GIS Oral Exam

Definition of GIS

Geographic Information System (GIS) is a system designed for gathering, managing, analyzing, and mapping all types of data that are connected to locations on Earth. GIS integrates spatial (location-based) data with descriptive information, enabling users to visualize, interpret, and understand spatial relationships, patterns, and trends.

Main Components of GIS

• Data: Includes both spatial data (location, geometry) and attribute data (descriptive information).

• Analysis: Tools and methods for spatial analysis, modeling, and querying data.

• Maps: Visualization of spatial data through maps and other graphical outputs.

• Software: Platforms like ArcGIS Pro, enabling data management, analysis, and visualization.

Vector Data

• Definition: Represents geographic features using points, lines, and polygons, each defined by coordinates in a known reference system.

Examples:

• Points: Locations of wells or trees

• Lines: Roads, rivers

Polygons: Land parcels, lakes.

Raster Data

• Definition: Represents geographic phenomena as a matrix of cells (pixels), each with a value indicating a specific attribute (e.g., elevation, temperature).

Examples:

• Satellite imagery

• Digital Elevation Models (DEMs)

Land cover classification maps.

Main Features of Raster Datasets

• Structure: Grid of cells (pixels), each with a single value.

• Resolution: Defined by cell size (spatial resolution); smaller cells = higher detail.

• Data Type: Suitable for continuous data (e.g., temperature, elevation).

• Spectral/Radiometric Resolution: Can store multiple bands (e.g., RGB, infrared) and various data types (integer, floating point).

• Use Cases: Ideal for overlays, environmental modeling, remote sensing, and representing phenomena that vary continuously across space.

Geodatabases vs Shapefile Data Formats: Pros and Cons

Feature Geodatabase Shapefile

• Storage: Supports complex data types and relationships, while shapefiles are simpler and less flexible.
• File Size: Geodatabases can handle larger datasets, while shapefiles have a 2GB size limit.
• Performance: Geodatabases offer better performance for larger datasets, while shapefiles are slower for complex analyses.

Definition of Metadata

Metadata is information that describes a dataset, including its name, creation date, geographic extent, coordinate reference system, resolution, and data format.

Importance of Metadata

• Sharing: Ensures others can understand, use, and trust the dataset.

• Accessing: Facilitates data discovery and retrieval in catalogs and portals.

• Interoperability: Helps integrate datasets from different sources.

Standardization: ISO19115 is the main international standard for GIS metadata.

Purpose of a Web Service

A web service allows users and applications to access, visualize, and interact with spatial data over the internet, often in real time. This enables sharing, updating, and analyzing geospatial data without needing local copies.

Two Main OGC Web Services

• WMS (Web Map Service): Provides map images for visualization; users can view but not directly interact with the underlying data.

WFS (Web Feature Service): Provides access to vector features, allowing users to query, retrieve, and sometimes edit spatial data directly.

What is a Datum?

A datum is a mathematical model of the Earth that defines the origin, orientation, and scale of a coordinate system used to calculate geographic positions.

Geoid: Earth’s true irregular surface defined by gravity and sea level

Ellipsoid: Smooth mathematical approximation of Earth’s shap.

A datum alligns the ellipsoid to the Earth’s surface

Global vs Local Datums: Orientation Differences

Global Datum: The ellipsoid is centered at the Earth's center of mass, providing a best-fit model for the entire planet (e.g., WGS84). This orientation allows for worldwide consistency in positioning.

Local Datum: The ellipsoid is oriented and positioned to best fit a specific region, often making it tangential to the geoid at a chosen location. This minimizes vertical deviation locally but may not be accurate globally (e.g., ED50 for Western Europe, Roma 40 for Italy).

Can the Same Ellipsoid Be Used in Different Datums? Why?

Yes, the same ellipsoid can be used in different datums. The difference between datums lies in the orientation, position, and reference frame of the ellipsoid, not just its shape. For example, the Hayford 1909 ellipsoid is used in both the European Datum 1950 (ED50) and Roma 40, but each datum orients the ellipsoid differently to best fit their respective regions.

Two Types of Coordinate Systems: Definitions

Geographic Coordinate System (GCS):

• Uses angular units (latitude and longitude) to define locations on the Earth's surface.

