Data Sources

Part 1: Data Input and Digitizing Errors (Section 4.1)

1. Primary Data Sources for Digitizing (2 Marks)

Digitizing is the process of converting analog (physical) information into digital GIS formats (points, lines, polygons). The primary sources you trace over to create this digital data are:

• Scanned Hardcopy Maps: Historical topographic maps or cadastral (parcel) blueprints that have been scanned and georeferenced.
• Satellite Imagery & Aerial Photography: High-resolution orthophotos where the digitizer visually traces roads, rivers, and building footprints directly from the image.

2. Location Errors vs. Topological Errors (4 Marks)

This is a critical distinction in GIS data quality:

• Location (Positional) Errors: This relates to geometric accuracy. It occurs when the geographic coordinates of a feature in the GIS do not match its true physical location on the Earth.
  • Example: A GPS error places a water well 15 meters away from its actual physical location. The point is structurally perfect, but it is in the wrong place.
• Topological Errors: This relates to logical relationships. It occurs when the geometry is drawn incorrectly, breaking the mathematical rules of how features connect and interact, regardless of their absolute coordinates.
  • Examples:
    • Overshoots/Undershoots: A digitized road line goes too far and crosses an intersection (overshoot), or stops just short of connecting to the next road (undershoot).
    • Slivers: Tiny, microscopic gap polygons created when two adjacent boundaries (like two forest stands) don't perfectly snap together.
    • Unclosed Polygons: A lake boundary whose start and end points do not perfectly connect.

3. Workflow: Integrating GPS Data into GIS (10 Marks)

This is a highly practical scenario question. Here are the exact steps a surveyor takes to bring field data into a GIS for analysis:

• Step 1: Data Collection & Export: The surveyor uses a handheld GPS or a Rover to collect waypoints (coordinates) and attributes in the field. This data is exported from the device usually as a .GPX file or a comma-separated .CSV Excel file.
• Step 2: Importing into GIS (Add XY Data): In the GIS software (like ArcGIS or QGIS), the surveyor uses the "Add Delimited Text" or "Display XY Data" tool. They map the spreadsheet's Longitude column to the X-axis and Latitude to the Y-axis.
• Step 3: Defining the Coordinate System: Because raw GPS data is collected globally, the surveyor must define the incoming spatial reference as GCS WGS 1984.
• Step 4: Conversion & Projection: The temporary plotted points are exported and saved permanently as a Point Shapefile or Feature Class. To prepare for engineering measurements, the "Project" tool is used to convert the data from WGS 1984 (degrees) into a local projected coordinate system like MUTM (meters).
• Step 5: Spatial Analysis: The data is now ready. The surveyor can buffer the GPS points, overlay them with municipal wards, or interpolate them to create a continuous surface.

Part 2: Major Data Feeds to GIS and Their Characteristics (Section 4.2)

1. Major Data Sources (Feeds) to GIS (5 to 6 Marks)

To secure full marks on this recurring question, divide your answer into Primary (direct measurement) and Secondary (derived) sources:

Primary Data Sources (Directly Collected):

• GNSS / GPS Field Surveys:
  • Characteristics: Highly accurate, point-based discrete data. Excellent for mapping utilities, manholes, and property corners.
• Remote Sensing (Satellites & Drones):
  • Characteristics: Provides massive, continuous raster coverage. Great for land-use classification, NDVI, and tracking environmental changes over time.
• Total Station / Conventional Surveying:
  • Characteristics: Offers millimeter-level precision. Ideal for small-scale, highly detailed engineering sites like a dam construction layout.

Secondary Data Sources (Derived / Existing):

• Digitized / Scanned Maps:
  • Characteristics: Cost-effective way to get historical data into GIS, but heavily prone to human digitizing errors and paper shrinkage.
• Digital Elevation Models (DEMs):
  • Characteristics: Continuous 3D raster surfaces downloaded from global missions (like SRTM or ASTER) used for watershed and slope analysis.
• Tabular Census / Survey Data:
  • Characteristics: Purely attribute data (non-spatial Excel sheets like population counts) that must be "joined" to existing spatial polygons (like ward boundaries) using a common ID.

