7.3 Terrain Representation & 3D Visualization: Shaded Relief, Slope, Aspect, Curvature & Oblique Views

Terrain Representation & 3D Visualization: Shaded Relief, Slope, Aspect, Curvature & Oblique Views

Quick Summary: Digital terrain modeling extracts primary morphometric derivatives (slope and aspect) and secondary derivatives (profile and plan curvature) from elevation surfaces using 3x3 moving focal window kernels. Visualizing three-dimensional topography on two-dimensional screens or printed maps requires specialized cartographic techniques, including analytical hillshading (shaded relief), hypsometric tinting, and oblique 3D perspective draping. Understanding illumination geometry, the pseudoscopic illusion, vertical exaggeration factors, and coordinate system unit conversions (Z-factors) is critical for accurate spatial analysis and landscape visualization.

---

Digital Elevation Models and Focal Operations

Digital terrain modeling represents continuous topographic surfaces using several primary data models:

• Digital Elevation Model (DEM): A regularly spaced raster grid where each cell stores a single elevation value representing the bare-earth ground surface (devoid of vegetation canopy and human structures).
• Digital Surface Model (DSM): Captures the elevation of the highest reflective surface, including tree canopies, building roofs, and infrastructure.
• Digital Terrain Model (DTM): A bare-earth elevation model augmented with vector breaklines (e.g., ridge crests, stream centerlines, road cuts) to maintain linear topographic discontinuities.
• Triangulated Irregular Network (TIN): A vector-based surface composed of contiguous, non-overlapping triangular facets constructed via Delaunay triangulation from irregularly spaced point samples.

The 3x3 Moving Window Kernel

Most raster terrain derivatives are calculated using a 3x3 cell moving window (kernel) centered over target cell $e_5$:

                    3x3 FOCAL NEIGHBORHOOD KERNEL
                    
                       [ e1   e2   e3 ]
                       [ e4   e5   e6 ]
                       [ e7   e8   e9 ]
                       
   * Cell e5 is the target processing cell.
   * Cell size in X direction = Δx (cell width)
   * Cell size in Y direction = Δy (cell height)

Using Horn's algorithm (the standard formulation in modern GIS software), the first partial derivatives of elevation with respect to the east-west ($X$) and north-south ($Y$) directions are computed using weighted finite differences:

$\frac{\partial z}{\partial x} = \frac{(e_3 + 2e_6 + e_9) - (e_1 + 2e_4 + e_7)}{8 \times \Delta x}$

$\frac{\partial z}{\partial y} = \frac{(e_7 + 2e_8 + e_9) - (e_1 + 2e_2 + e_3)}{8 \times \Delta y}$

Horn's algorithm weights orthogonal adjacent neighbors ($e_2, e_4, e_6, e_8$) twice as heavily as diagonal corner neighbors ($e_1, e_3, e_7, e_9$), providing superior noise damping over older simple difference algorithms.

---

Analytical Hillshading (Shaded Relief)

Analytical hillshading (shaded relief) simulates the visual appearance of sunlight casting shadows across three-dimensional terrain onto a two-dimensional map display.

Illumination Geometry

Hillshading is governed by two user-specified solar illumination parameters:

1. Solar Azimuth: The compass direction of the simulated light source, measured clockwise from true north ($0^\circ$ to $360^\circ$). The universal cartographic default is $315^\circ$ (Northwest).
2. Solar Altitude (Elevation Angle): The angle of the light source above the horizon, measured from $0^\circ$ (at the horizon) to $90^\circ$ (directly overhead at zenith). The standard default is $45^\circ$.

Zenith Angle: The complement of altitude, defined as $\text{Zenith} = 90^\circ - \text{Altitude}$. For a standard $45^\circ$ altitude, the zenith angle is also $45^\circ$ ($0.7854$ radians).

