    A --> D[Hillshade Calc]
    A --> E[Roughness Calc]
    A --> F[TPI Calc]

    B --> G[GDAL Slope]
    C --> H[GDAL Aspect]
    D --> I[GDAL Hillshade]
    E --> J[GDAL TRI]
    F --> K[GDAL TPI]

Step 3: Quality Control¶

• Check for artifacts
• Validate statistics
• Generate previews

Output Products¶

Raster Outputs¶

Statistics Report¶

{
  "terrain_stats": {
    "elevation": {
      "min_m": 420.5,
      "max_m": 685.2,
      "mean_m": 542.8,
      "std_m": 45.3
    },
    "slope": {
      "min_degrees": 0.0,
      "max_degrees": 45.8,
      "mean_degrees": 8.5,
      "std_degrees": 6.2,
      "percent_under_5": 35.2,
      "percent_5_to_15": 48.7,
      "percent_over_15": 16.1
    },
    "aspect": {
      "dominant_direction": "southeast",
      "percent_north_facing": 22.5,
      "percent_south_facing": 28.3
    }
  }
}

Slope Distribution¶

pie title Slope Distribution
    "Flat (0-5°)" : 35.2
    "Gentle (5-15°)" : 48.7
    "Moderate (15-25°)" : 12.1
    "Steep (>25°)" : 4.0

API Usage¶

Basic Request¶

curl -X POST "https://geoserver.core-stack.org/api/v1/generate_terrain_descriptor/" \
  -H "Content-Type: application/json" \
  -d '{
    "state": "karnataka",
    "district": "raichur",
    "block": "devadurga",
    "gee_account_id": 1
  }'

Response¶

{
  "Success": "generate_terrain_descriptor task initiated"
}

Python Example¶

import requests

response = requests.post(
    "https://geoserver.core-stack.org/api/v1/generate_terrain_raster/",
    json={
        "state": "karnataka",
        "district": "raichur",
        "block": "devadurga",
        "gee_account_id": 1
    }
)

print(response.status_code)
print(response.json())

Applications¶

Agriculture¶

graph TB
    A[Terrain Data] --> B[Suitability Analysis]
    A --> C[Erosion Risk]
    A --> D[Irrigation Design]

    B --> E[Crop Selection]
    C --> F[Conservation Planning]
    D --> G[Water Distribution]

Slope Classes for Agriculture:

Hydrology¶

• Flow routing: D8/D-infinity algorithms
• Watershed delineation: Pour point analysis
• Stream extraction: Contributing area threshold

Infrastructure¶

graph LR
    A[Road Design] --> B[Max Slope: 6-8%]
    C[Building Sites] --> D[Max Slope: 15%]
    E[Drainage] --> F[Min Slope: 0.5%]

Visualization¶

Color Ramps¶

Slope: - Green: 0-5° (Flat) - Yellow: 5-15° (Gentle) - Orange: 15-25° (Moderate) - Red: >25° (Steep)

Aspect: - Continuous color wheel - North = Blue - South = Red - East = Yellow - West = Green

3D Visualization¶

Combine hillshade with: - Slope (opacity overlay) - LULC (color overlay) - Custom color ramps

Troubleshooting¶

Artifacts in Output¶

Invalid Values¶

Best Practices¶

1. Choose appropriate resolution - Match to analysis scale
2. Use hydro-enforced DEMs - For watershed analysis
3. Validate outputs - Compare with known slopes
4. Consider vertical datum - MSL vs ellipsoid

See Also¶

• Hydrology Pipeline - Uses slope for runoff
• LULC Pipeline - Combined analysis
• API Reference - Programmatic access