• Based on a datum and ellipsoid.

• Example: WGS84 geographic coordinate system12.

Projected Coordinate System (PCS):

• Uses linear units (e.g., meters) to represent locations on a flat, two-dimensional surface (map).

• Derived from a geographic coordinate system using a map projection.

Example: UTM (Universal Transverse Mercator) system12.

Main Characteristics of the UTM Projection

• Type: Cylindrical, conformal projection (preserves local shapes/angles)

• Structure: Divides the world into 60 zones, each 6° of longitude wide

• Coordinates: Each zone has its own planar reference system; coordinates are given as Easting and Northing in meters

• False Origin: Each zone has a false easting (500,000 m) to avoid negative coordinates

• Coverage: Extends from 80°S to 84°N latitude

Usage: Widely used for large- and medium-scale mapping.

Importance of Knowing the UTM Zone

Knowing the correct UTM zone of a dataset is crucial because:

• Each zone is a separate planar reference system; coordinates are only meaningful within the correct zone.

• Using the wrong zone can result in significant positional errors, misalignment, and incorrect spatial analysis.

Proper data integration and interpretation depend on referencing the correct UTM zone

Goal of a Datum Transformation

A datum transformation converts coordinates from one datum (reference frame and ellipsoid) to another. The goal is to maintain positional accuracy when combining or comparing datasets that use different datums. This is essential for integrating spatial data from multiple sources.

Does Datum Transformation Impact Positional Accuracy?

A datum transformation converts coordinates from one datum (reference frame and ellipsoid) to another. The goal is to maintain positional accuracy when combining or comparing datasets that use different datums. This is essential for integrating spatial data from multiple sources.

What is a Spatial Reference System Identifier (SRID)?

A Spatial Reference System Identifier (SRID) is a unique integer code assigned to a specific coordinate reference system (CRS). SRIDs are used in GIS software and databases to unambiguously identify and manage different spatial reference systems, ensuring consistency and interoperability when sharing or integrating geospatial data

Explain the difference between large map scales and small map scales in terms of the level of detail and the geographic coverage they provide

• Large Map Scale:

• Example: 1:1,000 or 1:5,000

• Detail: High—shows small areas with great detail (e.g., city blocks, buildings).

• Coverage: Low—covers a small geographic area.

• Small Map Scale:

• Example: 1:100,000 or 1:1,000,000

• Detail: Low—shows large areas with less detail (e.g., entire countries).

• Coverage: High—covers a large geographic area.

Which is the relation between nominal map scale and map accuracy?

The nominal map scale (1:n) defines the intended level of detail and accuracy of the map. The smaller the denominator (n), the larger the scale and the higher the accuracy.

• Accuracy Formula:

Horizontal map accuracy (95%) is calculated as:

Accuracy=0.0004×nAccuracy=0.0004×n

where nn is the scale dxenominator.

If you require geospatial data with a specific accuracy, how can you determine the necessary nominal map scale according to the information provided

Positional accuracy ≈ 1/2,000 of the map scale denominator
Map Scale Denominator ≈ Required Accuracy × 2,000

If you need ±5 meters accuracy:
Scale = 5 m × 2,000 = 1:10,000

How does vertical map accuracy typically compare to horizontal map accuracy

• Vertical Accuracy: Typically lower than horizontal accuracy for the same scale.

• Example: At 1:10,000, horizontal accuracy ≈ 4 m, but vertical accuracy for buildings ≈ 4 m, and for tall vegetation ≈ 6 m.

• Reason: Elevation data sources are less precise than horizontal measurements.

Always check metadata for the exact values.

Is nominal map scale relevant for vector datasets?

Nominal map scale is relevant for vector datasets because it defines the original accuracy and intended use of the data, even though vectors can be displayed at any scale in digital form. Using vector data at a larger scale than its nominal scale does not improve accuracy.

Why it is crucial to know the nominal map scale (and metadata in general) of a dataset?

• Ensures data is used at an appropriate level of detail and accuracy.

• Prevents errors in analysis (e.g., using small-scale data for detailed urban planning).