2. Positional Accuracy vs. Attribute Accuracy (2 Marks)

GIS data quality is measured in two distinct ways:

• Positional Accuracy: Answers "Is it in the right place?" It is the statistical closeness (usually measured by RMS error) of the map features to their true ground coordinates.
• Attribute Accuracy: Answers "Is the description correct?" It is the correctness of the non-spatial data table attached to the geometry.
  • Example: A road line is perfectly mapped in the correct location (High Positional Accuracy), but the attribute table lists it as a "4-lane pitched highway" when it is actually a "2-lane dirt road" (Low Attribute Accuracy).

4.3 Maps, GPS, Images, Databases (Sources & Importance)

1. The Basic Sources of Data for GIS

While we touched on this earlier, if an exam question specifically asks for the "basic sources" under this chapter's context, group them into these four distinct categories:

• Maps (Analog to Digital): Hardcopy topographic or cadastral maps that are scanned and georeferenced. A technician then manually digitizes (traces) the features into vector formats.
• GPS/GNSS: Field surveys using handheld receivers or high-precision RTK rovers. This provides exact point data for infrastructure like manholes, light poles, or boundary pegs.
• Images (Remote Sensing): Passive satellite imagery (Landsat, Sentinel) or active sensors (LiDAR, Radar) and drone orthophotos. These provide continuous raster data for tracking land-use changes or extracting elevation.
• Databases & Commercial Data: Non-spatial tabular data (like national census population tables or municipal tax records) that are linked to spatial polygons (like ward boundaries) using a common ID number.

2. Why is Data Acquisition/Collection the Most Important Process?

You must mention these two specific principles to get full marks:

• The 80/20 Rule: In any real-world GIS engineering project, data collection, entry, and cleaning consume 80% of the project's total time and budget. The actual spatial analysis and map making only take the remaining 20%.
• The GIGO Principle (Garbage In, Garbage Out): GIS software is just a calculator. It assumes the data you feed it is perfect. If you input a heavily flawed Digital Elevation Model (DEM) into an advanced hydrological model, the software will confidently output a completely incorrect flood plain. In civil engineering, bad input data doesn't just make a bad map; it causes bridges to be built at the wrong elevation and budgets to explode.

4.4 Locating and Evaluating Data

No direct past exam questions have been asked from this section, but it is part of the syllabus and may appear in future exams.

Before starting any GIS project, engineers must find appropriate spatial data and assess whether it is fit for purpose. This process involves locating available datasets and evaluating them against project requirements.

Locating Data Sources

GIS data can be found through several channels:

• National Mapping Agencies: Survey Department of Nepal, USGS, Ordnance Survey — provide official topographic maps, DEMs, and boundary data
• Spatial Data Infrastructures (SDI): National and international portals (e.g., Nepal's National Geoportal, INSPIRE in Europe) that catalog and distribute standardized spatial datasets
• Open Data Portals: OpenStreetMap (OSM), NASA EarthData, DIVA-GIS, Natural Earth — free and community-maintained datasets
• Commercial Providers: DigitalGlobe (satellite imagery), Esri Living Atlas, Google Earth Engine
• In-house Survey: Primary data collected through GPS surveys, total station, drone/LiDAR flights, or field observations
• Metadata Catalogs: Searching metadata records that describe the content, quality, extent, and access conditions of spatial datasets

Evaluating Data Quality

Once located, data must be evaluated before use. Key evaluation criteria include:

4.5 Data Capture Methods, Data Quality, and Metadata

1. Direct vs. Indirect Methods of Spatial Data Capture

• Direct Data Capture: Collecting original data directly from the physical real world.
  • Examples: Standing in a field with a GPS rover to record property corners; flying a drone to capture fresh aerial photography; running a Total Station survey.
• Indirect Data Capture: Creating new GIS data by extracting or converting information from existing data sources.
  • Examples: Heads-up digitizing (tracing over a scanned historical map on your computer screen); converting an existing paper contour map into a digital DEM; scanning a hardcopy blueprint.

2. The 5 Macro-Level Components of Data Quality

When examiners ask for data quality parameters or macro-level components, they want you to list these exact five metrics. If a dataset fails even one of these, it is considered poor quality.