                     HILLSHADE ILLUMINATION GEOMETRY
                     
                       North (0° / 360°)
                              |
             Sun Azimuth      |
               315° (NW) \    |
                          \   |
                           \  |
      West (270°) ----------+---------- East (90°)
                              |
                              |
                              |
                        South (180°)
                        
   Solar Altitude: 45° above horizon (Zenith = 45°)
   Illumination formula: Intensity = 255 * [(cos(Zenith) * cos(Slope)) +
                                           (sin(Zenith) * sin(Slope) * cos(Azimuth - Aspect))]

Lambert's Cosine Law of Illumination

Hillshading algorithms compute the illumination value ($0$ to $255$ integer grayscale) of each cell based on Lambert's Cosine Law: the radiant intensity reflected from a diffuse surface is directly proportional to the cosine of the angle of incidence ($i$) between the incoming illumination vector and the surface normal vector:

$\text{Illumination} = 255 \times \cos(i)$

Where the angle of incidence is calculated from the cell's local slope and aspect relative to solar zenith and azimuth:

$\cos(i) = \left(\cos(\text{Zenith}) \times \cos(\text{Slope})\right) + \left(\sin(\text{Zenith}) \times \sin(\text{Slope}) \times \cos(\text{Azimuth} - \text{Aspect})\right)$

If the resulting illumination value is $\le 0$, the cell lies in complete cast shadow (assigned a value of $0$). Fully illuminated slopes facing directly into the light source approach $255$.

The Pseudoscopic Illusion (Relief Inversion)

One of the most critical perceptual rules in cartography is the prevention of the pseudoscopic illusion, commonly known as relief inversion:

• The human visual cortex has evolved under the subconscious perceptual assumption that overhead light arrives from above and slightly to the left (the upper-left quadrant of visual space).
• If a GIS analyst attempts to model "realistic" Northern Hemisphere solar illumination by setting the light source from the south (e.g., solar azimuth $135^\circ$ to $180^\circ$, representing true midday sun in North America or Europe), the human visual system inverts the topography.
• Mountain ridges appear as sunken canyons or valleys, while river valleys and meteor craters appear as elevated mountain ridges.

[!CAUTION] The Universal Northwest Rule: To prevent relief inversion, cartographers universally illuminate analytical hillshades from the Northwest ($315^\circ$), regardless of the map's actual geographic hemisphere or true solar ephemeris. Even in the Southern Hemisphere, where the real sun shines from the north, human cognitive physiology requires top-left illumination on printed or digital screens.

Advanced Hillshading: Multidirectional and Swiss-Style Shading

Standard single-source hillshades suffer from visual saturation on northwest-facing slopes and pitch-black shadows on southeast-facing slopes. Modern cartographers employ multidirectional hillshading, which computes weighted illumination from multiple light sources (e.g., $225^\circ, 270^\circ, 315^\circ,$ and $360^\circ$) simultaneously. When blended with hypsometric tinting (elevation color ramps) using a Multiply layer blending mode, the resulting display conveys rich relief without obscuring base vector themes.

---

Slope: Angular Degrees vs. Percent Rise

Slope represents the first spatial derivative of elevation, quantifying the maximum rate of change in elevation across each cell relative to its immediate neighbors.

Measurement Formats

Slope is expressed in two distinct formats that must never be confused:

1. Degree of Slope ($0^\circ$ to $90^\circ$): The angular incline of the terrain surface relative to a horizontal plane:

$\text{Slope}_{\text{degrees}} = \arctan\left(\sqrt{\left(\frac{\partial z}{\partial x}\right)^2 + \left(\frac{\partial z}{\partial y}\right)^2}\right) \times \frac{180^\circ}{\pi}$

1. Percentage Rise ($0%$ to $\infty%$): The ratio of vertical rise to horizontal run, multiplied by 100:

$\text{Slope}_{\text{percent}} = \tan(\text{Slope}_{\text{degrees}}) \times 100 = \left(\frac{\text{Rise}}{\text{Run}}\right) \times 100$

                     DEGREES VS. PERCENT RISE
                     
   Flat Surface:       0°  =    0% Rise  (Rise = 0, Run = 100)
   Moderate Slope:   14.04°=   25% Rise  (Rise = 25, Run = 100)
   Steep Slope:      26.57°=   50% Rise  (Rise = 50, Run = 100)
   45-Degree Incline: 45°  =  100% Rise  (Rise = 100, Run = 100) <-- EXAM TRAP!
   Precipitous:      63.43°=  200% Rise  (Rise = 200, Run = 100)
   Vertical Wall:     90°  =  Approaches Infinity

[!IMPORTANT] The 45-Degree / 100% Rise Trap: A common exam misconception is assuming that a $100%$ slope represents a vertical $90^\circ$ cliff. This is mathematically incorrect. A $100%$ slope occurs when vertical rise equals horizontal run ($\text{Rise} / \text{Run} = 1.0$), which corresponds to an angle of exactly $45^\circ$. A vertical wall ($90^\circ$) has a run of zero, resulting in a slope that approaches infinity.