Metadata documents scale, accuracy, and other properties, helping users avoid misuse and misinterpretation of data.

What are the three fundamental elements of the Entity-Relationship (ER) model?

1. Entities: Objects or items of interest (e.g., buildings, parcels).

2. Attributes: Properties or characteristics of entities (e.g., address, area).

Relationships: Associations between entities (e.g., a building is located on a parcel).

Define cardinality in the context of relationships between entities

Cardinality describes the number of possible associations between entities in a relationship (e.g., one-to-one, one-to-many, many-to-many).

Can you provide at least one example of 1:1, 1:N or N:N relationships

• One-to-One (1:1):

• Each county has one population change record.

• One-to-Many (1:N):

• One land-use type can be assigned to many polygons.

• Many-to-Many (N:M):

Many students can enroll in many courses, and vice versa.

What is the difference between Join and Relate?

Join: Combines two tables into one based on a common field.

Relate: Links two tables through a common field but keeps them separate.

What is the purpose of joining tables in GIS?

To enrich a spatial layer's attribute table with external data (e.g., adding census data to a map of districts) for querying, symbology, and analysis.

How does relating tables differ from joining tables?

• Join: Merges data for each matching record (e.g., 1:1 or M:1).

• Relate: Creates a navigable relationship, best for 1:N or M:N, without merging data.

Which types of cardinality are best suited for joining tables in ArcGIS Pro?

• One-to-One (1:1)

• Many-to-One (M:1)
  Because only one match per record is allowed in a Join.

In case of 1:N cardinality, would you opt for a Join or a Relate? Explain and provide an example.

A Join cannot handle multiple matches for one record; it will only attach the first match. Relate maintains the one-to-many link.

Example:

• A parcel layer (each parcel is unique)

• A sales table with multiple sale records per parcel
  Use Relate so each parcel can link to all its sale records.

Define Remote Sensing and Provide an Example

Remote sensing is the process of collecting information about Earth’s surface without direct contact, usually via satellites or aircraft.
Example: Using satellite imagery to monitor crop health in agriculture/ weather forecasting/ Google Maps navigation

What is the electromagnetic spectrum (EMS)?

The EMS is the full range of electromagnetic radiation, from gamma rays to radio waves.
The wavelength or frequency of the radiation determines its position in the spectrum (shorter wavelength = higher energy).

EMS regions detectable by remote sensing sensors

• Visible (400–700 nm)

• Near Infrared (NIR, ~700–1,300 nm)

• Shortwave Infrared (SWIR, ~1.3–2.5 μm)

• Thermal Infrared (TIR, ~8–14 μm)

• Microwave (Radar, ~1 mm–1 m)

Resolutions in Remote Sensing

• Spatial: Size of the smallest object detected (e.g., 10 m/pixel)

• Temporal: Frequency of image acquisition (e.g., every 5 days)

• Spectral: Number and width of wavelength bands (e.g., 4-band or hyperspectral)

• Radiometric: Sensor’s sensitivity to brightness differences (e.g., 8-bit = 256 levels)

Explain the concept of Ground Sample Distance (GSD) and its relationship to the level of detail visible in a satellite image

GSD is the distance between pixel centers on the ground.
Smaller GSD = higher detail (e.g., 0.5 m GSD sees roads clearly; 30 m GSD sees only land cover types)

Explain the concept of a spectral signature and how it can be useful in analyzing remotely sensed data

A spectral signature is the unique pattern of reflectance/emittance across spectral bands for a material (e.g., vegetation reflects strongly in NIR).
Useful for classification and identifying land cover types.

Describe the characteristics of a typical near-polar sun-synchronous orbit used by Earth Observation satellites and the impact on the revisiting time

• Passes near both poles

• Crosses the equator at the same local solar time daily

• Enables consistent lighting for images

• Typical revisit: ~1–16 days depending on swath width and orbit

Discuss the main relations among spatial, temporal, radiometric and spectral resolutions

• Higher spatial → lower temporal (due to narrow swath)

• Higher spectral/radiometric → larger file size, slower revisit

• Trade-offs: More detail (spatial/radiometric) vs. frequency (temporal) vs. spectral richness

Explain the difference between passive and active remote sensing sensors, providing an example of each.