1. Positional Accuracy: How closely the coordinate ($X, Y, Z$) of a feature in the GIS matches its true physical location on the Earth. (e.g., Is the water pipe actually where the map says it is?)
2. Attribute Accuracy: The correctness of the descriptive tabular data attached to the geometry. (e.g., The pipe is drawn in the right place, but does the table correctly list it as PVC, or mistakenly list it as Cast Iron?)
3. Logical Consistency (Topology): Do the geometric features follow mathematical and logical rules? (e.g., Do the road lines perfectly connect at the intersection, or is there a microscopic gap preventing the network routing tool from working?)
4. Temporal Accuracy: Is the data relevant for the current time period? (e.g., Using a 2015 land-use map to plan a 2026 highway alignment has terrible temporal accuracy because thousands of new houses have likely been built since then.)
5. Data Completeness: Is anything missing from the dataset? (e.g., Did the surveyors map 100% of the ward's fire hydrants, or did they stop early and only map 85%?)

3. The Significance of Data Accuracy in GIS Analysis

In civil engineering, spatial analysis is used for legally binding decisions. If a cadastral map lacks positional accuracy, land ownership disputes arise. If a utility network lacks logical consistency, excavation crews might strike an unmapped gas line. Accuracy ensures safety, legal compliance, and economic efficiency.

4. What is Metadata? (3 Marks)

• Definition: Metadata is simply "data about the data."
• Explanation: Think of it like the nutrition label on a can of soup. When you download a shapefile from the internet, you have no idea where it came from. The Metadata file (usually an .XML file attached to the shapefile) tells you the crucial history: Who created it? When was it published? What coordinate system does it use? What is its positional accuracy?
• Importance: Without metadata, a professional engineer cannot legally or ethically trust a dataset for a critical design project because they cannot verify its accuracy or origin.

Solved Numerical Examples (Past Exam Questions)

The following are fully worked solutions from IOE past examination papers. Three unique solution types cover every numerical variant that has appeared in this chapter.

List of All Numerical Questions from Past Papers

Solution Type 1: Allowable RMS Error Calculation

Example A — Solving for 1:5,000 scale map

Step 1: Identify the Condition.
Scale = 1:5,000. Since 5,000 < 20,000, this is a larger scale map → Condition 1 applies.

Step 2: Calculate CE90 at Map Scale.

Step 3: Convert to Ground Distance.

Step 4: Calculate RMSE.

Example B — Solving for 1:25,000 scale map

Step 1: Identify the Condition.
Scale = 1:25,000. Since 25,000 ≥ 20,000, this is a smaller scale map → Condition 2 applies.

Step 2: Calculate CE90 at Map Scale.

Step 3: Convert to Ground Distance.

Step 4: Calculate RMSE.

Example C — Solving for 1:20,000 scale map

Step 1: Scale = 1:20,000. Since 20,000 ≥ 20,000 → Condition 2 applies.

Quick Reference Table: All Scale Answers

Solution Type 2: Distance Between Two Points on a Sphere

This requires the Spherical Law of Cosines.

Given:

• Radius (R) = 6378 km
• φ₁ (Latitude of A) = 24.5°
• φ₂ (Latitude of B) = 24.6°
• Δλ (Longitude difference) = 84.5° − 83.5° = 1.0°

Step 1: Convert Degrees to Radians.

Step 2: Apply the Spherical Law of Cosines Formula.

Step 3: Calculate Each Term.

Step 4: Calculate Distance.

Solution Type 3: Difference Between 1° Latitude and 1° Longitude

This question tests your understanding of Earth as an oblate ellipsoid. Since the Earth bulges at the equator and flattens at the poles, the distance covered by 1° of longitude (measured along the equatorial circle) is slightly longer than 1° of latitude (measured along the polar meridian).

Given (WGS 84):

• Equatorial radius: a = 6,378,137 m
• Polar radius: b = 6,356,752.3142 m

Step 1: Distance of 1° Longitude (at the Equator).

The equator is a perfect circle with radius a.

Step 2: Average Distance of 1° Latitude.

The meridian (pole-to-pole) uses the polar radius b.

Step 3: Calculate the Difference.

End of Chapter 4 – Data Sources