The Critical Z-Factor (Unit Conversion Multiplier)

The Z-factor is an essential conversion parameter applied when the horizontal ground coordinate units ($X, Y$) differ from the vertical elevation units ($Z$):

$\text{Effective Surface Gradient} = \sqrt{\left(\frac{\partial z}{\partial x}\right)^2 + \left(\frac{\partial z}{\partial y}\right)^2} \times Z\text{-factor}$

Common unit mismatch scenarios include:

• Geographic Coordinates: Horizontal coordinates are stored in Decimal Degrees (WGS84 or NAD83), but elevation is recorded in meters or feet. Because one degree of latitude equals approximately 111,320 meters at the equator, running slope without a Z-factor treats one degree as one meter, resulting in slope calculations exaggerated by more than 100,000x! The analyst must calculate the latitude-appropriate Z-factor ($Z = 1 / [111,320 \times \cos(\text{latitude})]$) or reproject the DEM into a conformal projected coordinate system (e.g., UTM or State Plane).
• Feet vs. Meters: Horizontal coordinates are in UTM meters, but elevation is recorded in US Survey Feet ($1 \text{ foot} = 0.3048 \text{ meters}$). A Z-factor of $0.3048$ must be applied to convert feet to meters prior to slope computation.

---

Aspect: Compass Orientation and Azimuths

Aspect is the compass direction that the maximum downslope gradient faces. It represents the downhill orientation of the terrain surface face.

Mathematical Computation

Aspect is computed from the directional derivatives using the two-argument arctangent function:

$\text{Aspect} = \text{atan2}\left(-\frac{\partial z}{\partial y}, -\frac{\partial z}{\partial x}\right)$

The resulting angle is converted into standard compass azimuth degrees ranging clockwise from $0^\circ$ to $360^\circ$:

• North: $0^\circ$ (or $360^\circ$)
• East: $90^\circ$
• South: $180^\circ$
• West: $270^\circ$

Handling Flat Terrain: The -1 Value

When a cell and its 3x3 neighbors have zero elevation change ($\partial z/\partial x = 0$ and $\partial z/\partial y = 0$), slope is zero. On a flat surface, the downhill direction is undefined. GIS software assigns flat terrain cells a special flag value of $-1$ (or NoData / null).

Environmental Applications

• Microclimate and Insolation: In the Northern Hemisphere, south-facing slopes ($135^\circ$ to $225^\circ$) receive significantly higher solar irradiance, resulting in warmer, drier soils, sparse vegetation, and accelerated snowmelt. North-facing slopes ($315^\circ$ to $45^\circ$) remain cooler and wetter, fostering denser forest canopy.
• Wildfire Behavior: Aspect dictates fuel moisture levels and wind-exposure dynamics.
• Avalanche Risk Modeling: Windward versus leeward aspect orientations govern dangerous snow slab accumulation.

---

Terrain Curvature: Profile vs. Plan Curvature

Curvature represents the second spatial derivative of elevation (the rate of change of slope). Curvature characteristically decomposes into two orthogonal directional components that govern surface hydrology and sediment transport:

                       TERRAIN CURVATURE COMPONENTS
                       
        [Profile Curvature]                      [Plan (Planform) Curvature]
     Measured parallel to slope               Measured perpendicular to slope
            (Downslope)                              (Along Contours)
            
           Convex: \                                  Convex:   /   \
                    \_ (Accelerating Flow)                     /     \ (Diverging Flow)
                                                              --------- (Ridges/Spurs)
           Concave: \_                                Concave: \     /
                      \ (Decelerating Flow)                     \   /  (Converging Flow)
                                                              --------- (Gullies/Ravines)

Modern landscape morphometry software also derives Tangential Curvature (plan curvature multiplied by the sine of slope) and Total Curvature to model complex soil moisture indices.