• Passive sensors detect natural energy (usually sunlight) reflected/emitted by surfaces.
  Example: Landsat (optical imagery).

• Active sensors emit their own energy and measure the return signal.
  Example: Sentinel-1 (SAR radar).

Discuss the main pros and cons of optical vs radar remote sensing

Optical (e.g., Sentinel-2)

• ✅ Natural-looking images, useful for vegetation, land cover

• ❌ Affected by clouds, needs daylight

Radar (e.g., Sentinel-1)

• ✅ Works in all weather, day/night

• ❌ Harder to interpret visually, sensitive to surface roughness

List at least five different applications of satellite remote sensing

• Crop monitoring and precision agriculture

• Deforestation and land cover change

• Urban growth and planning

• Disaster assessment (e.g., floods, earthquakes)

• Glacier and sea ice tracking

Compare and contrast the Sentinel-1 and Sentinel-2 missions

Feature	Sentinel-1	Sentinel-2
Sensor type	Radar (SAR)	Optical (MSI)
Works in clouds/night	✅ Yes	❌ No
Resolution	~10 m	10–60 m
Main use	Surface deformation, flood mapping	Land cover, vegetation health

What is the significance of the Landsat program in the context of Earth observation history?

Launched in 1972, Landsat is the longest-running Earth observation program.
It provided the first consistent, global, multispectral data for environmental monitoring and remains essential for studying long-term changes (e.g., urbanization, climate impact).

True color/visible vs false colour composites

• True Color: Uses visible bands (R-G-B) → image looks like what we see with our eyes

• False Color: Swaps or adds bands (e.g., NIR-R-G) → enhances features like vegetation or water

Why does vegetation look reddish in a false colour composite?

Vegetation reflects strongly in the near-infrared (NIR) band.
In a common false color composite (NIR → red), this reflectance is shown as bright red, highlighting plant health.

In a common false color composite, the near-infrared (NIR) band is displayed as red, the red band as green, and the green band as blue.

• Reason: Healthy vegetation reflects strongly in the NIR region but less in the visible red. When NIR is displayed as red, areas with lush vegetation appear bright red, making them stand out clearly from other land covers.

What is the primary task of image classification in satellite remote sensing?

To assign each pixel (or group of pixels) in an image to a specific land cover class (e.g., forest, water, urban) based on its spectral properties.

Describe the main difference in approach between supervised and unsupervised classification

• Supervised: Analyst defines training areas with known classes → algorithm learns from these samples

• Unsupervised: Algorithm groups pixels into natural clusters based on spectral similarity, with no prior knowledge; classes are labeled after clustering.

What are the typical outputs of unsupervised classification before analyst interpretation?

• Several spectral clusters (e.g., Class 1, Class 2…)

• No meaningful land cover names yet — analyst must assign classes afterward based on interpretation

What are the typical outputs of supervised classification?

• A classified image with labeled classes (e.g., Forest, Water, Urban)

• Accuracy assessment reports using validation data (e.g., confusion matrix)

Classification Methods: Pixel-based vs. Object-Based (OBIA)

Method	Pixel-Based	Object-Based (OBIA)
Unit	Individual pixels	Groups of pixels (objects)
Sensitivity	High to noise	Lower due to spatial context
Suitable for	High-res: not ideal	High-res: more accurate
Contextual info	Not used	Shape, texture, context used

Pixel-Based Classification

• Approach: Each pixel is classified individually based on its spectral value, ignoring neighboring pixels.

• Use Case: Suitable for medium- to low-resolution imagery.

• Limitation: Can result in a "salt and pepper" effect due to isolated misclassifications.

Object-Based Image Analysis (OBIA)

• Approach: Groups neighboring pixels with similar properties into objects using segmentation, considering color, shape, and pattern.

• Advantages: Produces cleaner results that better represent real-world features, especially with very high-resolution (VHR) imagery.

• Segmentation: The process of dividing an image into meaningful regions or objects (e.g., fields, urban blocks).

What is the process known as segmentation in the object-based approach (OBIA)?

