Data Acquisition, Quality & Management in GIS

Core Concepts

1. Data Acquisition Methods

Data acquisition is the process of collecting raw spatial data from real-world sources. GIS relies on three primary methods:

A. Remote Sensing

Remote sensing uses sensors (satellites, drones, aircraft) to capture data without physical contact. Key sources:

• Satellite Imagery: Landsat, Sentinel, MODIS (for land cover, vegetation, disaster monitoring).
• Aerial Photography: High-resolution images for urban planning.
• LiDAR (Light Detection and Ranging): Laser pulses to measure elevation (used in terrain modeling).

graph LR
    A["Remote Sensing"] --> B["Satellite Imagery"]
    A --> C["Aerial Photography"]
    A --> D["LiDAR"]
    B --> E["Landsat 8"]
    C --> F["Drone Mapping"]
    D --> G["Digital Terrain Models"]

B. GPS (Global Positioning System)

GPS provides precise location data using satellite signals. Used in:

• Field Surveys: Collecting coordinates for features (roads, buildings).
• Mobile GIS Apps: Pathao uses GPS to track rider locations.
• Precision Agriculture: Farming drones navigate using GPS.

C. Ground Surveys & Digital Data Sources

• Total Station & GNSS: High-accuracy surveys for infrastructure projects.
• Existing Databases: Government sources (Nepal’s Department of Survey, OpenStreetMap).

2. Data Quality Assessment

GIS data must meet standards for accuracy, precision, completeness, consistency, and logical correctness.

Key Metrics

3. Data Management

Proper management ensures data is usable, shareable, and sustainable.

A. Metadata Standards

Metadata describes data (source, accuracy, date, format). Key standards:

• FGDC (Federal Geographic Data Committee)
• ISO 19115 (International standard for spatial metadata)

classDiagram
    class Metadata {
        +Title
        +Abstract
        +Spatial Reference
        +Accuracy
        +Date
    }
    Metadata --> "Describes" Data

B. Database Design for Spatial Data

Spatial databases store GIS data efficiently. Common models:

• Vector Data: Points, lines, polygons (PostGIS, ArcGIS).
• Raster Data: Grids (DEMs, satellite images).
• Topology: Rules defining spatial relationships (e.g., "roads connect at nodes").

C. Data Cleaning & Validation

• Topological Errors: Dangles, overlaps, gaps (fixed using tools like FME).
• Attribute Errors: Incorrect labels (e.g., "forest" misclassified as "urban").

In the Real World

1. eSewa & Khalti (Nepal)

   • GPS & Remote Sensing: These apps use GPS for merchant location tagging and satellite data for delivery route optimization (e.g., avoiding flooded areas during monsoon).

2. Pathao (Ride-Hailing)

   • Spatial Analysis + GPS: Pathao’s algorithm calculates the shortest path using real-time traffic data (from GPS sensors) and elevation models (from LiDAR).

3. Nepal Electricity Authority (NEA) & NTC

   • LiDAR for Infrastructure: NEA uses LiDAR to map power line corridors and avoid vegetation conflicts. NTC uses satellite imagery to plan telecom tower placements.

Worked Example: Daraz Delivery Route Optimization

Problem: Daraz needs to deliver 100 orders in Kathmandu with minimal fuel cost. Solution:

1. Data Acquisition: Use OpenStreetMap (vector data) + GPS traces of past deliveries.
2. Data Quality Check:
   • Validate road network completeness (no missing streets).
   • Ensure elevation data (from LiDAR) is accurate for hilly areas.
3. Spatial Analysis:
   • Apply network analysis (shortest path) in QGIS.
   • Use buffer analysis to avoid congested zones.
4. Output: Optimized routes with fuel savings of ~20%.

Exam Tip

• Short Answer (5 marks):
  • Define remote sensing and list two sources (e.g., Landsat, LiDAR).
  • Explain one data quality metric (e.g., accuracy) with an example.
• Long Answer (10 marks):
  • Compare vector vs. raster data (storage, use cases, examples).
  • Describe how GPS and satellite data are used in one Nepali app (e.g., Pathao).
• Practical (15 marks):
  • Given a dataset, identify three quality issues (e.g., missing attributes, topological errors).
  • Suggest two cleaning methods (e.g., spatial joins, manual editing).

Key Takeaways:

• GIS data comes from remote sensing, GPS, and surveys; each has trade-offs.
• Accuracy ≠ Precision: A GPS may be precise but inaccurate if miscalibrated.
• Metadata is critical for sharing data (e.g., NTC’s telecom maps).
• Topology rules prevent errors like "floating" polygons.
• Real-world apps (eSewa, Pathao) rely on clean spatial data for efficiency.