---

3D Perspective and Oblique Views

Visualizing geographic datasets in three dimensions involves projecting 3D $(X, Y, Z)$ coordinates onto a two-dimensional camera viewing plane using perspective projection geometry.

Primary Camera Parameters

                     3D PERSPECTIVE VIEWING GEOMETRY
                     
              Camera Position (Observer X,Y,Z)
                    \ 
                     \   Pitch / Tilt Angle
                      \  (0° = Horizon, 90° = Nadir Looking Down)
                       \ 
                        \ 
                         +------------------> Look-at Target Point
                                              (Center of View)
                                              
   * View Azimuth (Heading): Compass direction of camera line-of-sight.
   * Field of View (FOV): Angular viewing cone (narrow = telephoto, wide = perspective distortion).
   * Vertical Exaggeration (VE): Multiplier applied to Z values relative to X,Y.

1. Observer Coordinates $(X, Y, Z)$ and Target Point: Define camera location and focal look-at center.
2. Pitch / Tilt Angle: The inclination angle between the camera optical axis and the horizontal ground plane ($0^\circ$ represents looking horizontally toward the horizon; $90^\circ$ represents looking straight down at nadir).
3. View Azimuth (Heading): The compass direction of the camera line-of-sight ($0^\circ$ to $360^\circ$).
4. Field of View (FOV): The angular cone captured by the virtual camera. A narrow FOV ($15^\circ$ to $30^\circ$) resembles a telephoto lens, minimizing perspective distortion; a wide FOV ($60^\circ$ to $90^\circ$) expands peripheral visibility but causes distortion near screen edges.

Vertical Exaggeration (VE)

Vertical Exaggeration (VE) is an artificial scaling factor applied to the vertical elevation axis ($Z$) relative to the horizontal ground coordinates ($X, Y$):

$\text{VE} = \frac{\text{Vertical Scale Factor}}{\text{Horizontal Scale Factor}}$

• When to Apply VE: In flat or gently rolling regions (such as coastal plains or interior prairies), natural topography appears completely flat in 3D perspective views. Applying a VE of $2\times, 5\times,$ or $10\times$ accentuates subtle geomorphic features (paleo-channels, dunes, escarpments).
• Exam Trap & Risk: Applying excessive VE in mountainous terrain produces precipitous, unnatural needle-peaks, exaggerates apparent slope angles, and misleads geotechnical hazard assessments.

Surface Draping

Surface draping projects two-dimensional planimetric themes (such as multispectral satellite imagery, aerial orthophotographs, vector parcel boundaries, or classified thematic rasters) over a three-dimensional elevation mesh (DEM or TIN). The rendering engine applies texture coordinate mapping, interpolating $Z$-elevations for every draped vertex to conform the 2D imagery to the undulating terrain surface.

---

Summary of Terrain Derivatives & Visualization Rules

---

Common Exam Traps

[!CAUTION] Exam Trap 7.3.1: Pseudoscopic Relief Inversion from Southern Light. Questions frequently test the result of setting the hillshade solar azimuth to the south (e.g., $180^\circ$ or $135^\circ$). The correct answer is the pseudoscopic illusion (relief inversion): valleys appear as ridges, and ridges appear as valleys. Illumination must be set to the Northwest ($315^\circ$).

[!CAUTION] Exam Trap 7.3.2: Percent Slope vs. Degrees (The 45-Degree Trap). An exam question asks for the percent rise of a $45^\circ$ incline. Many examinees guess $50%$. The correct answer is $100%$ because $\text{rise} = \text{run}$ ($\tan(45^\circ) \times 100 = 100%$). A $90^\circ$ slope approaches infinity.

[!CAUTION] Exam Trap 7.3.3: Missing Z-Factor in Geographic Coordinate Systems. Calculating slope or shaded relief directly on a DEM projected in WGS84 decimal degrees without applying a Z-factor produces wildly distorted results (slopes near $89^\circ$ everywhere) because degrees are treated as linear meters. A conversion Z-factor or projected coordinate system is required.

[!CAUTION] Exam Trap 7.3.4: Confusing Profile vs. Plan Curvature. Remember: Profile curvature governs flow acceleration/deceleration (velocity downslope), while Plan curvature governs flow convergence/divergence (horizontal spreading or concentrating across contours).