Segmentation divides the image into homogeneous regions (objects) based on spectral and spatial similarity.
It is the first step in OBIA, where pixels are grouped into meaningful “image objects” for further classification.

What is the purpose of a scatter plot in the context of examining image bands?

A scatter plot visualizes the relationship between two image bands, helping to:

• Identify spectral separability between classes

• Choose bands or thresholds for classification

• Detect correlation or redundant information

What is the primary purpose of using Spectral Indices (SI)?

To enhance specific land surface features (e.g., vegetation, water, built-up areas) by combining spectral bands into a single value that highlights target phenomena while reducing noise from other factors.

How are spectral indices typically calculated, and why use ratios?

Most indices use band ratios or normalized differences, like:
(Band A – Band B) / (Band A + Band B)
Benefit: Ratios reduce effects of illumination, sensor differences, or topography, improving comparability and feature detection.

normalize the effects of shadows and differences in illumination, making the indices more robust and comparable across different images and acquisition conditions.

What does NDVI stand for and what is its main application?

NDVI = Normalized Difference Vegetation Index
Formula: (NIR – Red) / (NIR + Red)
Application: Assessing vegetation health, density, and greenness.

What does MNDWI stand for and what is its main application?

MNDWI = Modified Normalized Difference Water Index
Formula: (Green – SWIR) / (Green + SWIR)
Application: Enhancing open water bodies, especially in urban areas.

What is histogram thresholding in the context of spectral indices?

Histogram thresholding involves analyzing the distribution of index values (e.g., NDVI) and setting cutoff thresholds to classify pixels (e.g., water vs. land, vegetation vs. non-vegetation) based on their index values.

What is GIS and its main components?

GIS (Geographic Information System) is a framework for capturing, storing, analyzing, and visualizing spatial data to understand relationships, patterns, and trends.

Main components:

• Hardware: Computers, GPS units

• Software: ArcGIS, QGIS

• Data: Vector (roads), Raster (satellite imagery), tables

• People: Users, analysts

• Methods: Workflows, spatial models

Provide definitions and examples of the main GIS data types

• Vector: Points, lines, polygons (e.g., cities, rivers, land parcels)

• Raster: Grid-based data, with cells/pixels (e.g., elevation models, satellite images)

• Attribute table: Linked tabular data describing features (e.g., population, area)

Describe the main features of raster datasets

• Grid of cells/pixels, each with a value

• Represents continuous data (e.g., temperature, elevation)

• Resolution = size of each pixel

• Larger datasets than vector; fast in overlay analysis

• Common formats: GeoTIFF, IMG

Discuss the main pros and cons of Geodatabases vs Shapefile data formats

Feature	Shapefile	Geodatabase
Structure	Simple, flat	Advanced, organized
Max size	~2 GB	>1 TB (File GDB)
Topology & rules	❌	✅
Multiple layers	❌ One per file	✅ Many in one
Field name limit	10 chars	64+
Portability	✅ Easy	❌ Slightly harder

Definition of Metadata: discuss the importance of metadata when sharing/accessing datasets

Metadata = "Data about data"
Describes dataset details: source, date, projection, accuracy, usage rights

Importance:

• Helps users understand, evaluate, and reuse data

• Essential for data sharing, transparency, and interoperability

What is the purpose of a Web Service? List the two main OGC Web Services and their main feature

Web Services allow GIS data to be shared and accessed online without downloading.

Main OGC Web Services:

1. WMS (Web Map Service) – Serves maps as images (e.g., satellite basemaps)

2. WFS (Web Feature Service) – Serves vector features with full attribute data (editable, queryable)

What is a Datum?

A datum defines the origin, orientation, and shape of the Earth's model used to reference geographic coordinates. It combines an ellipsoid with an anchor point on Earth.

How does a global datum differ from a local datum in its orientation?

• Global datum (e.g., WGS84): Best fit for the entire Earth, centered at Earth's mass center.

• Local datum (e.g., NAD27, ED50): Optimized for a specific region, with an origin close to the local area for higher accuracy there.

Can the same ellipsoid be used in different datums? Why?

Yes.
The same ellipsoid (e.g., GRS80) can be used in different datums because a datum is defined by both the ellipsoid and its position/orientation relative to the Earth. Two datums may use the same ellipsoid but differ in anchor point and rotation.

