Projection Selection

This skill is a built-in skill of GeoAgent. Please read all the following content carefully and strictly follow the guidelines during task execution.

Projection Coordinate System Selection Guide

This skill provides a comprehensive knowledge framework for selecting projected coordinate systems, helping you make sound projection choices when faced with geographic data analysis or cartographic tasks.

Prerequisite: You can only begin selecting a projection once the study area's latitude/longitude extent is known. Many projection parameters (such as UTM/Gauss-Krüger zone numbers, conic projection standard parallels, central meridians, etc.) depend on the specific location and extent of the study area — without this information, a correct projection choice cannot be made.

---

I. Distortion Properties of Projections

This is the first decision dimension for projection selection — what geometric property does your task need to preserve?

Four Types of Distortion

Key Principles of Distortion

1. Conformal and equal-area are mutually exclusive — A projection cannot be both conformal and equal-area simultaneously; this is a mathematically proven theorem.
2. Distortion increases with distance — The farther from where the projection surface is tangent/secant to the sphere (standard lines), the greater the distortion.
3. Negligible at small scales — When the mapping area is sufficiently small (e.g., a single city), differences between projections are negligible.

---

II. Projection Families and Applicable Ranges

This is the second decision dimension for projection selection — how large is your map's coverage, what shape is the region, and at what latitude is it located?

Classification by Projection Surface

Projections are classified into families by the type of geometric surface, each naturally suited to different regional shapes:

1. Cylindrical Projections

Project the earth onto a cylinder wrapped around it, then unroll.

• Characteristics: Meridians are equally spaced parallel straight lines; parallels are also parallel straight lines
• Best suited for: Regions near the equator, extending in the east-west direction
• Distortion pattern: Minimum along the equator (or standard parallels), increasing toward the poles

Common projections:

2. Conic Projections

Project the earth onto a cone placed over it, then unroll.

• Characteristics: Meridians are straight lines radiating from the apex; parallels are concentric circular arcs
• Best suited for: Mid-latitude (30°–60°) regions extending in the east-west direction
• Distortion pattern: Minimum along the standard parallels, increasing to the north and south
• Key parameter: Two standard parallels, typically placed at approximately 1/6 inward from the north and south boundaries of the mapping area

Common projections:

3. Azimuthal Projections

Project the earth onto a plane tangent to a single point.

• Characteristics: Radial pattern from the tangent point; azimuth angles from the center point are always correct
• Best suited for: Polar regions, or circular areas centered on a specific point
• Distortion pattern: Increases outward from the tangent point

Common projections:

4. Pseudocylindrical Projections

Variants of cylindrical projections where parallels remain straight lines but meridians curve.

• Characteristics: Overall oval or similar shape, visually close to the earth's "natural feel"
• Best suited for: World maps
• Distortion pattern: Most accurate near the central meridian and equator, increasing toward edges

Common projections:

---

III. Differences in Projection Selection for GIS Analysis vs. Cartography

Projection selection is not purely a technical issue — it also depends on the purpose of what you're doing. GIS analysis and cartographic mapping have different logics for projection requirements.

Important: A single task can use multiple projections. Within the same project, the GIS analysis phase and the final cartographic phase can use entirely different projections. For example: use an equal-area projection for area statistical analysis, then use a projection conforming to local cartographic standards for the final map output. Don't try to use a single projection for all stages — instead, choose the most suitable projection for each stage based on its purpose.