List and define two different types of coordinates systems

• Geographic Coordinate System (GCS): Uses latitude & longitude based on a datum (e.g., WGS84)

• Projected Coordinate System (PCS): Converts 3D Earth onto a 2D plane using a projection (e.g., UTM, State Plane)

What is a map projection, and why is distortion an inherent consequence of this process?

A map projection mathematically transforms the Earth's curved surface to a flat map.
Distortion is inevitable because you can't flatten a sphere without stretching, tearing, or compressing — leading to errors in area, shape, distance, or direction.

Which are the main characteristics of a UTM projection?

• Universal Transverse Mercator

• Divides Earth into 60 zones, each 6° wide

• Uses meters as units

• Minimizes distortion within each zone

• Best for small to medium-scale regional maps

Why it is crucial to know the UTM Zone of a dataset referenced in a UTM cartographic reference system

Each zone has its own origin and grid, so using the wrong zone can shift your dataset hundreds of kilometers off. Always match the zone to your data’s location.

Which is the goal of a Datum Transformation? Does it impact on the positional accuracy of the dataset?

To convert coordinates from one datum to another accurately.
Example: Transforming WGS84 data to ED50 for Europe mapping.

✔ Yes, it affects positional accuracy — especially for high-precision work. Poor transformations can introduce meter-level errors.

What is a Spatial Reference System Identifier (SRID)

An SRID is a numeric code used to uniquely define a coordinate system (including its projection and datum).
Example:

• 4326 → WGS84 (lat/lon)

• 32633 → UTM Zone 33N, WGS84

SRIDs ensure spatial data is interpreted and transformed correctly across systems.

Explain the difference between large map scales and small map scales in terms of the level of detail and the geographic coverage they provide

• Large scale (e.g., 1:5,000):
  ✅ Shows more detail, but covers a smaller area

• Small scale (e.g., 1:500,000):
  ✅ Shows less detail, but covers a larger area

Tip: Think “large scale = large features.”

Which is the relation between nominal map scale and map accuracy?

The nominal scale indicates the scale at which data was created or is most accurate.
Positional accuracy depends on the map scale — a rule of thumb:

Positional accuracy ≈ 1/2,000 × map scale denominator

So, 1:10,000 map → ±5 m expected accuracy.

If you require geospatial data with a specific accuracy, how can you determine the necessary nominal map scale according to the information provided

Scale = Required Accuracy × 2,000
Example:
You need ±3 m accuracy →
3 × 2,000 = 1:6,000 nominal scale required

How does vertical map accuracy typically compare to horizontal map accuracy

Vertical accuracy (elevation) is usually worse than horizontal accuracy because:

• Elevation data is harder to measure precisely

• Terrain, sensor angle, and model resolution affect it more

Is nominal map scale relevant for vector datasets?

✅ Yes.
Even though vectors are scale-independent in appearance, their accuracy, generalization level, and intended use depend on the nominal scale at which they were digitized or created.

Why it is crucial to know the nominal map scale (and metadata in general) of a dataset?

• Ensures appropriate use of data (e.g., don’t zoom in beyond its accuracy)

• Helps in error assessment, data compatibility, and legal defensibility

• Metadata informs on source, projection, resolution, and limitations — vital for analysis and documentation

Can you provide at least one example of 1:1, 1:N or N:N relationships

• 1:1: A person ↔ passport

• 1:N: A school ↔ many students

• N:N: Students ↔ courses (each student takes many courses, each course has many students)

Which is the difference between join and relate?

Feature	Join	Relate
Type	Merges data into one table	Links tables, keeps them separate
Cardinality	1:1 or M:1	1:N or M:N
Output	One unified table	Related records (viewed on request)
Purpose	Add attributes for mapping/analysis	Explore multiple related records

What is the purpose of joining tables in GIS?

To enrich a spatial layer by adding attribute data (e.g., adding population data to districts) for analysis, symbology, or queries.

How Relate Differs from Join

• Join pulls in exactly one matching record per feature.

• Relate allows for multiple matching records, ideal for more complex relationships (e.g., a building with many maintenance logs)